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.

Using SQLite for great science, part 1 (or 3, I guess). | Stavros' Stuff

24 pointsby iamelgringoalmost 15 years ago

4 comments

gxtialmost 15 years ago
SQLite is so hardcore about being totally unbreakable that they have a section in their documentation dedicated to explaining all the horrible things the OS has to do just to get it to fail: <a href="http://www.sqlite.org/lockingv3.html#how_to_corrupt" rel="nofollow">http://www.sqlite.org/lockingv3.html#how_to_corrupt</a><p>Reading about their testing process is also informative: <a href="http://www.sqlite.org/testing.html" rel="nofollow">http://www.sqlite.org/testing.html</a>
sshconnectionalmost 15 years ago
1000 writes in 5 seconds for mongo? Running 1.4, I get ~2500/sec on an old P4 desktop that I was using to test on. I've only recently started working with mongo, and I have no experience working with 1.3, the unstable development version that he lost his data using. I'm having a hard time buying into his comparison.
zandorgalmost 15 years ago
I used SQLite until I found a query that locked it up seemingly forever. It had 2 inner joins. It worked with 1 inner join - NOT 2. The author of SQLite said "Not my problem buddy, it's yours". So I'm switching to PostGRE (which CAN handle those 2 inner joins) or a Microsoft SQL.
评论 #1433257 未加载
评论 #1433234 未加载
lsbalmost 15 years ago
3M rows, 70M relationships? Sounds like it could fit in a hash table in memory in your programming language of choice.
评论 #1432935 未加载
评论 #1432934 未加载