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.

120K distributed consistent writes per second with Calvin

85 pointsby zenithmover 8 years ago

9 comments

itpabout 8 years ago
This seems cool, and I sincerely wish them nothing but success. That said, I had a major sense of déjà vu while reading this post -- I worked at FoundationDB prior to the Apple acquisition, when we published a blog post with a very similar feel:<p><a href="http:&#x2F;&#x2F;web.archive.org&#x2F;web&#x2F;20150325003241&#x2F;http:&#x2F;&#x2F;blog.foundationdb.com&#x2F;databases-at-14.4mhz" rel="nofollow">http:&#x2F;&#x2F;web.archive.org&#x2F;web&#x2F;20150325003241&#x2F;http:&#x2F;&#x2F;blog.founda...</a><p>I&#x27;m not trying to make a comparison between a system I used to work on and one that I frankly know little to nothing about; rather, I&#x27;d suggest that building a system like this just isn&#x27;t enough to be compelling on its own.
评论 #13728218 未加载
评论 #13728483 未加载
评论 #13728295 未加载
g0del_was_wr0ngabout 8 years ago
Including your 9x write amplification in the number of &quot;consistent writes&quot; doesn&#x27;t count -- like at all. I&#x27;m amazed nobody called you out on this yet.<p>You&#x27;re doing 3k batches per second with 4 logical writes each, right? So that is at most 3-12k writes per second using the way that every other distributed database benchmark and paper counts.<p>Or otherwise - if you continue counting writes in this special&#x2F;misleading way - you&#x27;d have to multiply every other distributed db benchmark&#x27;s performance numbers with a factor of 3-15x to get an apples-to-apples comparison.<p>The 12k batched writes&#x2F;sek through what I assume is a paxos variant is still pretty impressive though! Good to get more competition&#x2F;alternatives for zookeeper &amp; friends!
评论 #13731968 未加载
评论 #13739672 未加载
评论 #13731191 未加载
web007about 8 years ago
This description is very misleading.<p>120,000 writes per second is accurate, talking about actual durable storage (disk) writes. But it&#x27;s only 3,330 transactions, which should be the number that a user cares about.<p>I don&#x27;t have proper data and I&#x27;m a bit rusty, but I feel like Cassandra could blow that away if you set similar consistency requirements on the client side (QUORUM on read, same for write?). Am I understanding this correctly, or does Fauna&#x2F;Calvin give you something functionally better than what C* can do?
评论 #13729284 未加载
qaqabout 8 years ago
Maybe I am missing some special point but a decent PG box will do 1,000,000+ TPS vs 3,000+ TPS here. When pgXact lands it will do close to 2,000,000 TPS. So reading all the posts about the amazing new db &quot;X&quot; that can do about N times less than PG on a multi-node cluster I get confused why the numbers are being presented as some sort of achievement.
评论 #13730153 未加载
评论 #13730920 未加载
评论 #13730637 未加载
zenithmover 8 years ago
This is a new one to me...the referenced paper is here: <a href="http:&#x2F;&#x2F;cs.yale.edu&#x2F;homes&#x2F;thomson&#x2F;publications&#x2F;calvin-sigmod12.pdf" rel="nofollow">http:&#x2F;&#x2F;cs.yale.edu&#x2F;homes&#x2F;thomson&#x2F;publications&#x2F;calvin-sigmod1...</a><p>How does this algorithm compare to whatever Google Spanner does?
评论 #13727797 未加载
imownbeyabout 8 years ago
&quot;Calvin&#x27;s primary trade-off is that it doesn&#x27;t support session transactions, so it&#x27;s not well suited for SQL. Instead, transactions must be submitted atomically. Session transactions in SQL were designed for analytics, specifically human beings sitting at a workstation. They are pure overhead in a high-throughput operational context.&quot;<p>Is this specifically for distributed SQL only? I think there are some scalable SQL systems that don&#x27;t support sessions either.
评论 #13728020 未加载
评论 #13728046 未加载
olegkikinabout 8 years ago
2011: Benchmarking Cassandra Scalability on AWS - Over a million writes per second<p><a href="http:&#x2F;&#x2F;techblog.netflix.com&#x2F;2011&#x2F;11&#x2F;benchmarking-cassandra-scalability-on.html" rel="nofollow">http:&#x2F;&#x2F;techblog.netflix.com&#x2F;2011&#x2F;11&#x2F;benchmarking-cassandra-s...</a><p>Also a single SSD from 2015 is rated at 120K writes per second:<p>PM1725: <a href="http:&#x2F;&#x2F;www.samsung.com&#x2F;semiconductor&#x2F;global&#x2F;file&#x2F;insight&#x2F;2015&#x2F;11&#x2F;pm1725-ProdOverview-2015-0.pdf" rel="nofollow">http:&#x2F;&#x2F;www.samsung.com&#x2F;semiconductor&#x2F;global&#x2F;file&#x2F;insight&#x2F;201...</a>
lngnmnabout 8 years ago
Consistent writes to a permanent storage or didn&#x27;t happen.
评论 #13728500 未加载
rystsovabout 8 years ago
Is it possible to download fauna to play with it on my own?
评论 #13729016 未加载