TE
TechEcho
Home24h TopNewestBestAskShowJobs
GitHubTwitter
Home

TechEcho

A tech news platform built with Next.js, providing global tech news and discussions.

GitHubTwitter

Home

HomeNewestBestAskShowJobs

Resources

HackerNews APIOriginal HackerNewsNext.js

© 2025 TechEcho. All rights reserved.

Valkey achieved one million RPS 6 months after forking from Redis

113 pointsby reconditerose8 months ago

7 comments

nonane8 months ago
One thing the article doesn’t mention is how they figured out that waiting for external memory access is the bottleneck. Are there any profiling tools available that would tell the developer that the cpu is waiting for external memory x% of the time?
评论 #41539759 未加载
评论 #41540878 未加载
评论 #41539658 未加载
评论 #41539665 未加载
评论 #41557772 未加载
MobiusHorizons8 months ago
This is really cool work! I am surprised to see this level of tuning without using cache profiling or other performance counters to identify the bottleneck and quantify the improvement.
评论 #41539734 未加载
secondcoming8 months ago
Redis&#x27; biggest flaw is its single threaded design. We end up having to run separate redis processes on each core and have client side sharding. We&#x27;re lucky that our data allows this.<p>We experiment with KeyDB too but I&#x27;m not sure what state that project is in.
评论 #41540844 未加载
PeterZaitsev8 months ago
Great to see Valkey team is making a progress well beyond keeping old Redis version Security Patched.
jacobgorm8 months ago
Who in their right mind uses linked lists for a database style workload? Try doing this with arrays to get a reasonable baseline.
评论 #41540890 未加载
throwaway815238 months ago
Nice, how does that compare with Pedis, which was written in C++ several years ago? It&#x27;s an incomplete Redis lookalike that isn&#x27;t getting current development, but it uses Seastar, the same parallelism framework as ScyllaDB.<p><a href="https:&#x2F;&#x2F;github.com&#x2F;fastio&#x2F;1store">https:&#x2F;&#x2F;github.com&#x2F;fastio&#x2F;1store</a>
评论 #41540017 未加载
评论 #41539946 未加载
评论 #41539735 未加载
评论 #41539585 未加载
评论 #41539599 未加载
tayo428 months ago
The interleave thing isnt intuitive to me.<p>The problem with linked lists is the memory address of nodes isn&#x27;t necessarily contiguous because of malloc and the value could be NULL? Why does interleave loop make it faster for the cpu? It still a linked list, arbitrary memory, could be NULL? Not sure what im missing here?
评论 #41540371 未加载