Re: Triggering Rebalancing Programmatically get error while requesting

Ilya, have you had a chance to look into threaddumps? --Yakov ср, 27 мар. 2019 г. в 06:18, luongbd.hust : > Thank you for your enthusiasm > > I attached the logs for a longer time after the error occurred. > > logs.rar > > > >

Re: JVM tuning parameters

If you have all tables colocated with each other, i.e. having the same affinity function and backups count then you can try using cache groups - https://apacheignite.readme.io/docs/cache-groups --Yakov пт, 29 мар. 2019 г. в 08:49, ashfaq : > We are testing and also the production data would be

Re: Partition eviction failed, this can cause grid hang. (Caused by: java.lang.IllegalStateException: Failed to get page IO instance (page content is corrupted))

Hi! The upcoming 2.6 should fix the problem. It is under vote now. --Yakov

Re: destroy cache holding residual metadata in memory (2.7)

Wayne, can you please share a reproducer for this problem that can be launched from IDE? --Yakov

Re: destroy cache holding residual metadata in memory (2.7)

Wayne, can you please share a reproducer for this problem that can be launched from IDE? --Yakov

Re: JDK 11 support

Ignite 2.7 should run on JDK 11. Vladimir, Denis, correct me if I'm wrong. Please try this link for information - https://apacheignite.readme.io/v2.7/docs/getting-started#section-running-ignite-with-java-9 --Yakov

Re: JDK 11 support

Of course jdk 8 is fine :). I meant that ignite should also be fine with jdk9 and newer. Yakov

Re: Local node SEGMENTED error causing node goes down for no obvious reason

It seems there were issues with network. You can see plenty of discovery warnings in logs of the kind (231.log): [2018-11-07T07:44:44,627][WARN ][grid-timeout-worker-#119][TcpDiscoverySpi] Socket write has timed out (consider increasing 'IgniteConfiguration.failureDetectionTimeout' configuration p

Re: Cannot start/stop cache within lock or transaction

Naresh, The case you described should work. Can you please create a reproducer that I can run locally? I also filed this to make best effort to avoid deadlocks - https://issues.apache.org/jira/browse/IGNITE-6380 --Yakov

Re: Cannot start/stop cache within lock or transaction

Thanks, we'll take a look! --Yakov

Re: Cannot start/stop cache within lock or transaction

I see that you call cache.clear() after acquiring explicit lock on key "fake". This cause cache clear to hang and this in turn prevents lock.unlock() call. Explicit lock also holds the exchange process. If I comment out cache.clear() test passes without deadlock. --Yakov

Re: Ignite cluster with persistent store enabled did not load from wal after restarting.

Guys, it seems that we have the same issue that was described in thread - Specifying location of persistent storage location I filed a ticket some time ago - https://issues.apache.org/jira/browse/IGNITE-6285, Alex G, can you take a look if this ticket makes sense? --Yakov

Re: Fetched result use too much time

-considerations Thanks! -- Yakov Zhdanov, Director R&D *GridGain Systems* www.gridgain.com 2017-09-15 4:50 GMT+03:00 Lucky : > Hi > I have a table with 25,000,000 records. > The sql like this: > select fdatabasedid from databasedassign where fassingcuid in(3589

Re: Ignite 2.0.0 GridUnsafe unmonitor

+1 for warning about potential deadlock and improving javadocs --Yakov

Re: Prohibit "Node is out of topology"

Lukas, what about GC pauses in your cluster? What is the max? What are network specs? Also, do you use bare metal servers or virtual machines? --Yakov

Re: Runtime.availableProcessors() returns hardware's CPU count which is the issue with Ignite in Kubernetes

Cross-posting to dev list. Guys, Suggestion below makes sense to me. Filed a ticket https://issues.apache.org/jira/browse/IGNITE-7310 Perhaps, Arseny would like to provide a PR himself ;) --Yakov 2017-12-26 14:32 GMT+03:00 Arseny Kovalchuk : > Hi guys. > > Ignite configures all thread pools,

Re: Runtime.availableProcessors() returns hardware's CPU count which is the issue with Ignite in Kubernetes

Ilya, agree. I like IGNITE_AVAILABLE_CPU more. Yakov Zhdanov, www.gridgain.com 2017-12-26 16:36 GMT+03:00 Ilya Lantukh : > Hi Yakov, > > I think that property IGNITE_NODES_PER_HOST, as you suggested, would be > confusing, because users might want to reduce amount of available res

Re: ZooKeeper Based Discovery

Dmitry, there were few threads on dev list discussing new discovery SPI implementation. You can find current code in "ignite-zk" branch. The work is still in progress, but is mostly ready and will be merged to master soon. Documentation and algorithms descriptions are still on TODO list. Stay tune

Re: Upgrade from 2.1.0 to 2.4.0 resulting in error within transaction block

Cross posting to dev. Vladimir Ozerov, can you please take a look at NPE from query processor (see below - GridQueryProcessor.typeByValue(GridQueryProcessor.java:1901))? --Yakov 2018-03-29 0:19 GMT+03:00 smurphy : > Code works in Ignite 2.1.0. Upgrading to 2.4.0 produces the stack trace > below

Re: IgniteDataStreamer.flush() returns before all futures are completed

David, can you try adding all the futures to a collection and change assert condition to check that for all futures in that collection fut.isDone() is true? This should be a proper assertion. Please let me know if it works. Yakov Zhdanov чт, 26 апр. 2018 г., 0:30 David Harvey : > We

Re: IgniteDataStreamer.flush() returns before all futures are completed

to make this > behavior clear. > > -DH > > On Wed, Apr 25, 2018 at 5:43 PM, Yakov Zhdanov > wrote: > >> David, can you try adding all the futures to a collection and change >> assert condition to check that for all futures in that collection >> fut.isDon

Re: C# client too slow to connect.

Hi, slow node join is possible on windows machines which is related to tcp/ip stack differences from linux - windows implementation waits for the entire timeout which is 3000ms by defaull and linux returns immediately with "connection refuse". Just narrow the range - 10.200.20.90:47500..47509, try

Re: [TcpCommunicationSpi] Close incoming connection, unknown node

Hi! >From this log [1] I see that local node (id=06b8bcb7) reported 18 node failed events 1 for server node and 17 for clients. Communication sub-system closes incoming connections from nodes that are not in topology. We need to understand: 1. Why this node failed? Probably it started segmentatio

Re: How to solve the 22 parameters' limit under scala 2.10 in the case class?

Hi! Can you please provide the code you are trying to execute and full logs of execution. Seems that IgniteContext was not properly initialized. -- Yakov

Re: Messaging Reliability

Philipp, Messaging works on top of Ignite internal communication sub-system. Message delivery is guaranteed, unless recipient node has not left the grid. If it leaves (i.e. topology gets EVT_NODE_LEFT or EVT_NODE_FAILED), then sender clears internal queue associated with left node. When node rece

Re: Messaging Reliability

n? > And what happens if a new node is joining the cluster while a message is > being sent? > > If it makes sense, I could create a pull request to add this information > to the javadoc? > > Philipp > > ------ > *From:* Yakov Zhdanov [yzhda...@ap

Re: How to solve the 22 parameters' limit under scala 2.10 in the case class?

Thanks for sharing the code. Alex Goncharuk, can you please take a look? --Yakov

Re: Behavior of init() for clustered singleton

Your examples seems correct to me. 1. What does it mean by "goes null"? 2. I do not see any assignments other than instantiation in init() method. 3. You confirm that service worked OK on some node but after some time with no topology changes it starts to throw NPE, correct? Can you please share th

Re: Easy way to determine a node has been segmented

You can subscribe to event - org.apache.ignite.events.EventType#EVT_NODE_SEGMENTED with org.apache.ignite.IgniteEvents#localListen and take action inside listener. Please note that listener is called in sensitive internal system thread, so make sure to move all complex and blocking processing to an

Re: Unexpected heap usage by GridDhtPartitionTopologyImpl

Guys, I like the idea of separate pool. But, please note that write-behind pool may be slower and may not be able to flush all cache updates to DB. We will have to force system threads to help with this. Tolga, I know that Nick is currently working on async cache callbacks and he will be introduci

Re: Behavior of init() for clustered singleton

Guys, It seems there can be a race condition between service methods call and initialization - org/apache/ignite/internal/processors/service/GridServiceProcessor.java:921 Alex G, Val, can you please check if service may be called prior to its initialization? Dan, can you please add service insta

Re: Behavior of init() for clustered singleton

Please provide app logs after the issue gets reproduced. --Yakov 2016-04-08 19:20 GMT+03:00 dstieglitz : > Ok I added the debug statements: > > > https://github.com/dstieglitz/grails-ignite/blob/v0.4.x/src/java/org/grails/ignite/DistributedSchedulerServiceImpl.java > > Let me know if you want me

Re: ODBC Driver?

It is possible. I think we can send 1.6 for vote as soon as odbc driver gets finished. --Yakov 2016-04-27 13:51 GMT+03:00 Tomek W : > Does it mean that in next week ODBC driver will be available ? > > 2016-04-27 5:41 GMT+02:00 arthi : > >> Thank you... >> >> will this driver support rowset bindi

Ignite Reentrant Lock

Guys, Vlad Jelisavcic has implemented Ignite reentrant lock ( https://issues.apache.org/jira/browse/IGNITE-642). I have finally reviewed and merged it today. Vlad, thank you for your efforts! I look forward to next contribution from you! I have removed origin/ignite-642 branch. Please close PR an

Re: jpoint presentation

Nick, I am not sure if apache lists allow attachments. So, I will send you personal email. Thanks! -- Yakov Zhdanov, Director R&D *GridGain Systems* www.gridgain.com 2016-05-05 13:52 GMT+06:00 Nick Pavlov : > Hi Yakov, > > Could you share your presentation from jpoint? >

Re: jpoint presentation

Tuesday, May 10, 2016 3:56 PM, Yakov Zhdanov > wrote: > > > Nick, I am not sure if apache lists allow attachments. So, I will send you > personal email. > > > Thanks! > -- > Yakov Zhdanov, Director R&D > *GridGain Systems* > www.gridgain.com > > 2016-0

Re: Data Nucleus plugin extensions aren't loaded by the grid class loader

Thank you for your interest to the project! Unfortunately, I have no idea what is wrong with your case. Can you please share logs from all Ignite nodes so I can get minimal idea on what is going on? Do you see exceptions in logs? --Yakov 2016-05-13 14:12 GMT+06:00 georgeg : > Hi everyone, > >

Re: How to troubleshoot a slow client node get()

What OS do do you use? Is this delay observed only initially or can also be observed after leaving client idle after doing some work? Can you please disable shared memory communication in case you use Linux or Mac - set org.apache.ignite.spi.communication.tcp.TcpCommunicationSpi#setSharedMemoryPor

Re: How to troubleshoot a slow client node get()

Tracel, can you please also take threaddumps of both client and server JVMs when you see that get() takes too long? --Yakov 2016-07-13 12:33 GMT+03:00 tracel : > Thanks yakov, > I am using a Linux box. > The delay can also be observed after leaving client idle for some 3 to 10 > minutes. > > I w

Re: DataStreamer performance question

Case of 1 client and 1 server should be about 2 times slower than 2 servers, since in the second case 50% of keys belong to local node and there is no need to send them over network. In the first case 100% keys and values go over network. 1. I would think you are running out of memory on server no

Re: How to troubleshoot a slow client node get()

I see these possible reasons of the delay 1. GC pause on client or server. Please enable GC logging and rerun test. 2. your machine runs out of resources - CPU, memory, disk IO, etc. Can you please check that there are enough resources in the system? 3. Is any energy saving stuff enabled? I also d

Re: Deadlock during Client Continuous Query deserialization

Guys, here are my comments. As far as ticket filed by Sam. If reads are paused at some point they should get unpaused when incoming message queue gets shorter. The issues's description makes me think that Ross has very heavy logic in the listener, so notification processing takes too long. Ross,

Re: Deadlock during Client Continuous Query deserialization

Ross, Deserialization may be heavy. When you deserialize object Ignite implicitly goes to some internal caches to get metadata of the type. This explains why your example works when you skip deserialization. I am not sure whether 2000 is a lot or not. For me even 1 cont query bringing all the upd

Re: Ignite performance

Manuel, The numbers you are seeing are pretty strange to me. Is it correct that you run your query in a loop giving enough time for the whole cluster to warmup and only then take the final measurements? I also do not understand why CPU load is 400% which may be interpreted as full (correct?). Thi

Re: Ignite performance

gt; > > > Ignite configuration: > > class="org.apache.ignite.configuration.IgniteConfiguration"> > > > > > > > > > > > > class="org.apache.ignite.spi.discovery.tcp.TcpDiscoverySpi"

Re: Javadoc at GridGain website is at version 1.6 instead of 1.7.1

Hi Kenan! I reached 1.7.0 API docs from ignite.apache.org site. Thanks for letting us know anyway. --Yakov 2016-09-15 19:14 GMT+03:00 Kenan Dalley : > The Javadoc link at the GridGain website hasn't been updated to the latest > release of 1.7.1 and is still referring back to v 1.6.0. This is

Re: Is Apache Tomcat 8 supported for Web Session Clustering yet?

I don't see any reason why Tomcat 8 should not be supported. We use servlet API only. Can you please give it a try and let us know? --Yakov 2016-09-19 11:44 GMT+03:00 yucigou : > According to the Web Session Clustering documentation: > https://apacheignite.readme.io/docs/web-session-clustering,

Re: java.util.concurrent.CompletionService implementation

Why don't use JDK provided service? ExecutorCompletionService service = new ExecutorCompletionService(dr1Hub.executorService()); --Yakov 2016-10-07 15:50 GMT+03:00 mnadeem : > Appreciate if java.util.concurrent.CompletionService implementation is > provided as in Infinispan > (org.infinispan.d

Ignite service for redhat/centos

Hello Igniters! As far as I know many of Ignite deployments run under rhel & centos I think it would be very convenient to include sample service configuration and service installation instructions to our distribution. Here is the link to redhat documentation. https://access.redhat.com/document

Re: Some problems in test case which comparing sql query performance between Ignite and Oracle

Bob, can you please also tell what is your benchmark scenario? Do you have enough warm-up cycles to bring your app to some steady state? --Yakov 2016-10-20 12:40 GMT+03:00 Vladislav Pyatkov : > Hi Bob, > > One way to do SQL faster this is adding indexes. > 1) I do not think what the estimation w

Re: Threads got stuck

Alper, There can be multiple reasons. Can you please reproduce the issue one more time, collect and share the following with us: 1. collect all the logs from all the nodes - clients and servers 2. take threaddumps of all JVMs (from all nodes) with jstack -l --Yakov 2016-10-25 10:49 GMT+03:00

Re: Threads got stuck

he > data to be replicated to others. Can that cause such a situation, too? > > On Tue, Oct 25, 2016 at 11:14 AM, Yakov Zhdanov > wrote: > >> Alper, >> >> There can be multiple reasons. >> >> Can you please reproduce the issue one more time, collect an

Re: Performance question

I think it is a bad idea to set number of threads greater than CPU count given you do not block on any IO operations However, benchmark itself seems totally incorrect to me. The most important issues are 1. total time is too low. try to put it to a loop body and measure each iterations. you can al

Re: Not able to join cluster with Zookeeper based IP finder

for some reason these two nodes register themselves with incorrect addresses or, to be more exact, incorrect ports - 47101 instead of 47500. Then they both get 2 addresses (1st is own address, 2nd is address for remote node) but cannot connect to them since port is incorrect. 2016-12-06 16:00:25,2

Re: Update a column of cache entities - Ignite 1.7

You can send IgniteCompute.affinityCall() to the primary node that holds the partition you are interested in. Inside the closure you can have iteration and invoke() calls that will be local to the primary copy. If you need to process the entire cache data set then send as many affinityCall's as you

Re: Not able to join cluster with Zookeeper based IP finder

, then remove it, since it is misuse. Ignite automatically picks unused port and propagates this info to the rest of the cluster. Thanks! -- Yakov Zhdanov, Director R&D *GridGain Systems* www.gridgain.com 2016-12-08 5:03 GMT+07:00 ghughal : > Yakov, > > We are configuring 47101 port

Re: AtomicSequence not performing well with 2 nodes cluster set up

GB, I am not sure if you see any difference since you will be updating the same cache entry which does not scale much. Sequence and IgniteAtomicLong work exactly in the same way. --Yakov 2016-12-08 22:15 GMT+07:00 GB : > Hi, > What if we create an ignite cache having "SeqId" and increment its va

Re: Not able to join cluster with Zookeeper based IP finder

Nikolay, can you please join this thread and point out how to build cluster using docker? --Yakov 2016-12-10 5:48 GMT+07:00 ghughal : > Yakov - I'll try to see if I can remove AddressResolver. The main reason > for > adding it was because we are running it inside docker container and if we > run

Re: Service RPC guarantees

Evgeny, you can take a look at org.apache.ignite.internal.processors.service.GridServiceProxy#invokeMethod - service method maybe invoked more than once on different service instances on different nodes. --Yakov 2016-12-27 11:59 GMT+03:00 Evgeniy Ignatiev : > Hello. > > I have a question on Serv

Re: Service RPC guarantees

: > Thanks for pointing out. As far as I can see the same applies to a > clustered singleton service - its method may be invoked more than once per > a single call? > > On 12/27/2016 2:42 PM, Yakov Zhdanov wrote: > > Evgeny, you can take a look at org.a

Re: Make changes to cache entry before data rebalanicng.

Igor, it is impossible to alter the value since it may cause data inconsistency. Imagine you have 2 nodes and cache with 1 backup. 3 node joins and you alter cache entries being sent from node 1 and 2 to 3, but those entries are currently backed up on 2 and 1 respectively. Please explain why you

Re: Efficient way to get partial data of a cache entry

I would suggest calling cache.invoke() and return needed result without altering the entry. Or compute.affinityCall() and do local get inside callable and computing and returning the result.

Re: Trigger query in ignite

You can also use jdk timers. --Yakov 2017-01-06 13:15 GMT+03:00 rishi007bansod : > Is there any way we can trigger query at some fixed time interval in > ignite(So that query is written only single time initially and it gets > called at defined time interval). Is there any event parameter where

Re: Locking behavior

1. Lock is acquired on per entry (per-key) basis. Can you please provide threaddumps and logs from all of the nodes. Note, that locks are bound to the thread they are acquired in. 2. It seems the answer is NO. One option though - you can stop the node holding the lock, i.e. node where cache.lock(k

Re: Locking behavior

This is the trace of the waiting thread. This means there is some thread in the cluster that holds the lock. Is it possible to determine which one and on which jvm and why this lock has not been released. > We did not use transactions because we did not want to have to handle the condition of what

Re: Data Rebalancing & Transactions

Val, NONE means rebalancing is disabled. I think we will remove ability to set rebalance mode in 2.0. I think async is always fine. --Yakov 2017-01-19 5:59 GMT+03:00 vkulichenko : > Sam, > > NONE mode is something that should be used very accurately. It means what > it > means - no rebalancing i

Re: Data Rebalancing & Transactions

Sam, you can control rebalancing with CacheConfiguration.setRebalance***() properties. Take a look and rebalanceDelay, batchsize and prefetch count. You can disable rebalancing by setting negative delay - this way you will have to manually trigger rebalancing from API or Visor. > 2. PUT will re-ba

Re: Timed out waiting for message delivery receipt

Hello! What environment do you use? Can you please make sure you do not have any firewall/connectivity/iptables issues? --Yakov

Re: Timed out waiting for message delivery receipt

Very good! Please keep us up to date. --Yakov 2017-01-24 9:30 GMT+03:00 hitendrapratap < hitendrapratapsingh.si...@target.com>: > I has been resolved when increased "ignite.failure.detection.timeout" to > 60sec from 10sec. > > > > -- > View this message in context: http://apache-ignite-users. >

Re: Asynchronous jobs (not the scheduling)

Sergei, Vlad, I think you also might be interested in compute continuations since the result of async execution may be sufficient for the job itself. Please see ComputeFibonacciContinuationExample under ignite/examples. --Yakov

Re: Cache write behind optimization

Tolga, this looks like you do cache.get() and key resides on remote node. So, yes, local node waits for response from remote node. --Yakov 2017-02-21 10:23 GMT+03:00 Tolga Kavukcu : > Hi Val,Everyone > > I am able to overcome with write behind issue and can process exteremly > fast in single nod

Re: 10X decrease in performance with Ignite 2.0.0

Chris, that's very surprising. Let's get to the bottom of it. 1. You have 10 nodes with 4 partitioned caches each configured with 2 backups. There are no updates to the caches. Correct? 2. Are caches atomic or transactional? 3. 5 nodes send compute jobs to the topology of 10 nodes. Correct? 4. How

Re: 10X decrease in performance with Ignite 2.0.0

How many values are in your cache? SLB? What is it? >Will there be any difference between 1.9 and 2.0 if you send the same >number of empty jobs? >None. Exact same code . I want you to check if compute engine became slower in your deployment. If you comment out all lookups inside the job and run

Re: 10X decrease in performance with Ignite 2.0.0

Yakov Zhdanov : > How many values are in your cache? > > SLB? What is it? > > >Will there be any difference between 1.9 and 2.0 if you send the same > >number of empty jobs? > >None. Exact same code > . > I want you to check if compute engine became slower in y

Re: 10X decrease in performance with Ignite 2.0.0

Cross-posting to devlist. --Yakov

Re: 10X decrease in performance with Ignite 2.0.0

Absolutely agree here. I think if we can add getAll() benchmark and run it with batch sizes of 5 and 10. Thanks! -- Yakov Zhdanov, Director R&D *GridGain Systems* www.gridgain.com 2017-05-12 10:48 GMT+03:00 Sergi Vladykin : > According to our benchmarks Ignite 2.0 is not slower

Re: 10X decrease in performance with Ignite 2.0.0

I think it will also be useful to switch to offheap tiered (cacheConfig.setMemoryMode()) in 1.9 and compare results again. --Yakov 2017-05-12 11:30 GMT+03:00 Alexey Goncharuk : > Hi Chris, > > One of the most significant changes made in 2.0 was moving to an off-heap > storage by default. This me

Re: 10X decrease in performance with Ignite 2.0.0

Chris, any news? --Yakov 2017-05-13 1:05 GMT+03:00 Denis Magda : > Chris, > > These are some links for reference: > > 1. BinaryObject and BinaryObjectBuilder interfaces usage: > https://apacheignite.readme.io/docs/binary-marshaller# > section-binaryobject-cache-api > https://apac

Re: How to return ID of Ignite task to stop this task in the future?

Couple of questions to Igniters (cross-posting to dev list). 1. Why IgniteCompute does not return ComputeTaskFuture from all methods? 2. In GridGain 6.x and earlier there was a method GridComputeImpl#cancelTask(GridUuid sesId) - https://github.com/gridgain/gridgain/blob/cf56a35cbae442d11181f58004

Re: CacheWriteSynchronizationMode

Reid, I don't think you can ensure that local node is primary for all updates. In theory you can implement affinity function having only 1 primary node for all partitions, but once you initiate replicated cache update from another node then primary node becomes remote. As far as LOCAL_SYNC - I am

Re: Why is custom cacheStore.write() being called in clientMode?

In order to commit DB transaction consistently with in-memory one it is committed from near node (which is client node in your case). Currently we don't have an option to write-through from dht nodes, but for now you can try write behind functionality - https://apacheignite.readme.io/docs/persiste

Re: Why is custom cacheStore.write() being called in clientMode?

In Ignite terminology near node is the node that coordinates transaction or cache update. As far as store initialization you can implement org.apache.ignite.lifecycle.LifecycleAware on your store and inject Ignite instance to your store with org.apache.ignite.resources.IgniteInstanceResource Init

Re: Transaction Boundary - Data Streamer

>>If you need transactional in your operations, you can use optimistic&serializable transactions by implementing and registering a receiver into your datastreamer as explained here >> https://apacheignite.readme.io/docs/transactions#deadlock- free-transactions >> and for datastreamer here https://

Re: Transaction Boundary - Data Streamer

2017-06-26 13:28 GMT+03:00 Yakov Zhdanov : > >>If you need transactional in your operations, you can use > optimistic&serializable transactions by implementing and registering a > receiver into your datastreamer as explained here > > >> https://apacheignite.readme.i

Re: Transaction Boundary - Data Streamer

2017-06-26 15:27 GMT+03:00 fatality : > Hi > > We have also implemented a generic receiver as > TimestampBasedUpdateStreamReceiver and also > AbstractTransactionalStreamReceiver. I send you the source code if you > want > and maybe can be added to the ignite source code. > > Please post your code

Re: Fetch column names in sql query results

Megha, this will be available in 2.1. You can see the ticket for details - https://issues.apache.org/jira/browse/IGNITE-5252 --Yakov

Re: Read-through cache performance

Pascal, Why do you have expiry policy in one run and do not have it in another? Please remove it and rerun your test. Also, can you please wrap your loop in another loop and take 60 measurements in a row. --Yakov

Re: Read-through cache performance

Any reason why you left expiry policy? --Yakov

Re: About Apache Ignite in Japanese

Wow! Very cool! Great job, Roman! --Yakov

Re: Apache Ignite Site Changed its Face

I like the new design. Looks good! --Yakov

Re: Ignite Benchmark has very low CPU utilization, why?

Michael, according to the first message in this thread all machines have 36 cores. The thing is that client threads call cache operation and wait for response from server. So, most of the time client is waiting. I would have at least 36 * 4 threads on client side. And also start more clients. I thi

Re: Possible starvation in striped pool. message

Hi! Ignite prints stacktrace of an internal thread if it detects that thread runs the same task (single get response in your case) for a long time, which is 30 seconds by default. Of course, getting large objects from remote nodes is expensive. Can you please estimate the value size? Possible wa

Re: Possible starvation in striped pool. message

ot;f"));}); This should work fine for both transactional and atomic caches. Thanks! -- Yakov Zhdanov, Director R&D *GridGain Systems* www.gridgain.com 2017-08-09 16:30 GMT+03:00 kestas : > I did some testing on #invoke vs #get performance. It works as expected on > ATOMIC cache,

Testing Ignite Applications Locally

Guys, I want to discuss the subject again. It is pretty vivid that having wide set of local unit and simple integration tests most likely help to avoid many failures and bugs when going to server environment. I participated in many POC and I can say for sure - if developers are not implementing l

Re: Testing Ignite Applications Locally

Sergey, can you please elaborate? --Yakov 2017-08-26 23:51 GMT+03:00 Sergey Kozlov : > The idea is great! > > Also I would suggest an ability to run new (modified) tests 100 times in > loop on the CI server to make sure that they don't cause no sporadic > failures (we can include that as part of

Re: Testing Ignite Applications Locally

>As far as Maven archetype, Yakov, is the only purpose of it to load a project, so users can add tests to it? Yes, but not just add. We will also be providing test examples, data loading application sample, shell scripts, etc. This can be pretty handy thing to create project with all dependencies

Re: Specifying location of persistent storage location

+ dev Pavel Tupitsin, can you please check that org.apache.ignite.configuration.IgniteConfiguration#setConsistentId has it platform counterpart? I could not find it. Raymond, you can explicitly set a bind address for Ignite with public string Localhost { get; set; }. This will make consistent ID t

Re: Specifying location of persistent storage location

Raymond, I think most of Ignite users run on ipv4. So, issues with v6 in this case are hardly possible =) --Yakov

Re: ignite.active(true) blocking forever

Slava, this looks like the issue with start cache operation throwing exception on some nodes and cache start rollback. I remember you implemented that fix. Can we apply the same machinery to rollback the activation with proper exception? --Yakov

Re: Task management - MapReduce & ForkJoin performance penalty

Guys, I see the following issues with the benchmark: 1. There is only one iteration. I would put it in a loop and measure at least hundred of iterations. 2. no-op jobs are not real world example at all =) job requests are processed in thread pool executor which is not very much effective for such

  1   2   >