Return-Path: Return-Path: Sender: Tejun Heo Date: Thu, 2 Aug 2018 08:59:17 -0700 From: Tejun Heo To: Dennis Zhou Cc: Jens Axboe , Josef Bacik , Johannes Weiner , kernel-team@fb.com, linux-block@vger.kernel.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH v3] block: make iolatency avg_lat exponentially decay Message-ID: <20180802155917.GT1206094@devbig004.ftw2.facebook.com> References: <20180802061541.49173-1-dennisszhou@gmail.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <20180802061541.49173-1-dennisszhou@gmail.com> List-ID: Hello, Dennis. On Wed, Aug 01, 2018 at 11:15:41PM -0700, Dennis Zhou wrote: > From: "Dennis Zhou (Facebook)" > > Currently, avg_lat is calculated by accumulating the mean of every > window in a long running cumulative average. As time goes on, the metric > becomes less and less useful due to the accumulated history. > > This patch reuses the same calculation done in load averages to make the > avg_lat metric more lively. Unlike load averages, the avg only advances > when a window elapses (due to an io). Idle periods extend the most > recent window. Bucketing is used to limit the history of avg_lat by > binding it to the window size. So, the window range for 1/exp (decay > rate) is [1 min, 2.5 min) when windows elapse immediately. > > The current sample window size is exposed in the debug info to enable > calculation of the window range. > > Signed-off-by: Dennis Zhou > Acked-by: Tejun Heo > Acked-by: Johannes Weiner > Acked-by: Josef Bacik Heh, sorry about not thinking this up from the beginning but I think it'd be great to move this to blkcg core so that the stat is always available regardless of blk-iolatency. It's a really important metric and can be used for both monitoring and policy implementation. Thanks! -- tejun