[ 
https://issues.apache.org/jira/browse/IGNITE-15448?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
 ]

Alexey Scherbakov updated IGNITE-15448:
---------------------------------------
    Description: 
We should test the performance of a JRAFT port in comparison with original 
implementation [1].

Where are two possible scenarios:

1. Testing single group performance using simple state machine and various 
client threads count.

[2] can be used as a starting point

2. Testing key-value API performance in scenarios like [3]

For optimal performance per-partition batches must be used together with putAll 
operation.

[1] [https://github.com/sofastack/sofa-jraft]

[2] 
org.apache.ignite.raft.server.ITJRaftCounterServerTest#testCounterCommandListener

[3] [https://www.sofastack.tech/en/projects/sofa-jraft/benchmark-performance/]

  was:
We should test the performance of a JRAFT port in comparison with original 
implementation [1].

Where are two possible scenarios:

1. Testing single group performance using simple state machine and single 
thread.

[2] can be used as a reference for a test scenario

2. Testing key-value API performance in scenarios like [3]

For optimal performance per-partition batches must be used together with putAll 
operation.

[1] [https://github.com/sofastack/sofa-jraft]

[2] 
org.apache.ignite.raft.server.ITJRaftCounterServerTest#testCounterCommandListener

[3] [https://www.sofastack.tech/en/projects/sofa-jraft/benchmark-performance/]


> Implement RAFT benchmarks
> -------------------------
>
>                 Key: IGNITE-15448
>                 URL: https://issues.apache.org/jira/browse/IGNITE-15448
>             Project: Ignite
>          Issue Type: Task
>            Reporter: Alexey Scherbakov
>            Priority: Major
>              Labels: ignite-3
>
> We should test the performance of a JRAFT port in comparison with original 
> implementation [1].
> Where are two possible scenarios:
> 1. Testing single group performance using simple state machine and various 
> client threads count.
> [2] can be used as a starting point
> 2. Testing key-value API performance in scenarios like [3]
> For optimal performance per-partition batches must be used together with 
> putAll operation.
> [1] [https://github.com/sofastack/sofa-jraft]
> [2] 
> org.apache.ignite.raft.server.ITJRaftCounterServerTest#testCounterCommandListener
> [3] [https://www.sofastack.tech/en/projects/sofa-jraft/benchmark-performance/]



--
This message was sent by Atlassian Jira
(v8.3.4#803005)

Reply via email to