TE
科技回声
首页24小时热榜最新最佳问答展示工作
GitHubTwitter
首页

科技回声

基于 Next.js 构建的科技新闻平台,提供全球科技新闻和讨论内容。

GitHubTwitter

首页

首页最新最佳问答展示工作

资源链接

HackerNews API原版 HackerNewsNext.js

© 2025 科技回声. 版权所有。

The future of databases is RAM clouds and SSD

38 点作者 petewarden超过 15 年前

6 条评论

Andys超过 15 年前
The problem for RethinkDB is that the firmware on Intel SSDs is advanced enough that it can speed up the normal, free MySQL engine (or postgresql, or oracle, ...) to almost the same sort of speeds.<p>I am yet to see any evidence that RethinkDB's approach is anything but a niche for those who need to squeeze an extra 5-10% that advances in filesystems and flash firmware cannot.
评论 #983040 未加载
评论 #982958 未加载
评论 #983130 未加载
fauigerzigerk超过 15 年前
That's interesting and very promising. But what I miss in most discussions of relational vs NoSQL is some kind of separation of data model related, performance related and scalability related arguments. These are three rather different issues.<p>There is no way to have a benchmark alone show me what it means to denormalize my data model. Denormalization can help scalability and sometimes performance, but it has huge implications for maintainability, data quality, code complexity and even organisational matters.<p>I do understand the advantages of being able to run any algorithm on data and not be limited by SQL. But abolishing normalized data models altogether and replacing them with application code and custom data structures everywhere makes me very skeptical as someone who had to deal with data quality issues for many years.<p>So that's why I'm glad to see some performance issues of RDBMS improved by new storage architectures and also the innovation going on in terms of recursive queries to deal with hierarchical and graph data structures.
评论 #983738 未加载
chime超过 15 年前
One of the most actively-used databases at my company is about 1GB in size and grows at about 100MB/month. It is a Windows-based DB and is very heavy on the reads. I installed SuperVolume <a href="http://www.superspeed.com/desktop/supervolume.php" rel="nofollow">http://www.superspeed.com/desktop/supervolume.php</a> on the DB server and created a 4GB disk partition to store the DB files. SuperVolume mirrors the entire 4GB disk partition in RAM and works flawlessly and transparently.<p>I was going to get an SSD for performance enhancement but I don't much see a reason to do that anymore. All the reads happen at the speed of RAM and writes are delayed and non-blocking. Nested queries that used to take 15 minutes now take 45 seconds.<p>Right now the speed issues we have are solely due to unoptimized algorithms and not due to hardware limitations. This is a wonderful spot for a coder to be in because there is so much room for improvement now. If you have Windows systems serving files and data on slow systems, I would highly recommend giving SuperVolume and its related apps a shot before investing in expensive SSDs.
Luyt超过 15 年前
I put SSDs in all my development machines, and it indeed makes compiling faster (about three times). When prices drop a little, I'll also equip our database server with SSDs. Luckily the database is only 60G.
评论 #983670 未加载
akamaka超过 15 年前
I'd like to see a proper quantitative analysis done on this subject.<p>When you take money into account, are SSDs really worth it, or should I just buy more RAM?
评论 #984417 未加载
评论 #983604 未加载
评论 #983230 未加载
richcollins超过 15 年前
"RDBMS systems are headed into main memory, which changes the playing field all together"<p>It doesn't change the fact that RDBMSs don't do a good job of returning representations of graphs.
评论 #983376 未加载