[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r305198230 ## File path: integration/spark2/src/main/scala/org/apache/carbondata/indexserver/IndexServer.scala ## @@ -178,6 +171,8 @@ object IndexServer extends ServerInterface { server.stop() } }) + CarbonProperties.getInstance().addProperty(CarbonCommonConstants Review comment: Yeah, but this is added as a flag to tell the common logic that index-server was enabled. The user can still use set to control per table This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r305198086 ## File path: docs/index-server.md ## @@ -0,0 +1,226 @@ + + +# Distributed Index Server + +## Background + +Carbon currently prunes and caches all block/blocklet datamap index information into the driver for +normal table, for Bloom/Index datamaps the JDBC driver will launch a job to prune and cache the +datamaps in executors. + +This causes the driver to become a bottleneck in the following ways: +1. If the cache size becomes huge(70-80% of the driver memory) then there can be excessive GC in +the driver which can slow down the query and the driver may even go OutOfMemory. +2. LRU has to evict a lot of elements from the cache to accommodate the new objects which would +in turn slow down the queries. +3. For bloom there is no guarantee that the next query goes to the same executor to reuse the cache +and hence cache could be duplicated in multiple executors. + +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server service in form of +a request. The request will consist of the table name, segments, filter expression and other +information used for pruning. + +In IndexServer service a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. Refer: [query-data-with-specified-segments](https://github.com/apache/carbondata/blob/6e50c1c6fc1d6e82a4faf6dc6e0824299786ccc0/docs/segment-management-on-carbondata.md#query-data-with-specified-segments). +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the executor where the segment is cached. + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. + +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment(version: 1.1) the index size is not available +therefore all the legacy segments would be processed in a round robin +fashion. + +After the job is completed the tasks would return the cache size held by +each executor which would be updated to the executorToCacheMapping and +the pruned blocklets which would be further used for result fetching. + +## Reallocation of executor +In case executor(s) become dead/unavailable then the segments that were +earlier being handled by those would be reassigned to some other +executor using the distribution logic. + +**Note:** Cache loading would be done again in the new executor for the +current query. + +## MetaCache DDL +The show metacache DDL has a new column called cache location will indicate whether the cache is +from executor or driver. To drop cache the user has to enable/disable the index server using the +dynamic configuration to clear the cache of the desired location. + +Refer: [MetaCacheDDL](https://github.com/apache/carbondata/blob/master/docs/ddl-of-carbondata.md#cache) + +## Fallback +In case of any failure the index server would fallback to embedded mode +which means that the JDBCServer would take care of distributed pruning. +A similar job would be fired by the JDBCServer which would take care of +pruning using its own executors. If for any reason the embedded mode +also fails to prune the datamaps then the job would be passed on to +driver. + +**NOTE:** In case of embedded mode a job would be fired after pruning to clear
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r305198024 ## File path: integration/spark2/src/main/scala/org/apache/carbondata/indexserver/IndexServer.scala ## @@ -178,6 +171,8 @@ object IndexServer extends ServerInterface { server.stop() } }) + CarbonProperties.getInstance().addProperty(CarbonCommonConstants +.CARBON_ENABLE_INDEX_SERVER, "true") LOGGER.info(s"Index cache server running on ${ server.getPort } port") Review comment: Exception is already thrown in org.apache.carbondata.core.util.CarbonProperties#getIndexServerPort This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r303736131 ## File path: docs/index-server.md ## @@ -0,0 +1,238 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) +``` +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. +``` +executorToCacheMapping = Map(HostAddress -> Map(ExecutorId -> cacheSize)) +``` + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment the index size if not available +therefore all the legacy segments would be processed in a round robin +fashion. + +After the job is completed the tasks would return the cache size held by +each executor which would be updated to the executorToCacheMapping and +the pruned blocklets which would be further used for result fetching. + +## Reallocation of executor +In case executor(s) become dead/unavailable then the segments that were +earlier being handled by those would be reassigned to some other +executor using the distribution logic. + +**Note:** Cache loading would be done again in the new executor for the +current query. + +## MetaCache DDL +The show metacache DDL has a new column called cache location will indicate whether the cache is +from executor or driver. To drop cache the user has to enable/disable the index server using the +dynamic configuration to clear the cache of the desired location. + +## Fallback +In case of any failure the index server would fallback to embedded mode +which means that the JDBCServer would take care of distributed pruning. +A similar job would be fired by the JDBCServer which would take care of +pruning using its own executors. If for any reason the embedded mode +also fails to prune the datamaps then the job would be passed on to +driver. + +**NOTE:** In case of embedded mode a job would be fired to clear the +cache as data cached in JDBCServer executors would be of no use. + +## Writing splits to a file +If the response is too huge then it is better to write the splits to a file so that the driver can +read this file and create the splits. This can be controlled using the property 'carbon.index.server +.inmemory.serialization.threshold.inKB'. By default, the minimum value for
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r303736104 ## File path: docs/index-server.md ## @@ -0,0 +1,238 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) +``` +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. +``` +executorToCacheMapping = Map(HostAddress -> Map(ExecutorId -> cacheSize)) +``` + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment the index size if not available +therefore all the legacy segments would be processed in a round robin +fashion. + +After the job is completed the tasks would return the cache size held by +each executor which would be updated to the executorToCacheMapping and +the pruned blocklets which would be further used for result fetching. + +## Reallocation of executor +In case executor(s) become dead/unavailable then the segments that were +earlier being handled by those would be reassigned to some other +executor using the distribution logic. + +**Note:** Cache loading would be done again in the new executor for the +current query. + +## MetaCache DDL +The show metacache DDL has a new column called cache location will indicate whether the cache is +from executor or driver. To drop cache the user has to enable/disable the index server using the +dynamic configuration to clear the cache of the desired location. + +## Fallback +In case of any failure the index server would fallback to embedded mode +which means that the JDBCServer would take care of distributed pruning. +A similar job would be fired by the JDBCServer which would take care of +pruning using its own executors. If for any reason the embedded mode +also fails to prune the datamaps then the job would be passed on to +driver. + +**NOTE:** In case of embedded mode a job would be fired to clear the +cache as data cached in JDBCServer executors would be of no use. + +## Writing splits to a file +If the response is too huge then it is better to write the splits to a file so that the driver can +read this file and create the splits. This can be controlled using the property 'carbon.index.server +.inmemory.serialization.threshold.inKB'. By default, the minimum value for
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r303731363 ## File path: docs/index-server.md ## @@ -0,0 +1,238 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) +``` +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. +``` +executorToCacheMapping = Map(HostAddress -> Map(ExecutorId -> cacheSize)) +``` + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment the index size if not available +therefore all the legacy segments would be processed in a round robin +fashion. + +After the job is completed the tasks would return the cache size held by +each executor which would be updated to the executorToCacheMapping and +the pruned blocklets which would be further used for result fetching. + +## Reallocation of executor +In case executor(s) become dead/unavailable then the segments that were +earlier being handled by those would be reassigned to some other +executor using the distribution logic. + +**Note:** Cache loading would be done again in the new executor for the +current query. + +## MetaCache DDL +The show metacache DDL has a new column called cache location will indicate whether the cache is +from executor or driver. To drop cache the user has to enable/disable the index server using the +dynamic configuration to clear the cache of the desired location. + +## Fallback +In case of any failure the index server would fallback to embedded mode +which means that the JDBCServer would take care of distributed pruning. +A similar job would be fired by the JDBCServer which would take care of +pruning using its own executors. If for any reason the embedded mode +also fails to prune the datamaps then the job would be passed on to +driver. + +**NOTE:** In case of embedded mode a job would be fired to clear the +cache as data cached in JDBCServer executors would be of no use. + +## Writing splits to a file +If the response is too huge then it is better to write the splits to a file so that the driver can +read this file and create the splits. This can be controlled using the property 'carbon.index.server +.inmemory.serialization.threshold.inKB'. By default, the minimum value for
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r303731403 ## File path: docs/index-server.md ## @@ -0,0 +1,238 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) +``` +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. +``` +executorToCacheMapping = Map(HostAddress -> Map(ExecutorId -> cacheSize)) +``` + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment the index size if not available +therefore all the legacy segments would be processed in a round robin +fashion. + +After the job is completed the tasks would return the cache size held by +each executor which would be updated to the executorToCacheMapping and +the pruned blocklets which would be further used for result fetching. + +## Reallocation of executor +In case executor(s) become dead/unavailable then the segments that were +earlier being handled by those would be reassigned to some other +executor using the distribution logic. + +**Note:** Cache loading would be done again in the new executor for the +current query. + +## MetaCache DDL +The show metacache DDL has a new column called cache location will indicate whether the cache is +from executor or driver. To drop cache the user has to enable/disable the index server using the +dynamic configuration to clear the cache of the desired location. + +## Fallback +In case of any failure the index server would fallback to embedded mode +which means that the JDBCServer would take care of distributed pruning. +A similar job would be fired by the JDBCServer which would take care of +pruning using its own executors. If for any reason the embedded mode +also fails to prune the datamaps then the job would be passed on to +driver. + +**NOTE:** In case of embedded mode a job would be fired to clear the +cache as data cached in JDBCServer executors would be of no use. + +## Writing splits to a file +If the response is too huge then it is better to write the splits to a file so that the driver can +read this file and create the splits. This can be controlled using the property 'carbon.index.server +.inmemory.serialization.threshold.inKB'. By default, the minimum value for
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r303731446 ## File path: docs/index-server.md ## @@ -0,0 +1,238 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) +``` +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. +``` +executorToCacheMapping = Map(HostAddress -> Map(ExecutorId -> cacheSize)) +``` + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment the index size if not available +therefore all the legacy segments would be processed in a round robin +fashion. + +After the job is completed the tasks would return the cache size held by +each executor which would be updated to the executorToCacheMapping and +the pruned blocklets which would be further used for result fetching. + +## Reallocation of executor +In case executor(s) become dead/unavailable then the segments that were +earlier being handled by those would be reassigned to some other +executor using the distribution logic. + +**Note:** Cache loading would be done again in the new executor for the +current query. + +## MetaCache DDL +The show metacache DDL has a new column called cache location will indicate whether the cache is +from executor or driver. To drop cache the user has to enable/disable the index server using the +dynamic configuration to clear the cache of the desired location. + +## Fallback +In case of any failure the index server would fallback to embedded mode +which means that the JDBCServer would take care of distributed pruning. +A similar job would be fired by the JDBCServer which would take care of +pruning using its own executors. If for any reason the embedded mode +also fails to prune the datamaps then the job would be passed on to +driver. + +**NOTE:** In case of embedded mode a job would be fired to clear the +cache as data cached in JDBCServer executors would be of no use. + +## Writing splits to a file +If the response is too huge then it is better to write the splits to a file so that the driver can +read this file and create the splits. This can be controlled using the property 'carbon.index.server +.inmemory.serialization.threshold.inKB'. By default, the minimum value for
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r303731456 ## File path: docs/index-server.md ## @@ -0,0 +1,238 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) +``` +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. +``` +executorToCacheMapping = Map(HostAddress -> Map(ExecutorId -> cacheSize)) +``` + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment the index size if not available +therefore all the legacy segments would be processed in a round robin +fashion. + +After the job is completed the tasks would return the cache size held by +each executor which would be updated to the executorToCacheMapping and +the pruned blocklets which would be further used for result fetching. + +## Reallocation of executor +In case executor(s) become dead/unavailable then the segments that were +earlier being handled by those would be reassigned to some other +executor using the distribution logic. + +**Note:** Cache loading would be done again in the new executor for the +current query. + +## MetaCache DDL +The show metacache DDL has a new column called cache location will indicate whether the cache is +from executor or driver. To drop cache the user has to enable/disable the index server using the +dynamic configuration to clear the cache of the desired location. + +## Fallback +In case of any failure the index server would fallback to embedded mode +which means that the JDBCServer would take care of distributed pruning. +A similar job would be fired by the JDBCServer which would take care of +pruning using its own executors. If for any reason the embedded mode +also fails to prune the datamaps then the job would be passed on to +driver. + +**NOTE:** In case of embedded mode a job would be fired to clear the +cache as data cached in JDBCServer executors would be of no use. + +## Writing splits to a file +If the response is too huge then it is better to write the splits to a file so that the driver can +read this file and create the splits. This can be controlled using the property 'carbon.index.server +.inmemory.serialization.threshold.inKB'. By default, the minimum value for
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r303731579 ## File path: docs/index-server.md ## @@ -0,0 +1,238 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) +``` +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. +``` +executorToCacheMapping = Map(HostAddress -> Map(ExecutorId -> cacheSize)) +``` + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment the index size if not available +therefore all the legacy segments would be processed in a round robin +fashion. + +After the job is completed the tasks would return the cache size held by +each executor which would be updated to the executorToCacheMapping and +the pruned blocklets which would be further used for result fetching. + +## Reallocation of executor +In case executor(s) become dead/unavailable then the segments that were +earlier being handled by those would be reassigned to some other +executor using the distribution logic. + +**Note:** Cache loading would be done again in the new executor for the +current query. + +## MetaCache DDL +The show metacache DDL has a new column called cache location will indicate whether the cache is +from executor or driver. To drop cache the user has to enable/disable the index server using the +dynamic configuration to clear the cache of the desired location. + +## Fallback +In case of any failure the index server would fallback to embedded mode +which means that the JDBCServer would take care of distributed pruning. +A similar job would be fired by the JDBCServer which would take care of +pruning using its own executors. If for any reason the embedded mode +also fails to prune the datamaps then the job would be passed on to +driver. + +**NOTE:** In case of embedded mode a job would be fired to clear the +cache as data cached in JDBCServer executors would be of no use. + +## Writing splits to a file +If the response is too huge then it is better to write the splits to a file so that the driver can +read this file and create the splits. This can be controlled using the property 'carbon.index.server +.inmemory.serialization.threshold.inKB'. By default, the minimum value for
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r303731377 ## File path: docs/index-server.md ## @@ -0,0 +1,238 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) +``` +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. +``` +executorToCacheMapping = Map(HostAddress -> Map(ExecutorId -> cacheSize)) +``` + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment the index size if not available +therefore all the legacy segments would be processed in a round robin +fashion. + +After the job is completed the tasks would return the cache size held by +each executor which would be updated to the executorToCacheMapping and +the pruned blocklets which would be further used for result fetching. + +## Reallocation of executor +In case executor(s) become dead/unavailable then the segments that were +earlier being handled by those would be reassigned to some other +executor using the distribution logic. + +**Note:** Cache loading would be done again in the new executor for the +current query. + +## MetaCache DDL +The show metacache DDL has a new column called cache location will indicate whether the cache is +from executor or driver. To drop cache the user has to enable/disable the index server using the +dynamic configuration to clear the cache of the desired location. + +## Fallback +In case of any failure the index server would fallback to embedded mode +which means that the JDBCServer would take care of distributed pruning. +A similar job would be fired by the JDBCServer which would take care of +pruning using its own executors. If for any reason the embedded mode +also fails to prune the datamaps then the job would be passed on to +driver. + +**NOTE:** In case of embedded mode a job would be fired to clear the +cache as data cached in JDBCServer executors would be of no use. + +## Writing splits to a file +If the response is too huge then it is better to write the splits to a file so that the driver can +read this file and create the splits. This can be controlled using the property 'carbon.index.server +.inmemory.serialization.threshold.inKB'. By default, the minimum value for
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r303731258 ## File path: docs/index-server.md ## @@ -0,0 +1,238 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) +``` +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. +``` +executorToCacheMapping = Map(HostAddress -> Map(ExecutorId -> cacheSize)) +``` + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment the index size if not available +therefore all the legacy segments would be processed in a round robin +fashion. + +After the job is completed the tasks would return the cache size held by +each executor which would be updated to the executorToCacheMapping and +the pruned blocklets which would be further used for result fetching. + +## Reallocation of executor +In case executor(s) become dead/unavailable then the segments that were +earlier being handled by those would be reassigned to some other +executor using the distribution logic. + +**Note:** Cache loading would be done again in the new executor for the +current query. + +## MetaCache DDL +The show metacache DDL has a new column called cache location will indicate whether the cache is Review comment: index server is a spark-submit application. It cannot accept sql commands This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r303731282 ## File path: docs/index-server.md ## @@ -0,0 +1,238 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) +``` +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. +``` +executorToCacheMapping = Map(HostAddress -> Map(ExecutorId -> cacheSize)) +``` + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment the index size if not available +therefore all the legacy segments would be processed in a round robin +fashion. + +After the job is completed the tasks would return the cache size held by +each executor which would be updated to the executorToCacheMapping and +the pruned blocklets which would be further used for result fetching. + +## Reallocation of executor +In case executor(s) become dead/unavailable then the segments that were +earlier being handled by those would be reassigned to some other +executor using the distribution logic. + +**Note:** Cache loading would be done again in the new executor for the +current query. + +## MetaCache DDL +The show metacache DDL has a new column called cache location will indicate whether the cache is +from executor or driver. To drop cache the user has to enable/disable the index server using the +dynamic configuration to clear the cache of the desired location. + +## Fallback +In case of any failure the index server would fallback to embedded mode +which means that the JDBCServer would take care of distributed pruning. +A similar job would be fired by the JDBCServer which would take care of +pruning using its own executors. If for any reason the embedded mode +also fails to prune the datamaps then the job would be passed on to +driver. + +**NOTE:** In case of embedded mode a job would be fired to clear the Review comment: changed the line This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards,
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r303731159 ## File path: docs/index-server.md ## @@ -0,0 +1,238 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) +``` +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. +``` +executorToCacheMapping = Map(HostAddress -> Map(ExecutorId -> cacheSize)) +``` + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment the index size if not available +therefore all the legacy segments would be processed in a round robin +fashion. + +After the job is completed the tasks would return the cache size held by +each executor which would be updated to the executorToCacheMapping and +the pruned blocklets which would be further used for result fetching. + +## Reallocation of executor +In case executor(s) become dead/unavailable then the segments that were +earlier being handled by those would be reassigned to some other +executor using the distribution logic. Review comment: distribution logic is already mentioned above This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r303731106 ## File path: docs/index-server.md ## @@ -0,0 +1,238 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) +``` +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. +``` +executorToCacheMapping = Map(HostAddress -> Map(ExecutorId -> cacheSize)) +``` + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment the index size if not available Review comment: added version This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r303731088 ## File path: docs/index-server.md ## @@ -0,0 +1,238 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) Review comment: removed This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r303730940 ## File path: docs/index-server.md ## @@ -0,0 +1,238 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, Review comment: removed This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r303730996 ## File path: docs/index-server.md ## @@ -0,0 +1,238 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the Review comment: added link to set segments doc This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r303731022 ## File path: docs/index-server.md ## @@ -0,0 +1,238 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). Review comment: changed This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r303730835 ## File path: docs/index-server.md ## @@ -0,0 +1,238 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom Review comment: changed This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r302915771 ## File path: docs/index-server.md ## @@ -0,0 +1,231 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) +``` +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. +``` +executorToCacheMapping = Map(HostAddress -> Map(ExecutorId -> cacheSize)) +``` + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment the index size if not available +therefore all the legacy segments would be processed in a round robin +fashion. + +After the job is completed the tasks would return the cache size held by +each executor which would be updated to the executorToCacheMapping and +the pruned blocklets which would be further used for result fetching. + +## Reallocation of executor +In case executor(s) become dead/unavailable then the segments that were +earlier being handled by those would be reassigned to some other +executor using the distribution logic. + +**Note:** Cache loading would be done again in the new executor for the +current query. + +## MetaCache DDL +The show/drop metacache DDL have been modified to operate on the +executor side cache as well. So when the used fires show cache a new +column called cache location will indicate whether the cache is from +executor or driver. For drop cache the user has to enable/disable the Review comment: changed This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r302915735 ## File path: docs/index-server.md ## @@ -0,0 +1,231 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) +``` +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. +``` +executorToCacheMapping = Map(HostAddress -> Map(ExecutorId -> cacheSize)) +``` + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment the index size if not available +therefore all the legacy segments would be processed in a round robin +fashion. + +After the job is completed the tasks would return the cache size held by +each executor which would be updated to the executorToCacheMapping and +the pruned blocklets which would be further used for result fetching. + +## Reallocation of executor +In case executor(s) become dead/unavailable then the segments that were +earlier being handled by those would be reassigned to some other +executor using the distribution logic. + +**Note:** Cache loading would be done again in the new executor for the +current query. + +## MetaCache DDL +The show/drop metacache DDL have been modified to operate on the +executor side cache as well. So when the used fires show cache a new +column called cache location will indicate whether the cache is from +executor or driver. For drop cache the user has to enable/disable the +index server using the dynamic configuration to clear the cache of the +desired location. + +## Fallback +In case of any failure the index server would fallback to embedded mode +which means that the JDBCServer would take care of distributed pruning. +A similar job would be fired by the JDBCServer which would take care of +pruning using its own executors. If for any reason the embedded mode +also fails to prune the datamaps then the job would be passed on to +driver. + +**NOTE:** In case of embedded mode a job would be fired to clear the +cache as data cached in JDBCServer executors would be of no use. + +## Writing splits to a file +If the response is too huge then it is better to write the splits to a file so that the driver can +read this file and create the splits. This can be controlled using the
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r302915756 ## File path: docs/index-server.md ## @@ -0,0 +1,231 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) +``` +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. +``` +executorToCacheMapping = Map(HostAddress -> Map(ExecutorId -> cacheSize)) +``` + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment the index size if not available +therefore all the legacy segments would be processed in a round robin +fashion. + +After the job is completed the tasks would return the cache size held by +each executor which would be updated to the executorToCacheMapping and +the pruned blocklets which would be further used for result fetching. + +## Reallocation of executor +In case executor(s) become dead/unavailable then the segments that were +earlier being handled by those would be reassigned to some other +executor using the distribution logic. + +**Note:** Cache loading would be done again in the new executor for the +current query. + +## MetaCache DDL +The show/drop metacache DDL have been modified to operate on the +executor side cache as well. So when the used fires show cache a new +column called cache location will indicate whether the cache is from +executor or driver. For drop cache the user has to enable/disable the +index server using the dynamic configuration to clear the cache of the +desired location. + +## Fallback +In case of any failure the index server would fallback to embedded mode +which means that the JDBCServer would take care of distributed pruning. +A similar job would be fired by the JDBCServer which would take care of +pruning using its own executors. If for any reason the embedded mode +also fails to prune the datamaps then the job would be passed on to +driver. + +**NOTE:** In case of embedded mode a job would be fired to clear the +cache as data cached in JDBCServer executors would be of no use. + +## Writing splits to a file +If the response is too huge then it is better to write the splits to a file so that the driver can +read this file and create the splits. This can be controlled using the
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r302915788 ## File path: docs/index-server.md ## @@ -0,0 +1,231 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) +``` +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. +``` +executorToCacheMapping = Map(HostAddress -> Map(ExecutorId -> cacheSize)) +``` + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment the index size if not available +therefore all the legacy segments would be processed in a round robin +fashion. + +After the job is completed the tasks would return the cache size held by +each executor which would be updated to the executorToCacheMapping and +the pruned blocklets which would be further used for result fetching. + +## Reallocation of executor +In case executor(s) become dead/unavailable then the segments that were +earlier being handled by those would be reassigned to some other +executor using the distribution logic. + +**Note:** Cache loading would be done again in the new executor for the +current query. + +## MetaCache DDL +The show/drop metacache DDL have been modified to operate on the +executor side cache as well. So when the used fires show cache a new Review comment: changed This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. For queries about this service, please contact Infrastructure at: us...@infra.apache.org With regards, Apache Git Services
[GitHub] [carbondata] kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server
kunal642 commented on a change in pull request #3294: [CARBONDATA-3462][DOC]Added documentation for index server URL: https://github.com/apache/carbondata/pull/3294#discussion_r299929028 ## File path: docs/index-server.md ## @@ -0,0 +1,216 @@ + + +# Distributed Index Server + +## Background + +Carbon currently caches all block/blocklet datamap index information into the driver. For bloom +datamap, it can prune the splits in a distributed way. In the first case, there are limitations +like driver memory scale up and cache sharing between multiple applications is not possible. In +the second case, there are limitations like, there is +no guarantee that the next query goes to the same executor to reuse the cache and hence cache +would be duplicated in multiple executors. +Distributed Index Cache Server aims to solve the above mentioned problems. + +## Distribution +When enabled, any query on a carbon table will be routed to the index server application using +the Hadoop RPC framework in form of a request. The request will consist of the table name, segments, +filter expression and other information used for pruning. + +In IndexServer application a pruning RDD is fired which will take care of the pruning for that +request. This RDD will be creating tasks based on the number of segments that are applicable for +pruning. It can happen that the user has specified segments to access for that table, so only the +specified segments would be applicable for pruning. + +IndexServer driver would have 2 important tasks, distributing the segments equally among the +available executors and keeping track of the cache location(where the segment cache is present). + +To achieve this 2 separate mappings would be maintained as follows. +1. segment to executor location: +This mapping will be maintained for each table and will enable the index server to track the +cache location for each segment. +``` +tableToExecutorMapping = Map(tableName -> Map(segmentNo -> uniqueExecutorIdentifier)) +``` +2. Cache size held by each executor: +This mapping will be used to distribute the segments equally(on the basis of size) among the +executors. +``` +executorToCacheMapping = Map(HostAddress -> Map(ExecutorId -> cacheSize)) +``` + +Once a request is received each segment would be iterated over and +checked against tableToExecutorMapping to find if a executor is already +assigned. If a mapping already exists then it means that most +probably(if not evicted by LRU) the segment is already cached in that +executor and the task for that segment has to be fired on this executor. + +If mapping is not found then first check executorToCacheMapping against +the available executor list to find if any unassigned executor is +present and use that executor for the current segment. If all the +executors are assigned with some segment then find the least loaded +executor on the basis of size. + +Initially the segment index size would be used to distribute the +segments fairly among the executor because the actual cache size would +be known to the driver only when the segments are cached and appropriate +information is returned to the driver. + +**NOTE:** In case of legacy segment the index size if not available +therefore all the legacy segments would be processed in a round robin +fashion. + +After the job is completed the tasks would return the cache size held by +each executor which would be updated to the executorToCacheMapping and +the pruned blocklets which would be further used for result fetching. + +## Reallocation of executor +In case executor(s) become dead/unavailable then the segments that were +earlier being handled by those would be reassigned to some other +executor using the distribution logic. + +**Note:** Cache loading would be done again in the new executor for the +current query. + +## MetaCache DDL +The show/drop metacache DDL have been modified to operate on the +executor side cache as well. So when the used fires show cache a new +column called cache location will indicate whether the cache is from +executor or driver. For drop cache the user has to enable/disable the +index server using the dynamic configuration to clear the cache of the +desired location. + +## Fallback +In case of any failure the index server would fallback to embedded mode +which means that the JDBCServer would take care of distributed pruning. +A similar job would be fired by the JDBCServer which would take care of +pruning using its own executors. If for any reason the embedded mode +also fails to prune the datamaps then the job would be passed on to +driver. + +**NOTE:** In case of embedded mode a job would be fired to clear the +cache as data cached in JDBCServer executors would be of no use. + + +## Configurations + +# carbon.properties(JDBCServer) + +| Name | Default Value| Description | +|:--:|:-:|:--: | +| carbon.enable.index.server | false