Dream5 experiment
log in

Advanced search

Message boards : Number crunching : Dream5 experiment

Author Message
Profile valterc
Project administrator
Project tester
Send message
Joined: 30 Oct 13
Posts: 320
Credit: 16,290,071
RAC: 4,086
Italy
Message 832 - Posted: 28 Jan 2017, 11:08:13 UTC
Last modified: 28 Jan 2017, 11:29:55 UTC

We are adding a different experiment: E. coli based on Dream5 expression dataset. There should be around 300 batches. The name of the workunit will contains something like "oge_dream5_net3", the dataset will be download once, its name is net3_expression_data_pc.csv. This kind of workunit should be faster than usual, I will monitor this and try to adjust the running time, if needed.

More scientific information about this experiment will be posted in the Science section of the forum.

If you catch one of those workunits and have some kind of problem, please let me know.

Dave Peachey
Send message
Joined: 6 Nov 16
Posts: 7
Credit: 2,364,725
RAC: 34
United Kingdom
Message 833 - Posted: 28 Jan 2017, 13:19:11 UTC - in response to Message 832.

If you catch one of those workunits and have some kind of problem, please let me know.

I've had no problems with the eleven that recently turned up in my queue and which I selected to turn around immediately.

In comparison to the "ecm" units:
- the "dream5" WUs complete in around 6 to 7 mins (although one finished in 2 mins!) vs. 30 to 32mins for the "ecm" WUs
- the output file size is just over 1MB vs. approx. 350Kb for the "ecm" WUs

This is on a Win7 machine running an i7 6900K CPU and using the FMA version of Daniel's latest optimised application.


Post to thread

Message boards : Number crunching : Dream5 experiment


Main page · Your account · Message boards


Copyright © 2017 CNR-TN & UniTN