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。

How does a relational database work?

377 点作者 mgachka超过 9 年前

18 条评论

why-el超过 9 年前
Good write up. Another excellent resource straight out of the UC Berkeley Database Group that I keep close by is &quot;Architecture of a Database System&quot;[1] by three researchers in the field. It is very readable.<p>[1] <a href="http:&#x2F;&#x2F;db.cs.berkeley.edu&#x2F;papers&#x2F;fntdb07-architecture.pdf" rel="nofollow">http:&#x2F;&#x2F;db.cs.berkeley.edu&#x2F;papers&#x2F;fntdb07-architecture.pdf</a>
评论 #10091761 未加载
jandrewrogers超过 9 年前
Good overview of traditional OLTP architectures. As complicated as they look from the article, it is just scratching the surface of a sophisticated implementation. There are many internals common to more advanced designs that are not even mentioned, and the article is already quite long!<p>The thing I love most about database engines is that there is probably more hardcore computer science per line of code than any other software system of similar scope. It is a very rich ecosystem for an algorithms and data structures geek.
faragon超过 9 年前
Be careful with theoretical asymptotic complexity (big O) related to execution time. E.g. if your algorithm time complexity is O(1), but internally calls a higher complexity function, e.g. malloc(), implemented with higher complexity, e.g. O(log n), your algorithm time complexity would be O(log n) and not O(1). It could be even worse: <i>on average</i> or <i>typical</i> constant time algorithm could be in reality an O(n) one: e.g. case of hash table reindexation (that&#x27;s the reason of why many big data structures, including most SQL databases, requiring real time behavior, are implemented as trees, tree hierarchies&#x2F;division&#x2F;clustering, instead of big hash tables).
评论 #10090024 未加载
评论 #10090528 未加载
brudgers超过 9 年前
Because I am interested in databases, I found the Se-radio&#x27;s 2013 interview with Michael Stonebreaker [1] interesting, particularly in regard to traditional database design and more recent ideas:<p><a href="http:&#x2F;&#x2F;www.se-radio.net&#x2F;2013&#x2F;12&#x2F;episode-199-michael-stonebraker&#x2F;" rel="nofollow">http:&#x2F;&#x2F;www.se-radio.net&#x2F;2013&#x2F;12&#x2F;episode-199-michael-stonebra...</a><p>[1]: <a href="http:&#x2F;&#x2F;www.theregister.co.uk&#x2F;2015&#x2F;03&#x2F;25&#x2F;mike_stonebraker_wins_turing_award&#x2F;" rel="nofollow">http:&#x2F;&#x2F;www.theregister.co.uk&#x2F;2015&#x2F;03&#x2F;25&#x2F;mike_stonebraker_win...</a>
otis_inf超过 9 年前
&gt; When it comes to relational databases, I can’t help thinking that something is missing. They’re used everywhere. There are many different databases: from the small and useful SQLite to the powerful Teradata. But, there are only a few articles that explain how a database works.<p>That&#x27;s because the inner workings are really old, as in: emerged before blogging etc. was popular, hell before the internet was invented.<p>In the &#x27;before&#x2F;early internet days&#x27;, we read books like &#x27;An introduction to Database Systems&#x27; by C.J. Date. (I had to blow the dust off my copy to read the exact title ;)), which are more in depth than this article, but I like the article better, because it&#x27;s more to the point and easier to understand. Well done!
评论 #10088144 未加载
mgrennan超过 9 年前
Not that Cassandra and Hadoop don&#x27;t have a place. But because NO-SQL is hot I see lots of young coders (I&#x27;m and old DBA) try to turn document store systems into relational databases. They should all be made to read this post.
评论 #10087146 未加载
n0us超过 9 年前
Actually this is the best post I have ever seen on this website.
jlees超过 9 年前
This is also a pretty accessible quick intro to complexity and data structures, nicely done. Definitely the sort of thing I would include as further reading in a beginner course -- some beginners love to understand &quot;why&quot; and this post answers pretty much all the &quot;why&quot; possible.
codezero超过 9 年前
I decided to spend some time digging into SQLite. I highly recommend the overviews of their architecture and the details about each part of the puzzle.<p>It&#x27;s really understandable, very straight forward, even if a lot of it refers to SQLite v2, it still seems very relevant.<p><a href="http:&#x2F;&#x2F;www.sqlite.org&#x2F;arch.html" rel="nofollow">http:&#x2F;&#x2F;www.sqlite.org&#x2F;arch.html</a>
emehrkay超过 9 年前
I&#x27;m I crazy for wanting to write a database after reading this? Noting too serious, just to flex that dev muscle
评论 #10090567 未加载
评论 #10088694 未加载
aikah超过 9 年前
Great Article , I wish a book was written where a simple database with a query language was implemented from start to finish , even a nosql one, I always wanted to implement my own.
buckbova超过 9 年前
I&#x27;ve read sql server internals cover to cover and in many respects this is a much better read. Thank you.
Cakez0r超过 9 年前
&gt; Nowadays, many developers don’t care about time complexity … and they’re right!<p>That&#x27;s a pretty bold statement...<p>Very thorough explanations though!
评论 #10089630 未加载
njharman超过 9 年前
I believe that to be the best technical document I&#x27;ve ever read. Surely biased as I learned so much.
评论 #10092475 未加载
0xCMP超过 9 年前
Wow, I wanted this recently. Anyone know stuff related to graph databases and how those work?
评论 #10089400 未加载
mgrennan超过 9 年前
Good read. How long do your keep your transactions logs and how often do you make backups?
0xCMP超过 9 年前
Anyone know of any rust based databases being worked on? Relational or otherwise...
评论 #10089831 未加载
beenpoor超过 9 年前
Great article!