From: Paolo Bonzini <pbonzini@redhat.com>
To: Stefan Hajnoczi <stefanha@gmail.com>
Cc: kwolf@redhat.com, "Benoît Canet" <benoit@irqsave.net>,
stefanha@redhat.com, qemu-devel@nongnu.org
Subject: Re: [Qemu-devel] [PATCH V5 1/5] throttle: Add a new throttling API implementing continuous leaky bucket.
Date: Mon, 19 Aug 2013 13:27:50 +0200 [thread overview]
Message-ID: <52120136.4070401@redhat.com> (raw)
In-Reply-To: <20130816114505.GA22193@stefanha-thinkpad.redhat.com>
Il 16/08/2013 13:45, Stefan Hajnoczi ha scritto:
>> > +#define BUCKETS_COUNT 6
>> > +
>> > +typedef enum {
>> > + THROTTLE_BPS_TOTAL = 0,
>> > + THROTTLE_BPS_READ = 1,
>> > + THROTTLE_BPS_WRITE = 2,
>> > + THROTTLE_OPS_TOTAL = 3,
>> > + THROTTLE_OPS_READ = 4,
>> > + THROTTLE_OPS_WRITE = 5,
>> > +} BucketType;
Please remove the "= N" from the enums, and add BUCKETS_COUNT here.
>> > +typedef struct LeakyBucket {
>> > + double ups; /* units per second */
>> > + double max; /* leaky bucket max in units */
>> > + double bucket; /* bucket in units */
> These comments aren't very clear to me :). So I guess bps or iops would
> be in ups. Max would be the total budget or maximum burst. Bucket
> might be the current level.
I also suggest replacing "ups" with "avg", since it's the average
throughput that the leaky bucket allows after the initial burst has
emptied the bucket.
>> + uint64_t unit_size; /* size of an unit in bytes */
>> + uint64_t op_size; /* size of an operation in units */
>
> It's not clear yet why we need both unit_size *and* op_size. I thought
> you would have a single granularity field for accounting big requests as
> multiple iops.
IIUC the ops buckets account operations in op_size / unit_size units,
while the bps buckets account operations in 1 / unit_size units, or
something like that. But it needs clarification indeed.
Paolo
next prev parent reply other threads:[~2013-08-19 11:28 UTC|newest]
Thread overview: 17+ messages / expand[flat|nested] mbox.gz Atom feed top
2013-08-12 16:53 [Qemu-devel] [PATCH V5 0/5] Continuous Leaky Bucket Throttling Benoît Canet
2013-08-12 16:53 ` [Qemu-devel] [PATCH V5 1/5] throttle: Add a new throttling API implementing continuous leaky bucket Benoît Canet
2013-08-14 8:52 ` Fam Zheng
2013-08-16 11:45 ` Stefan Hajnoczi
2013-08-19 11:27 ` Paolo Bonzini [this message]
2013-08-12 16:53 ` [Qemu-devel] [PATCH V5 2/5] throttle: Add units tests Benoît Canet
2013-08-12 16:53 ` [Qemu-devel] [PATCH V5 3/5] block: Enable the new throttling code in the block layer Benoît Canet
2013-08-14 9:31 ` Fam Zheng
2013-08-14 9:50 ` Fam Zheng
2013-08-16 12:02 ` Stefan Hajnoczi
2013-08-12 16:53 ` [Qemu-devel] [PATCH V5 4/5] block: Add support for throttling burst max in QMP and the command line Benoît Canet
2013-08-16 12:07 ` Stefan Hajnoczi
2013-08-12 16:53 ` [Qemu-devel] [PATCH V5 5/5] block: Add iops_sector_count to do the iops accounting for a given io size Benoît Canet
2013-08-14 9:48 ` Fam Zheng
2013-08-14 18:31 ` Benoît Canet
2013-08-16 12:10 ` Stefan Hajnoczi
2013-08-16 12:14 ` [Qemu-devel] [PATCH V5 0/5] Continuous Leaky Bucket Throttling Stefan Hajnoczi
Reply instructions:
You may reply publicly to this message via plain-text email
using any one of the following methods:
* Save the following mbox file, import it into your mail client,
and reply-to-all from there: mbox
Avoid top-posting and favor interleaved quoting:
https://en.wikipedia.org/wiki/Posting_style#Interleaved_style
* Reply using the --to, --cc, and --in-reply-to
switches of git-send-email(1):
git send-email \
--in-reply-to=52120136.4070401@redhat.com \
--to=pbonzini@redhat.com \
--cc=benoit@irqsave.net \
--cc=kwolf@redhat.com \
--cc=qemu-devel@nongnu.org \
--cc=stefanha@gmail.com \
--cc=stefanha@redhat.com \
/path/to/YOUR_REPLY
https://kernel.org/pub/software/scm/git/docs/git-send-email.html
* If your mail client supports setting the In-Reply-To header
via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line
before the message body.
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).