Comments on: A Peer to Peer test distribution system (TestBot)?

By: weight loss diets

weight loss diets — Fri, 06 Feb 2009 17:56:47 +0000

Many now interests how correctly to eat. The number of the people dissatisfied with the figure or health recently has
increased and, as consequence, trying to get rid of excess weight. You should pick up a diet approaching you, and also learn to make correctly balanced diet.

By: nutrition foods

nutrition foods — Thu, 29 Jan 2009 17:03:17 +0000

Thanks for the info. May God have mercy on us all.

By: nutrition foods

nutrition foods — Thu, 29 Jan 2009 16:56:21 +0000

Thanks for the info. May God have mercy on us all.

By: jnoller

jnoller — Wed, 10 Sep 2008 21:20:51 +0000

Yeah – I'm taking the distributed build thing a bit farther. I want a (globally)disparate series of test clients available to test any application, where those clients might test the app “locally” – in the case of desktop apps for instance, or they might execute a test passed to them which uses the client as CPU to bind to a larger test “in the swarm”.

I've written three different “manager passes tests to clients” and gets results back – those are easy, and relatively dumb.

I would much rather have the slaves register with the server and provide the weight number which indicates the “desirability” of the client – a simple way of doing this would be to pass an object across the wire containing the .network_hops_from_target, .ram, .cpu and so on attributes, and do the calculations on manager-side.

Doing it the simple way has the additional benefit of allowing a test-in-the-queue to dictate what attribute it's more interested in. Of course, if you calculate the weight numbers the clients pass back correctly – you don't need the tests to dictate what it “wants”.

Argh, it's still a jumble of ideas.

By: Orestis Markou

Orestis Markou — Wed, 10 Sep 2008 21:08:58 +0000

We have something like that in Resolver Systems – we call it “distributed build”.

It builds your working copy, copies it to the machines you specify in the LAN and publishes the list of tests it wants to be run.

A central server assigns tests to machines and gathers the result back. It's not clever, but it does its job. I've always wanted to make it more smart so that new machines could be added transparently when they are idle, but it was always too much work…

By: jnoller

jnoller — Mon, 08 Sep 2008 17:42:30 +0000

Yup. I wanted to go to GTAC last year (and this year too) but didn't have the chance. Skoll is interesting for a variety of reasons, and there are some parallels to what I am talking about in the distributed-slave-sense. I'm gleaning what seems to be the more meaty parts of Skoll from the papers and publications.

By: terry peppers

terry peppers — Mon, 08 Sep 2008 17:14:24 +0000

Jesse -

You mention Skoll. And I'm sure you've already seen Adam Porter and Atif Memon talk from GTAC last year. Very interesting concepts.

http://www.youtube.com/watch?v=OiE9zRPD6ps