TiDB Sysbench Performance Test Report -- v5.2.0 vs. v5.1.1

Test overview

This test aims at comparing the Sysbench performance of TiDB v5.2.0 and TiDB v5.1.1 in the Online Transactional Processing (OLTP) scenario. The results show that compared with v5.1.1, the Point Select performance of v5.2.0 is improved by 11.03%, and the performance of other scenarios is slightly reduced.

Test environment (AWS EC2）

Hardware configuration

Software version

Parameter configuration

TiDB v5.2.0 and TiDB v5.1.1 use the same configuration.

TiDB parameter configuration

log.level: "error"
performance.max-procs: 20
prepared-plan-cache.enabled: true
tikv-client.max-batch-wait-time: 2000000

TiKV parameter configuration

storage.scheduler-worker-pool-size: 5
raftstore.store-pool-size: 3
raftstore.apply-pool-size: 3
rocksdb.max-background-jobs: 8
raftdb.max-background-jobs: 4
raftdb.allow-concurrent-memtable-write: true
server.grpc-concurrency: 6
readpool.unified.min-thread-count: 5
readpool.unified.max-thread-count: 20
readpool.storage.normal-concurrency: 10
pessimistic-txn.pipelined: true
server.enable-request-batch: false

TiDB global variable configuration

set global tidb_hashagg_final_concurrency=1;
set global tidb_hashagg_partial_concurrency=1;
set global tidb_enable_async_commit = 1;
set global tidb_enable_1pc = 1;
set global tidb_guarantee_linearizability = 0;
set global tidb_enable_clustered_index = 1;

Test plan

1. Deploy TiDB v5.2.0 and v5.1.1 using TiUP.
2. Use Sysbench to import 16 tables, each table with 10 million rows of data.
3. Execute the analyze table statement on each table.
4. Back up the data used for restore before different concurrency tests, which ensures data consistency for each test.
5. Start the Sysbench client to perform the point_select, read_write, update_index, and update_non_index tests. Perform stress tests on TiDB via HAProxy. The test takes 5 minutes.
6. After each type of test is completed, stop the cluster, overwrite the cluster with the backup data in step 4, and restart the cluster.

Prepare test data

Execute the following command to prepare the test data:

sysbench oltp_common \
    --threads=16 \
    --rand-type=uniform \
    --db-driver=mysql \
    --mysql-db=sbtest \
    --mysql-host=$aws_nlb_host \
    --mysql-port=$aws_nlb_port \
    --mysql-user=root \
    --mysql-password=password \
    prepare --tables=16 --table-size=10000000

Perform the test

Execute the following command to perform the test:

sysbench $testname \
    --threads=$threads \
    --time=300 \
    --report-interval=1 \
    --rand-type=uniform \
    --db-driver=mysql \
    --mysql-db=sbtest \
    --mysql-host=$aws_nlb_host \
    --mysql-port=$aws_nlb_port \
    run --tables=16 --table-size=10000000

Test results

Point Select performance

Compared with v5.1.1, the Point Select performance of v5.2.0 is improved by 11.03%.

Update Non-index performance

Compared with v5.1.1, the Update Non-index performance of v5.2.0 is reduced by 1.98%.

Update Index performance

Compared with v5.0.2, the Update Index performance of v5.1.0 is reduced by 4.33%.

Read Write performance

Compared with v5.0.2, the Read Write performance of v5.1.0 is reduced by 1.24%.