Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

TiDB v5.1.0 consumes more CPU than v5.0.2 in ycsb workload-e #25726

Open
youjiali1995 opened this issue Jun 24, 2021 · 2 comments
Open

TiDB v5.1.0 consumes more CPU than v5.0.2 in ycsb workload-e #25726

youjiali1995 opened this issue Jun 24, 2021 · 2 comments
Labels
severity/minor sig/execution SIG execution type/bug The issue is confirmed as a bug.

Comments

@youjiali1995
Copy link
Contributor

Bug Report

Please answer these questions before submitting your issue. Thanks!

1. Minimal reproduce step (Required)

Topo:

          spec:
            machines:
              - name: m1
                exclusive: true
            items:
              - name: tidb
                spec:
                  memory: 32GB
                  cores: 16
                  machine: m1
              - name: pd
                spec:
                  memory: 8GB
                  cores: 2
                  machine: m1
              - name: tikv-1
                spec:
                  memory: 64GB
                  cores: 16
                  disks:
                    disk1:
                      kind: nvme
                      mountPath: /disk1
                  machine: m1
              - name: workload
                spec:
                  memory: 8GB
                  cores: 4
                  machine: m1

ycsb workload-e:

                      echo "recordcount=100000000" >> global.conf
                      echo "operationcount=10000000" >> global.conf
                      echo "workload=core" >> global.conf
                      echo "fieldcount=10" >> global.conf
                      echo "fieldlength=100" >> global.conf
                      echo "threadcount=500" >> global.conf
                      echo "requestdistribution=zipfian" >> global.conf
                      go-ycsb run mysql \
                          -P /ycsb/workloads/{{workflow.parameters.workload}} \
                          -P global.conf \
                          -p mysql.host=$tidb \
                          -p mysql.port=4000

2. What did you expect to see? (Required)

Same throughput and CPU usage.

3. What did you see instead (Required)

The throughput is the same(?) in the two versions but CPU usage is different.

v5.1
image

v5.0.2
image

4. What is your TiDB version? (Required)

v5.1.0 go1.16.4
v5.0.2 go1.13

@youjiali1995 youjiali1995 added the type/bug The issue is confirmed as a bug. label Jun 24, 2021
@youjiali1995
Copy link
Contributor Author

/cc @sticnarf

@youjiali1995
Copy link
Contributor Author

BTW, the ycsb workload-e is a TiDB CPU-bound workload. I guess there are many potiential optimizations here.
image

Related to #25573 (comment).

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
severity/minor sig/execution SIG execution type/bug The issue is confirmed as a bug.
Projects
None yet
Development

No branches or pull requests

2 participants