Skip to content

Commit ecdc554

Browse files
committed
Strengthen passthrough proxy section with L7 and contention insights
- Reframe the takeaway: the proxy boils the latency-sensitive path to near-TCP-stack overhead while operating at Layer 7 — that's the win - Add paragraph explaining why overhead holds across 10/100 topics: the proxy doesn't contend between topics (unlike a broker which juggles disk I/O, partition leaders, and replication); the connection sweep validates linear throughput scaling Assisted-by: Claude Sonnet 4.6 <noreply@anthropic.com> Signed-off-by: Sam Barker <sam@quadrocket.co.uk>
1 parent 1e3de97 commit ecdc554

1 file changed

Lines changed: 5 additions & 2 deletions

File tree

_posts/2026-05-21-benchmarking-the-proxy.md

Lines changed: 5 additions & 2 deletions
Original file line numberDiff line numberDiff line change
@@ -67,8 +67,11 @@ Good news first. The proxy itself — with no filter chain, just routing traffic
6767

6868
**The headline: ~0.2 ms additional average publish latency. Throughput is unaffected.**
6969

70-
What did I take away from this entirely unsurprising result? Not much, honestly — without filters the proxy is little more than a couple of hops through the TCP stack, but we now have data rather than a hunch.
71-
The end-to-end (E2E) p99 figure is dominated by the Kafka consumer fetch timeouts, as it should be. That said, it is reassuring to have a sub-ms impact on the p99.
70+
What did I take away from this entirely unsurprising result? Not much, honestly — without filters the proxy boils the latency-sensitive path down to little more than a couple of hops through the TCP stack. We replaced a hunch with data. The remarkable part: the proxy is doing this at Layer 7.
71+
72+
The overhead holding across 10 and 100 topics makes sense for the same reason: the proxy doesn't contend between topics. A Kafka broker juggles disk I/O, partition leaders, and replication across everything it manages; the proxy treats each connection independently. Topics don't contend for shared resources: throughput scales linearly across them, and the connection sweep validates it.
73+
74+
The end-to-end p99 figure is dominated by Kafka consumer fetch timeouts, as it should be. That said, it is reassuring to have a sub-ms impact on the p99.
7275

7376
---
7477

0 commit comments

Comments
 (0)