Benchmark script for base64 ingest

[JoshMock]

Uses work from elastic/elastic-transport-js#337. Part of #3080.

Should be run with env vars ES_URL, ES_USERNAME, and ES_PASSWORD. Example:

npm install
env ES_URL=http://localhost:9200 \
  ES_USERNAME=elastic \
  ES_PASSWORD=changeme \
  node test/ingest/base64.mjs

[prodsecmachine]

✅ Snyk checks have passed. No issues have been found so far.

Status	Scanner	Critical	High	Medium	Low	Total (0)
✅	Open Source Security	0	0	0	0	0 issues
✅	Licenses	0	0	0	0	0 issues

💻 Catch issues earlier using the plugins for VS Code, JetBrains IDEs, Visual Studio, and Eclipse.

[margaretjgu]

parsing of dataset should be moved before the start

corpus deserialization and file IO should not be part of the benchmark result

[JoshMock]

I guess this is a question of whether splitting a string into an array of strings is considered deserialization or serialization. 😆

I download, unzip and read the file into a string before starting to measure the time, but this feels like a gray area that we should clear up. Opened up a question about it in the Google doc.

[margaretjgu]

yea also l120 for JSON.parse() would count as deserialization.

[margaretjgu]

should add a refresh call to ensure the index is fully created before starting the ingestion

[JoshMock]

Added in 63277f6

[margaretjgu]

The index is deleted but there's no check/delete at the start of each pass before client.indices.create(). If a previous run crashed, the index might already exist...to be extra safe

[JoshMock]

Added in 47044ff.

[margaretjgu]

think we need documents/second metric instead as int he doc

[JoshMock]

Asked a clarifying question about that in the doc.

[margaretjgu]

lgtm just when we're testing to use http instead of https as suggested in the doc

[JoshMock]

Will wait to get another review once questions get answered in the benchmark methodology document.