Why the difference?

Betreger
Betreger
Joined: 25 Feb 05
Posts: 991
Credit: 1,551,095,179
RAC: 714,417
Topic 197725

I have been sent Gamma-ray pulsar search tasks with est run times. Some are over 10 hrs and some are a bit over 2. Why is the project splitting 2 differnt length jobs? OBTW the ests are reasonably accurate and the credit earned per hr is consistent so this is not a complaint.

archae86
archae86
Joined: 6 Dec 05
Posts: 3,156
Credit: 7,175,504,931
RAC: 713,233

Why the difference?

I've seen considerably more than two lengths. If you look at Stderr output listing on the task page after an FGRP4 job is finished, you can see two things which so far agree with each other and which come pretty close to classifying the relative execution time needed:

One is stated as showing how many skypoints there will be, in a line like this which can be found a dozen or so lines past the very beginning shortly before the big gaps start:
% Sky point 1/3

The other is near the very end, and is the last of a sequential set indicating checkpoints, with the last one looking like this:

% checkpoint 3
% Time spent on semicoherent stage: 4148.8073s
% Writing semicoherent output file.

% Following up candidate number: 1


and so on for five candidates...

Once I noticed the skypoints number, which is recently, it so far has matched the maximum checkpoint number, or been one more.

On the "production" FGRP4 work, I have seen work units with at least the following number of skypoints.:

3
5
7
12
13
16
19
20
21
23
30

Of these, the 30 checkpoint/30 skypoint type is by far the most common, so far.

Gary Roberts or another guru has indicated that these smaller jobs are termed "short ends" and perhaps a search on that term might give some insight into why these exist. I confess I don't know.

mikey
mikey
Joined: 22 Jan 05
Posts: 12,538
Credit: 1,838,585,330
RAC: 3,585

RE: Gary Roberts or

Quote:

Gary Roberts or another guru has indicated that these smaller jobs are termed "short ends" and perhaps a search on that term might give some insight into why these exist. I confess I don't know.

I believe the 'short ends' are the units that either had problems for some reason and the project is retrying to get them crunched, had been aborted and needed resending but just hadn't been, and lastly just general clean-up of the odds and end units that need crunching before the Project moves on to another type of unit. The 'short ends' meaning the short end of the stick as far as getting closer to the end of this type of unit.

I'm also guessing that's why the units are different sizes too, they are the leftovers that need crunching, but just haven't been yet for one reason or another. With some units having lots of data in them, and others having very little, the more usable data the software finds the longer the unit takes to finish. That's NOT to say the short units aren't usable, they are, and the longer ones are not MORE usable because of all of their possible data in them. ALL the units are needed to make them all useful.

Holmis
Joined: 4 Jan 05
Posts: 1,118
Credit: 1,055,935,564
RAC: 0

My understanding of "short

My understanding of "short ends" is that the data files that the work unit generator slices up into work units are of different lengths but the work units are "all" of the same size/lenght, when the generator reaches the end of the file there will be a "short end" left that might vary in size.

Comment viewing options

Select your preferred way to display the comments and click "Save settings" to activate your changes.