Function Hub Scalability Test Datasets
收藏资源简介:
<strong>Function Hub - characterization</strong> Running Benchmarks Use the scripts <code>run_sequential_benchmark.sh</code> and <code>run_load_test_benchmark.sh</code> to orchestrate running the benchmark, this will create a docker container, that will first deploy a cloudstash instance, then run the benchmark, output the results to .csv saved in the <code>output</code> directory, and finally remove the created cloudstash instance. <pre><code>>>> THE SCRIPTS EXPECTS TO BE CALLED FROM THE ROOT OF THE REPOSITORY <<< </code></pre> You may add the following <code>option</code> flags after the required arguments: <code>--local</code> runt the script locally, will bindmount the benchmark code from the local directory to make development easier <code>--tail</code> will automatically follow the generated log file, note that when invoking the option, when you exit tail with Ctrl-c, you will the container process too, so do not use in production! <code>--debug</code> will enable some debug prints <code>--verbose</code> will enable more verbose prints Sequential Uploads Benchmark This benchmark will sequentially upload the specified number of artifacts. You must pass the <code>number of artifacts to be sequentially uploaded</code> as the first argument: <pre><code>bash run_sequential_benchmarks <number of artifacts (int)> [options] </code></pre> Load Test Benchmark This benchmark will run a predefined load test, consisting of listing artifacts, getting (downloading) artifacts and uploading artifacts. You must pass the <code>number of artifacts to already be in the cloudstash instance</code> before the load test is started, these will be uploaded to the deployed cloudstash instance before the load test starts. <pre><code>bash run_load_test_benchmark.sh <number of artifacts (int)> [options]</code></pre>



