jrmccluskey commented on code in PR #33702:
URL: https://github.com/apache/beam/pull/33702#discussion_r1941387833


##########
sdks/python/apache_beam/testing/benchmarks/README.md:
##########
@@ -0,0 +1,117 @@
+<!--
+    Licensed to the Apache Software Foundation (ASF) under one
+    or more contributor license agreements.  See the NOTICE file
+    distributed with this work for additional information
+    regarding copyright ownership.  The ASF licenses this file
+    to you under the Apache License, Version 2.0 (the
+    "License"); you may not use this file except in compliance
+    with the License.  You may obtain a copy of the License at
+
+      http://www.apache.org/licenses/LICENSE-2.0
+
+    Unless required by applicable law or agreed to in writing,
+    software distributed under the License is distributed on an
+    "AS IS" BASIS, WITHOUT WARRANTIES OR CONDITIONS OF ANY
+    KIND, either express or implied.  See the License for the
+    specific language governing permissions and limitations
+    under the License.
+-->
+
+# Writing a Dataflow Cost Benchmark
+
+Writing a Dataflow Cost Benchmark to estimate the financial cost of executing 
a pipeline on Google Cloud Platform Dataflow requires 4 components in the 
repository:
+
+1. A pipeline to execute (ideally one located in the examples directory)
+1. A text file with pipeline options in the 
`.github/workflows/cost-benchmarks-pipeline-options` 
[directory](../../../../../.github/workflows/cost-benchmarks-pipeline-options)
+1. A test class inheriting from the `DataflowCostBenchmark` 
[class](../load_tests/dataflow_cost_benchmark.py)
+1. An entry to execute the pipeline as part of the cost benchmarks workflow 
action
+
+### Choosing a Pipeline
+Pipelines that are worth benchmarking in terms of performance and cost have a 
few straightforward requirements.
+
+1. The transforms used in the pipeline should be native to Beam *or* be 
lightweight and readily available in the given pipeline
+1. The pipeline itself should run on a consistent data set and have consistent 
internals (such as model versions for `RunInference` workloads.)

Review Comment:
   This is referring to keeping the same version of a model in a RunInference 
pipeline rather than doing something like automatically updating to the latest 
version. A fully specified benchmark should be running on an identical 
configuration every time, from details like model version and framework all the 
way up to the GCP region the job runs in. I'll see if I can nail down better 
wording



-- 
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.

To unsubscribe, e-mail: [email protected]

For queries about this service, please contact Infrastructure at:
[email protected]

Reply via email to