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.

Unmapped I/O improves userland direct disk performance by 35% ~ 135%

85 pointsby antoniosalmost 9 years ago

6 comments

pritambaralalmost 9 years ago
From the title, I thought Microsoft had sponsored an improvement in generic FreeBSD.<p>From the revision, it seems the improvement is in a Hyper-V storage driver.
评论 #12186651 未加载
评论 #12187099 未加载
exeliusalmost 9 years ago
I mean, this only makes sense -- OSes are no longer a core product in the server world; they&#x27;re just a part of the application stack. The hypervisor &#x2F; orchestrator is the critical part, and that&#x27;s Hyper-V in the Microsoft world.<p>Also, this patch doesn&#x27;t strike me as monumental. Really, it strikes me as one of the first things you would do when configuring a disk driver for a VM (mapped I&#x2F;O is unnecessary overhead for most VM configurations as the hypervisor and&#x2F;or SAN remaps it anyway). Has Microsoft not supported FreeBSD well under HyperV in the past?
评论 #12186934 未加载
评论 #12186865 未加载
sitkackalmost 9 years ago
I don&#x27;t understand mapped vs unmapped io in terms of Hyper-V. Does that mean that io from a user process goes directly to a paravirtualized io device and bypasses the freebsd kernel?
评论 #12188917 未加载
mrfusionalmost 9 years ago
Can anyone eli5?
评论 #12187831 未加载
ajamesmalmost 9 years ago
I see now that the patch was sponsored by Microsoft, but for a moment I thought that this was a new kind of HN text ad.
评论 #12187241 未加载
评论 #12187244 未加载
honkhonkpantsalmost 9 years ago
The style of this diff is bewildering. What&#x27;s with the gratuitous reformatting of the whitespace? Makes it difficult for the reader to pick out the meaningful difference.
评论 #12188071 未加载
评论 #12187640 未加载