PostHog / PostHog/posthog

Partition S3 batch export files by event name

Open
#26,316 1 comment 1 reaction 0 assignees View on GitHub

Nobody has claimed this yet.

team/batch-exports
Dominant language
Python
Stars
39.9k
Forks
3.4k
Avg merge
7h 27m
Merged PRs (30d)
222

Description

As a user of S3 batch exports, it can be easier to process individual files per event, but only partitioning by timestamp and table fields is currently supported.

The main challenge with this feature is that we do not know which event names to partition by before we query ClickHouse. So, we may need to delay the creation of an S3 upload until we start seeing events, and then maintain one S3 upload per event name. The question pending with this is how to allow recovery in the event of a worker crash now that we have a potentially very large number of simultaneous S3 uploads. Maybe Temporal heartbeating is not enough to support this feature and we will need to look into new ways of tracking progress.

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start by tracing the existing S3 batch export flow, including timestamp and table partitioning, the ClickHouse query, and Temporal worker recovery. Determine how event names can drive uploads and how progress can be recovered after a worker crash. Done means exports are partitioned by event name without losing recoverability.

Written by the indexing model from the issue text.

Assessment

Tech stack
aws, clickhouse, python
Domain
backend, cloud, data-engineering
Issue type
Feature
Difficulty
5/5
Estimated time
Over a week
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.