launchpad-dev team mailing list archive

[Martin Pool <mbp@xxxxxxxxxxxxx>]

On 4 August 2010 17:33, Stuart Bishop <stuart.bishop@xxxxxxxxxxxxx> wrote:
> On Wed, Aug 4, 2010 at 9:46 AM, Robert Collins
> <robert.collins@xxxxxxxxxxxxx> wrote:
>
>> Doing this will immediately close a half-dozen bugs, and focus our
>> timeout and performance efforts closer to the actual source of our
>> problems.
>
> Which bugs? The only open ones I'm aware of are feature requests which
> I don't think are on anyone's radar to address.

Milestones pages don't update when you close bugs targetted to the
milestone.  It rather spoils the moment of closing your last bug.

Likewise bug counts <https://bugs.edge.launchpad.net/malone/+bug/602936>.

There are others.

They can in principle be squashed individually but I don't know if
it's worth it.

> I'm not sure why turning the facility off will help peoples focus -
> memcached was never about stopping timeouts and this has been
> repeatedly stressed. Using the facility will slow down initial loads,
> and may improve the median page load time. It won't fix timeouts. It
> may reduce the frequency of timeouts. The memcached infrastructure is
> about improving scalability and overall performance, and throwing away
> our 24% hit rate seems rather pointless (better rate than I was
> expecting actually, but I guess the places it is being used have been
> specifically targeted). The same rationales for turning off would be
> applied to turning off Squid, which fills pretty much the same role
> but more so by caching 100% of our unauthorized access.

It might be helping with scalability but I don't think at the moment
it's helping much with user perceived performance.  If something is
slow (but not timing out) 76% of the time then being faster 24% of the
time is not going to make Launchpad feel fast.  (It may be that it's
much better than 24% on some particular pages and really poor on
others; data welcome.)

If it was making pages really dazzlingly fast 99% of the time and then
occasionally they'd be slow that would be perhaps a good tradeoff; at
least it might be lost in the noise of page loads sometimes being slow
for reasons other than server side time.  Even then I don't think that
giving confusing incorrect results is worth it.

Perhaps we could turn memcached off entirely for a day or an hour and
see if the oops rate or server load changes?

-- 
Martin