From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from mail-ot0-f175.google.com ([74.125.82.175]:38424 "EHLO mail-ot0-f175.google.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752832AbeAFSKj (ORCPT ); Sat, 6 Jan 2018 13:10:39 -0500 Received: by mail-ot0-f175.google.com with SMTP id h2so6365132oti.5 for ; Sat, 06 Jan 2018 10:10:39 -0800 (PST) MIME-Version: 1.0 In-Reply-To: References: From: James Courtier-Dutton Date: Sat, 6 Jan 2018 18:09:58 +0000 Message-ID: Subject: Re: btrfs balance problems To: Duncan <1i5t5.duncan@cox.net> Cc: linux-btrfs@vger.kernel.org Content-Type: text/plain; charset="UTF-8" Sender: linux-btrfs-owner@vger.kernel.org List-ID: On 28 December 2017 at 00:39, Duncan <1i5t5.duncan@cox.net> wrote: > > AFAIK, ionice only works for some IO schedulers, not all. It does work > with the default CFQ scheduler, but I don't /believe/ it works with > deadline, certainly not with noop, and I'd /guess/ it doesn't work with > block-multiqueue (and thus not with bfq or kyber) at all, tho it's > possible it does in the latest kernels, since multi-queue is targeted to > eventually replace, at least as default, the older single-queue options. > > So which scheduler are you using and are you on multi-queue or not? > Thank you. The install had defaulted to deadline. I have now switched it to CFQ, and the system is much more responsive/interactive now during a btrfs balance. I will test it when I next get a chance, to see if that has helped me. After reading about it: deadline: more likely to complete long sequential reads/writes and not switch tasks.Thus reducing the amount of seeking but impacting concurrent tasks. cfq: more likely to break up long sequential reads/writes to permit other tasks to do some work. Thus increasing the amount of seeking but helping concurrent tasks. This would explain why "cfq" is best for me. I have not yet looked at "multi-queue".