[
https://issues.apache.org/jira/browse/IGNITE-20618?page=com.atlassian.jira.plugin.system.issuetabpanels:all-tabpanel
]
Konstantin Orlov updated IGNITE-20618:
--------------------------------------
Attachment: main-fix.png
> Sql. Degradation of SELECT operations performance over time
> -----------------------------------------------------------
>
> Key: IGNITE-20618
> URL: https://issues.apache.org/jira/browse/IGNITE-20618
> Project: Ignite
> Issue Type: Bug
> Components: sql
> Reporter: Ivan Artiukhov
> Assignee: Konstantin Orlov
> Priority: Major
> Labels: ignite-3, performance
> Attachments: Screenshot 2023-10-12 at 16.34.16.png, Screenshot
> 2023-10-12 at 16.34.58.png, Screenshot 2023-10-12 at 17.20.05.png, Screenshot
> 2023-10-12 at 17.20.20.png, baseline.png,
> gc-poc-tester-SERVER-192.168.1.43-id-0-2023-10-09-16-17-07.log,
> ignite-config.json, main-fix.png,
> poc-tester-SERVER-192.168.1.43-id-0-2023-10-09-16-17-07.log,
> sql-degr-insert.png, sql-degr-select.png
>
> Time Spent: 10m
> Remaining Estimate: 0h
>
> Ignite 3, rev. 7d188ac7ae068bd69ff0e6e6cfe5a32ac5749d17
> Benchmark:
> [https://github.com/gridgain/YCSB/blob/ycsb-2023.3/ignite3/src/main/java/site/ycsb/db/ignite3/IgniteSqlClient.java]
>
> The benchmark establishes an SQL {{Session}} and perform {{SELECTs}} via
> {{{}preparedStatement{}}}.
> {*}Steps{*}:
> * Start a 1 node cluster with the attached [^ignite-config.json]
> ** {*}raft.fsync = false{*}{*}{{*}}
> * Start the benchmark in pre-load mode – preload 100k entries:
> ** {{-db site.ycsb.db.ignite3.IgniteSqlClient -load -P
> /opt/pubagent/poc/config/ycsb/workloads/workloadc -threads 1 -p
> recordcount=10000 -p dataintegrity=true -p measurementtype=timeseries -p
> hosts=192.168.1.43 -p recordcount=100000 -p operationcount=100000}}
> * Start the benchmark in 100% read mode:
> ** {{-db site.ycsb.db.ignite3.IgniteSqlClient -load -P
> /opt/pubagent/poc/config/ycsb/workloads/workloadc -threads 1 -p
> recordcount=10000 -p dataintegrity=true -p measurementtype=timeseries -p
> hosts=192.168.1.43 -p recordcount=100000 -p operationcount=100000}}
> The following *stable* throughput is observed in preload mode
> ({{{}INSERT{}}}):
> !sql-degr-insert.png!
> The following *unstable* throughput is observed on reads ({{{}SELECT):{}}}
> {{!sql-degr-select.png!}}
--
This message was sent by Atlassian Jira
(v8.20.10#820010)
