Re: CacheStore with replication

In this case you should mark the implementation with @CacheLocalStore annotation, which will force Ignite to pass entry version along with the value during write-through. You can then use it to maintain consistency. But this is a pretty challenging task and implementing local store is way more

Re: Two-phase commit failure recovery

This blog is not about consistency within the cache, but about keeping cache and the database consistent with each other. I.e., you can't end up with an update that was successful in cache, but not successful in the database. And automatic reloading is used to guarantee that in case of coordinator

Re: CacheStore with replication

That's the problem, the store is not shared between nodes. I have a separate native process on every cache node. This process should be notified about cache entries, that reside on this particular node. I was able to implement this behaviour by implementing CacheStore interface and calling JNI

"Tuning Apache Ignite for Optimal Performance" webinar

Ignite Users, In couple of weeks (Nov 30) I'm going to conduct a webinar about performance tuning of Apache Ignite: https://www.gridgain.com/resources/webinars/tuning-apacher-ignitetm-optimal-performance Feel free to attend and ask tough questions :) -Val

Re: Two-phase commit failure recovery

We use SQL queries that's why we have to pre-load all data. For data loading we use IgniteDataStreamer and don't use CacheStore. In case of two-phase commit failure Ignite will not be able to automatically reload invalidated data -- View this message in context:

Re: CacheStore with replication

Hi, The CacheStore implementation is assumed to be shared across nodes, so you should not have such issues. Actually I don't see any difference in this case between JNI and non-JNI implementation. What is the actual storage you're writing to? -Val -- View this message in context:

Re: "Cluster group is empty" exception thrown after client reconnect

The client should reconnect automatically after the server node is back. What is the log showing? Is it happening or not? -Val -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/Cluster-group-is-empty-exception-thrown-after-client-reconnect-tp9086p9094.html Sent

Re: Apache Spark & Ignite Integration

IgniteRDD can be used with any dataset as long as it fits in memory. If you didn't get performance from pure Ignite, I doubt that Spark will help. It's Ignite who speeds up Spark, not other way around :) I believe there are other reasons for not having acceptable performance. As for the APIs, you

Re: Partitioning on a non-uniform cluster

Hi Krzysztof, Can you give a bit more details on the business use case? What are you trying to achieve? -Val -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/Partitioning-on-a-non-uniform-cluster-tp8940p9092.html Sent from the Apache Ignite Users mailing list

Re: Binarylizable interface in apache ignite

Binary marshaller already provides very compact serialization protocol, Binarylizable will not help here, unless you want to sacrifice some of the functionality (like SQL queries, for example). Binarylizable should be used when you have some kind of a custom logic for serialization (if-else or

Re: Explicit lock whithin a transaction

What do you mean by "inter-thread async communication"? This is not happening in getAllOutTx, the whole purpose of this method is to allow reading outside of transaction without delegating it to another thread. -Val -- View this message in context:

Re: setAtomicSequenceReserveSize(0) doesnt work. It reserves a minimum of 2.

Hi, Please properly subscribe to the mailing list so that the community can receive email notifications for your messages. To subscribe, send empty email to user-subscr...@ignite.apache.org and follow simple instructions in the reply. Rashmika wrote > AtomicConfiguration a = new

CacheStore with replication

Hi, all! I am trying to use CacheStore for making a JNI call on every put opperation. It works fine, but i am concerned about replication and rebalancing. Local native process i am calling should be synchronized with cache entries located on this exact node, but when rebalancing occurs neither

"Cluster group is empty" exception thrown after client reconnect

Hi, I have a 3 node cluster and connect to it with a client. All works well until I kill one of the nodes on the cluster. I can catch the fact that the client has been disconnected/reconnected by listening to the events: internal class LocalEventListener : IEventListener {

Re: Failed to send message to remote node error when remote node has been terminated

Hi Val, Thanks for clearing that up. I am catching the exception by checking the exception message with: if (ex.Message.Contains("Failed to send message to remote node:")) { ... } as I only have the "catch all" ApacheIgniteException class - there seems to be no specific exception I can trap.

Re: Can java code look up cache in server without Ignition.start(cfg)

Hi Val, That was the perfect one for me. It worked with XML using Ignition.loadSpringBean() . Appreciate all your help in this regard. Thank You ! Thanks Saji On Thu, Nov 17, 2016 at 3:27 PM, vkulichenko [via Apache Ignite Users] < ml-node+s70518n9057...@n6.nabble.com> wrote: > Saji, > >

Re: Memory consumption in apache ignite

Please describe how data are stored on disk that you measured as 370 Mbytes? Are they compressed? How many fields each entry has? The comparison may not be correct. Ignite stores entries in a different way than relational database, f.e. for each field it spends 4 bytes, so if you have 10 fields,

Re: Apache Spark & Ignite Integration

a. No. I have not executed any tests. I am doing a theoretical study first to understand the memory footprint and data movement between the spark & ignite nodes c. So basically there is no use case when working with spark (for data processing) and ignite (for in-memory data storage) that can

Re: Memory consumption in apache ignite

I did modifications as you suggested. But, still memory consumption is about 2 GB, what might be reason for this? Memory consumption before data loading Memory consumption after data loading

Re: Binarylizable interface in apache ignite

Hi. Actually to make your data just a bit more compact you may use org.apache.ignite.binary.BinaryRawWriter and org.apache.ignite.binary.BinaryRawReader (writer.rawWriter() and reader.rawReader()), but in this case you'll not be able to use SQL queries. Binarylizable interface gives you a way to

Re: Memory consumption in apache ignite

You don't have to deploy your data classes on server, because it may work with binary types. Just configure cache as usual but use QueryEntity[1] for that. On client side you may put and query for data as you did before. [1]:

Binarylizable interface in apache ignite

I am trying to use Binarylizable interface in apache ignite for reducing memory utilization(for compact representation of data). Following is data class that I have made binarylizable. class order_line implements Binarylizable{ @QuerySqlField(index = true) int ol_o_id;

Re: Partitioning on a non-uniform cluster

For the second option: would not it cause the over-utilisation of CPUs on the nodes with two servers running? -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/Partitioning-on-a-non-uniform-cluster-tp8940p9076.html Sent from the Apache Ignite Users mailing list

Re: Low self-descriptiveness of query exceptions.

All we can do here is to propagate error message from H2. Probably it will give more details. Sergi 2016-11-18 13:31 GMT+03:00 Alexey Kuznetsov : > Vasiliy, Pavel. > > I think it is a usability issue. > Not always user has a full stack trace. > For example, if I execute

Re: Low self-descriptiveness of query exceptions.

Vasiliy, Pavel. I think it is a usability issue. Not always user has a full stack trace. For example, if I execute this query via some UI tool - in most cases I will see only top level message. I have no idea how hard to implement such functionality. Sergi (as a most expert in SQL engine) could

Re: Memory consumption in apache ignite

I have measured database size on disk 370MB -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/Memory-consumption-in-apache-ignite-tp9035p9073.html Sent from the Apache Ignite Users mailing list archive at Nabble.com.

Re: Low self-descriptiveness of query exceptions.

Vasiliy, did you examine the full stack trace? I've tried an invalid query and got the following in one of the inner exceptions: Caused by: org.h2.jdbc.JdbcSQLException: Table "ORGANIZATION2" not found However, I agree that this should be present in the top level exception message. On Fri, Nov

Re: Memory consumption in apache ignite

But for data storage required data classes must be defined on server node. What do you mean by "without data classes"? Can you give me example? -- View this message in context: http://apache-ignite-users.70518.x6.nabble.com/Memory-consumption-in-apache-ignite-tp9035p9071.html Sent from the

Re: Memory consumption in apache ignite

For sure, if you disable indexing SQL query will not be available. It was just recommendation if it's suitable for you. You may try as Vlad suggested: use server node without data classes, so it will not be able to cache deserialized values, and run your application on client node. On Fri, Nov

Re: Explicit lock whithin a transaction

I do know objects which I'm going to update in advance, but I don't know objects which I'm going to read. The getAllOutTx() can be used, but I don't like it. First of all performance would be much worse (it involves inter-thread async communication). I need read lot of objects but only very few

Re: Two-phase commit failure recovery

Problem is we can't use standart cache loading due to http://apache-ignite-users.70518.x6.nabble.com/SQL-query-against-cache-with-read-through-CacheStore-td9039.html We need implement some custom logic to handle this situation: report the problem to administrator, switch-off cache usage,