|
From: | Eric Blake |
Subject: | Re: [Qemu-devel] qemu-nbd performance |
Date: | Wed, 19 Sep 2018 10:29:53 -0500 |
User-agent: | Mozilla/5.0 (X11; Linux x86_64; rv:60.0) Gecko/20100101 Thunderbird/60.0 |
On 9/19/18 1:46 AM, lampahome wrote:
All images are on the RAID0(3 SSD).Below is the performance:image number seq. read(MB/s) seq. write(MB/s) 1 1480 60 16 1453 36 32 1450 36 64 1430 36 128 1400 36 It shows reading better than writing. I confused about why, does qemu did something to delay the write?
Not intentionally. But you'd have to profile things to learn where the hotspots are, in order to figure out how to reduce the overhead.
-- Eric Blake, Principal Software Engineer Red Hat, Inc. +1-919-301-3266 Virtualization: qemu.org | libvirt.org
[Prev in Thread] | Current Thread | [Next in Thread] |