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.

PolarFS: Alibaba Distributed File System for Shared Storage Cloud Database [pdf]

137 pointsby blopeuralmost 7 years ago

8 comments

antoncohenalmost 7 years ago
Man, reading that made me wish Clustrix (YC &#x27;06) open sourced their database (<a href="https:&#x2F;&#x2F;www.clustrix.com&#x2F;" rel="nofollow">https:&#x2F;&#x2F;www.clustrix.com&#x2F;</a>). They had a MySQL compatible scale our DB nearly 10 years ago, wireline compatible with MySQL without using any MySQL code, could participate in a MySQL replication cluster with normal MySQL servers (made migration easy). It was scale out shared-nothing, writes would scale linearly as you added nodes, unlike POLARDB which is shared-everything with a single master. It used RDMA 10 years ago, and custom PCIe devices because NVMe didn&#x27;t exist.<p>But they didn&#x27;t open source it, so only a small handful of companies get to use it. Sad.
评论 #17818957 未加载
KaiserProalmost 7 years ago
The hardest part of a Distributed File system (and I mean File system here) is managing the Meatadata (where a file is, where the directory is, who last did something to it.)<p>Lustre, GFS2 and GPFS all have centralised metadatstores, which is both a boon and a drawback.<p>What I can&#x27;t figure out is what they&#x27;ve done here. It appears like metadata is stored in a special partition (&quot;journal&quot;) which is shared? But there is a control process as well.
评论 #17817885 未加载
评论 #17818455 未加载
j16sdizalmost 7 years ago
The protocol is interesting.<p>But given how often the Alibaba cloud fails in production, I won&#x27;t hold my breath.
评论 #17816477 未加载
评论 #17816188 未加载
antongribokalmost 7 years ago
Kind of disappointing that they compared it with Ceph non-RDMA vs. PolarFS on RMDA, unless I misread this part.<p>Still, this is all very interesting.
sekh60almost 7 years ago
I am rather sleep deprived, so I may have misread things, but this doesn&#x27;t seem to me to be the best benchmark to evaluate Ceph for database work.<p>From what I understand best practice in ceph for databases is to make a rbd image and format that with your filesystem of choice. I believe. The rbd stripe size should be tuned to you database writes in mind.<p>I believe ceph rbd supports rdma, but I cannot find much current details about it.
评论 #17816669 未加载
RobLachalmost 7 years ago
Most interesting bit is their consensus protocol.
评论 #17816456 未加载
评论 #17817476 未加载
toolslivealmost 7 years ago
link is down. Is there an alternative link?
cbsmithalmost 7 years ago
RDMA is still an &quot;emerging technology&quot;?
评论 #17818130 未加载