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.

Sign up for GitHub

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

Jump to bottom

Add instruction for running benchmark test locally #3825

Open
wants to merge 1 commit into
base: main
Choose a base branch
from
Open

Add instruction for running benchmark test locally #3825

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
57 changes: 57 additions & 0 deletions src/test_workflow/README.md
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -217,6 +217,63 @@ H 9 * * * %BUNDLE_MANIFEST_URL=https://ci.opensearch.org/ci/dbc/distribution-bui

Once you have added the configuration in the jenkinsfile please raise the PR and opensearch-infra team will review it.

#### Run the test locally
1. Comment out the tailing characters `+ role_params` in [this line](https://github.com/opensearch-project/opensearch-build/blob/2.9.0/src/test_workflow/benchmark_test/benchmark_test_cluster.py#L70) to remove the parameters used for CI workflow.

Add back `+ " --require-approval never"` into this line to bypass the prompt "Do you wish to deploy these changes (y/n)" when deploying CDK application.

2. Prepare `config.yml` file.
The file contains some parameters for the CDK application of OpenSearch cluster.
- Please see below sample, and create a file called `config.yml` with all the fields in the sample.
- Do not change the value of the following fields: "Repository", "Role" and "isInternal".
- Modify the "VpcId" to make it correspond with the VPC created in the specific region of the AWS account.
- About "serverAccessType", see the [document for opensearch-cluster-cdk](https://github.com/opensearch-project/opensearch-cluster-cdk?tab=readme-ov-file#restricting-server-access) for reference.

A sample content for `config.yml` file:
```
Description: Configuration file to store constants required to run performance test
Constants:
Repository: https://github.com/opensearch-project/opensearch-cluster-cdk
VpcId: <VPC ID>
AccountId: <AWS account ID>
serverAccessType: ipv4 / ipv6 / prefixList / securityGroupId
restrictServerAccessTo:
Region: <AWS Region>
Role: cfn-set-up
isInternal: false
```

3. Prepare build manifest file.
The official opensearch bundle build manifest file is going to be used.
- Download the build manifest file directly from the following URL:
- for bundle distribution: https://ci.opensearch.org/ci/dbc/distribution-build-opensearch/2.10.0/latest/linux/x64/tar/dist/opensearch/manifest.yml
- for minimal distribution: https://ci.opensearch.org/ci/dbc/distribution-build-opensearch/2.10.0/latest/linux/x64/tar/builds/opensearch/manifest.yml
- Change the version number or the build number in the URL as per demand.
- See the document [here](https://github.com/opensearch-project/opensearch-build/tree/2.9.0?tab=readme-ov-file#build-numbers) to learn more about the file URL and build number.

4. Prepare `benchmark.ini` file.

This step aims to declare the OpenSearch cluster that used for storing the benchmark result and the metrics data.
Because the additional statistic data, such as JVM and CPU data collected from node stats API, can only be stored in OpenSearch cluster as index instead of text file, using OpenSearch cluster as the data store is necessary.
- Create a file `benchmark.ini`, and copy the content from the [default file](https://github.com/opensearch-project/opensearch-benchmark/blob/1.1.0/osbenchmark/resources/benchmark.ini).
- Remove the value after `=` for the fields `root.dir`, `src.root.dir` ,and `local.dataset.cache`
- See [opensearch-benchmark docs](https://opensearch.org/docs/2.9/benchmark/configuring-benchmark/#examples) to configure `[results_publishing]` section.

5. Go to the root directory of local cloned `opensearch-build` repository, and run the command to start the benchmark test.
- See the [source code](https://github.com/opensearch-project/opensearch-build/blob/main/src/test_workflow/benchmark_test/benchmark_args.py) for all supported parameters in the script.
For the meaning of parameters used for OpenSearch cluster, please refer the [docs](https://github.com/opensearch-project/opensearch-cluster-cdk?tab=readme-ov-file#required-context-parameters) in `opensearch-cluster-cdk` repository.
- Recommend to run the command in a `tmux` session to keep the process running in the background.

Sample command:
```
./test.sh benchmark-test --bundle-manifest ~/manifest-2.10.0.yml --config ~/config.yml \
--suffix 2100-cluster-multinode --without-security \
--additional-config "opensearch.experimental.feature.segment_replication_experimental.enabled:true cluster.indices.replication.strategy:SEGMENT" \
--workload http_logs --workload-params '{"index_settings":{"number_of_shards":1,"number_of_replicas":1}}' \
--benchmark-config ~/benchmark.ini --data-node-count 2 --data-node-storage 100 \
--user-tag="segrep:enabled,remote-store:enabled" --capture-node-stat --use-50-percent-heap --min-distribution
```

## Testing in CI/CD

The CI/CD infrastructure is divided into two main workflows - `build` and `test`. The `build` workflow automates the process to generate all OpenSearch and OpenSearch Dashboards artifacts, and provide them as distributions to the `test` workflow, which runs exhaustive testing on the artifacts based on the artifact type. The next section talks in detail about the test workflow.
Expand Down
Loading