[tahoe-dev] Benchmarks - 1.7.1 vs. 1.8.0c3

Kyle Markley kyle at arbyte.us
Thu Sep 9 05:29:24 UTC 2010


Brian,

> Please eyeball a couple of the download-status pages and see if you can
> learn anything about the number of servers being selected. I imagine
> that on your local grid, all the servers are identical, but since the
> performance variations we've seen on the testgrid/v-grid have been
> closely correlated to server-selection, it'd be good to find a way to
> rule that out here.

My environment destroys the grid as soon as the testing is done.  I could
artificially keep it alive, but then I have very little experience
interpreting the download information pages.


> What I'd love to have is a test framework in which we could simulate
> multiple servers (of varying CPU speeds) connected by virtual pipes with
> controllable latency and bandwidth limits. Then we could run an
> experiment like measuring download speed as the bandwidth limits are
> raised, where I'd expect to see it taper off as we approach the
> CPU-speed limit. (I remember writing this desire up on a whiteboard at
> allmydata about four years ago and labelling it "and also, I want a
> pony"). A lot of the performance measurements we've been making are at
> unlabled points along some axes of CPU-speed, network-speed, and share
> distribution.. being able to label those points and compare apples to
> apples would probably help us understand what's going on a lot better.

Your pony interests me.  A lot.

For the past couple years, actually, as I've been disappointed by some
software at work I've been thinking occasionally about an auto-prefetching
data stream that would prefetch at the measured rate of consumption in
order to eliminate the read latency without prefetching too much (which
hurts CPU caching).  A bandwidth goal is a fundamental part of this; fixing
it to a given value is a subset of what I'd wanted to do anyway.

I haven't gone beyond sketching some notes for this idea.  (I'm on the
wrong team at work to implement this there.)  But having a real-world
application for it could give me an incentive to go create this on my own. 
It's not a large project but I don't have a lot of hobby time anymore. 
Hmmm...

-- 
Kyle Markley


More information about the tahoe-dev mailing list