Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1532580 > unrolled thread

Re: [PATCH V4 13/15] blk-throttle: add a mechanism to estimate IO latency

Back to article view | Back to linux.kernel

This discussion starts older than the indexed window; earlier articles aren't shown. The article labeled Started by below is the oldest one visible, not the original post.

Contents

  Re: [PATCH V4 13/15] blk-throttle: add a mechanism to estimate IO  latency Tejun Heo <tj@kernel.org> - 2016-11-29 18:30 +0100
    Re: [PATCH V4 13/15] blk-throttle: add a mechanism to estimate IO  latency Tejun Heo <tj@kernel.org> - 2016-11-29 23:40 +0100

#1532580 — Re: [PATCH V4 13/15] blk-throttle: add a mechanism to estimate IO latency

Hello, Shaohua.

On Mon, Nov 14, 2016 at 02:22:20PM -0800, Shaohua Li wrote:
> To do this, we sample some data, eg, average latency for request size
> 4k, 8k, 16k, 32k, 64k. We then use an equation f(x) = a * x + b to fit
> the data (x is request size in KB, f(x) is the latency). Then we can use
> the equation to estimate IO target latency for any request.

As discussed separately, it might make more sense to just use the avg
of the closest bucket instead of trying to line-fit the buckets, but
it's an implementation detail and whatever which works is fine.

> Hard disk is completely different. Latency depends on spindle seek
> instead of request size. So this latency target feature is for SSD only.

I'm not sure about this.  While a disk's latency profile is way higher
and more erratic than SSDs, that doesn't make latency target useless.
Sure, it'll be more crude but there's a significant difference between
a cgroup having <= 20ms overall latency and experiencing multi-sec
latency.

Thanks.

-- 
tejun

[toc] | [next] | [standalone]

#1532826

Hello,

On Tue, Nov 29, 2016 at 10:30:44AM -0800, Shaohua Li wrote:
> > As discussed separately, it might make more sense to just use the avg
> > of the closest bucket instead of trying to line-fit the buckets, but
> > it's an implementation detail and whatever which works is fine.
> 
> that is still like a line fit. Don't think there is big difference.

Yeah, just wondering whether that'd be simpler.

> > > Hard disk is completely different. Latency depends on spindle seek
> > > instead of request size. So this latency target feature is for SSD only.
> > 
> > I'm not sure about this.  While a disk's latency profile is way higher
> > and more erratic than SSDs, that doesn't make latency target useless.
> > Sure, it'll be more crude but there's a significant difference between
> > a cgroup having <= 20ms overall latency and experiencing multi-sec
> > latency.
> 
> Sure, latency target is useful for hardisk too. But we need a different
> stragety. For hard disk, the latency highly depends on seek. Probably we can
> make the latency target the same for all request size. Not sure if average
> latency makes sense. Need more tests with hard disk. I'd like to forcus on SSD
> in current stage.

Sure, it's fine to focus on SSDs for now but with either line fitting
or bucketed avg, it should be fine, right?  The slope of the line
would be way lower and the deviation would be higher but that doesn't
really get in the way here.

Thanks.

-- 
tejun

[toc] | [prev] | [standalone]

Back to top | Article view | linux.kernel

csiph-web