[cpp] Setup a near cache on client and server nodes

Hello. Configuration: 1. N1 и N2 - data nodes (in different k8s clusters). 2. C1 - client node (outside of k8s clusters; c++ client node). 3. All caches are replicated. Caches create through REST/caches templates. Need: 1. C1 has to have a near cache for better performance. Question:

Race condition and conflicts during cache modifications?

Hello everyone, I have a few questions: 1) How are cache entry merge conflicts handled? Does the newer topology version of the entry always win? Can the cache conflict resolver be changed to a custom implementation? 2) Imagine a *cache* which uses key=String and *value=List* in *replicated* mode

Slowness in initial data load using COPY command

HI, I was trying to load data using csv file (each file contains 5 million rows of record which is approx.. 4 GB of data) using COPY command. In initial stage of loading it was quit fast, but later the load process start slowing down and showing hardly 1% of CPU usage. My cluster has 5 nodes, ea

Re: Slowness in initial data load using COPY command

Hello. The same thing I see when I move data from MariaDB to Apache Ignite (PutAll). I have tens of databases, but such things happens for a 100k rows database and for 400k+ rows databases as well. I tried to: 1. Increase threads 2. Disable WaL/persistance. 3. Put WaL/data on a SSD. Nothi

RE: Slowness in initial data load using COPY command

Thanks Oleg, I need to have persistence enabled for my system. I believe due to the locking/wait issue, the data load is slowing down. Can anyone suggest a solution on how to avoid this situation and make data load faster? Note: I am using COPY command to lad data into table from csv. Regards,

Re: Slowness in initial data load using COPY command

Hi, You can try to investigate your cluster state first: 1)First of all try to take a look at your logs and find there long JVM pauses. Also, you can collect the GC log and see the memory usage. It possible that the provided memory for your node isn't enough because it was used for something

Ignite query performance with lots of joins

Hi, I posted a message related to H2 & Ignite performances on stackoverflow. As the main issue is about a difference in execution times between H2 & Ignite so I thought it would be good to relay it here. Thanks for you

Re: Cache expiry policy not deleting records from disk(native persistence)

Shiva, Hopefully, someone from the dev community will pick this ticket up soon and solve the task. In the meantime, Artem, would you mind documenting this limitation referring to ticket 10862? - Denis On Tue, Sep 10, 2019 at 12:50 AM Shiva Kumar wrote: > I have filed a bug https://issues.apac