KAKafka · English
Community homeHot discussions Latest posts Best of Knowledge base Resources About

Kafka · English · Interview notes: how I answered a question about the kafka-en concurrency model

44
KAr/kafka-en·posted by swoole_lee·6 hours agoTooling

Interview notes: how I answered a question about the kafka-en concurrency model

It took me two weeks of on-and-off digging and plenty of wrong turns. Writing the process down as it happened so the next person spends less time.

One last trap: in container environments remember to adjust the memory-related parameters in step. Otherwise the host limit and the process expectation disagree, and the symptom is intermittent, unreproducible failure.

Performance44%
Maintainability28%
Ecosystem and community17%
Hiring difficulty11%

75 votes total

26 comments

26 comments

M
Kkernel_panic·12 minutes ago

There is actually a simpler fix that needs no architecture change: move this check up to the gateway and the problem disappears. The cost is one extra lookup at the gateway.

454
Hhuang_ke·3 minutes ago

I see point 3 differently. The trade-off depends on your read/write ratio: read-heavy with little writing means caching actually widens the inconsistency window.

228
Hhuang_ke·2 days ago

I just read the kafka-en source — the author actually explains the reasoning in a comment, roughly "so that it degrades into predictable behaviour in extreme cases".

91
Nnikic·2 days ago

One counter-example: below kafka-en 7.4 the semantics of that code are different, so do not copy it verbatim. We got burned in staging and rolled back once.

402
Wwinter·yesterdayedited

This matches what we see in production. We only hit it past 3k QPS; the earlier load tests showed nothing — the test traffic was too clean, with no long-tail requests.

286
Rrase·1 hour ago

We have run this in production for two years without hitting it. That said, we never reached this scale, so our experience is not really evidence here.

130
Hhuang_ke·28 minutes agoedited

Thanks for sharing real numbers — far more useful than the articles that only cover concepts.

126
Aalice_dev·1 hour ago

Agreeing with the above. One addition: with this option enabled the GC count in your metrics doubles, so adjust the alert threshold at the same time or it will keep firing.

57
Zzhou_yi·just now

Sharing our numbers, 8 cores 16GB, same scenario:

| Concurrency | P50 | P99 |
|---|---|---|
| 200 | 12ms | 88ms |
| 500 | 31ms | 340ms |

P99 clearly collapses at 500 concurrency, which lines up with your knee point.

361
Llinlin·just now

I just read the kafka-en source — the author actually explains the reasoning in a comment, roughly "so that it degrades into predictable behaviour in extreme cases".

50
Ddev_zhou·3 minutes ago

Can you give a minimal reproduction? I ran it locally for ten minutes and could not reproduce on macOS with the latest version.

36
Sslow_queryOP·1 hour ago

Saved. I am reworking this area this week — this saves a lot of wrong turns.

67
Wwinter·28 minutes ago

Worth learning from this debugging approach. We went straight at the logs and took a much longer route.

458
Kkernel_panic·2 days ago

Has anyone run a controlled experiment? I did, reducing it to a single variable, and the difference was 4% — within noise. So I suspect the main cause is something else.

255
NnikicOP·2 days ago

There is actually a simpler fix that needs no architecture change: move this check up to the gateway and the problem disappears. The cost is one extra lookup at the gateway.

6
Kkernel_panic·2 days ago

We have run this in production for two years without hitting it. That said, we never reached this scale, so our experience is not really evidence here.

2
Llinlin·just now

This is not a kafka-en problem, it is a usage problem. The docs say this API is not thread-safe and you must lock around it yourself.

7
Mmike_xuOP·1 hour ago

A question: what changes in a container with a 512Mi memory limit? That is how we run it in production.

242
Oops_wang·2 days ago

I see point 3 differently. The trade-off depends on your read/write ratio: read-heavy with little writing means caching actually widens the inconsistency window.

133
Sswoole_lee·1 hour ago

Has anyone run a controlled experiment? I did, reducing it to a single variable, and the difference was 4% — within noise. So I suspect the main cause is something else.

285
Nnikic·2 days agoedited

This matches what we see in production. We only hit it past 3k QPS; the earlier load tests showed nothing — the test traffic was too clean, with no long-tail requests.

19
Rrase·2 days ago

Sharing our numbers, 8 cores 16GB, same scenario:

| Concurrency | P50 | P99 |
|---|---|---|
| 200 | 12ms | 88ms |
| 500 | 31ms | 340ms |

P99 clearly collapses at 500 concurrency, which lines up with your knee point.

339
Lli_ming·2 days ago

A question: what changes in a container with a 512Mi memory limit? That is how we run it in production.

2
Mmike_xu·2 days ago

Saved. I am reworking this area this week — this saves a lot of wrong turns.

1
Kkite·just now

Thanks for sharing real numbers — far more useful than the articles that only cover concepts.

19
Zzhou_yi·yesterdayedited

Agreeing with the above. One addition: with this option enabled the GC count in your metrics doubles, so adjust the alert threshold at the same time or it will keep firing.

2

This is the post detail page /en/c/kafka-en/post/p9. Posts and comments are generated deterministically from a seeded PRNG, so the same post always renders the same content and the link can be shared, reloaded and indexed. In production this page reads MySQL for the post, Redis for hot-post caching, and fetches the whole comment tree in a single query on the path column.

See the database schema →