S5R5 plans
Anonymous
30 Oct 2008 21:48:05 UTC
Topic 13713
(moderation:
Quote:
Any news about the S5R5 run mentioned earlier in this thread?? it was supposed to begin soon, because of some kind of bug on the results of the S5R4...
Will definitely come. We're still testing and tweaking the setup. Some facts so far:
- slightly increased memory requirement
- larger "dwell time" per sky location. This is limiting the maximum checkpoint rate to about once per 3 min (on current average CPUs, longer on slower ones)
- workunits will run roughly half as long as S5R4 ones
BM
S5R5 plans
)
The S5R5 Windows App will feature the code that makes the S5R4 6.05 App as fast as it is.
There is no change in crediting targeted for S5R5.
BM
BM
The "deadline" is something
)
The "deadline" is something that can rather easily be adjusted on-the-fly during a run (whereas the average workunit duration is not). With current workunits the average is set to be 18d, for what I recall from previous discussions I think most people feel comfortable with about 14d for 6-8h WUs.
BM
BM
RE: Any idea when the cut
)
Currently there's a telecon scheduled for tomorrow which covers the subject. Once the final decision is made actually starting the run is a matter of days.
I'll keep you posted.
BM
BM
RE: RE: Any idea when the
)
Directly from telecon: Further (internal) testing / simulations needed, expected timeframe for this is another week.
BM
BM
Update: Simulations show
)
Update:
Simulations show that with the currently planned (and preliminarily implemented) S5R5 setup we would miss some signals. More tuning needed, will take at least another week.
BM
BM
Hi Gary! Yes, with a bit of
)
Hi Gary!
Yes, with a bit of luck (i.e. if we don't find a problem in the very last minute) we'll start S5R5 around the weekend. We were just waiting for a green light that we got last night.
Due to some remaining uncertainties we will restrict the frequency range of S5R5 to 1000Hz and then take a look at the results to decide whether we should push it up to the 1250Hz of the current S5R4. This reduces the projected total runtime of (the possibly first part of) S5R5 to about 6 months. We are working on some hoefully even more sensitive and efficient analysis methods (improving the "Hough transform"). If a new program is ready to be used by then, we'll also not extend S5R5, but start an S5R6.
The only consequence of this decision right now is that clients currently running S5R4 workunits above 1000Hz will need to get a new set of data files, while the others will just get S5R5 work for the data files they already have.
We'll stop the S5R4 workunit generator, but the workunits generated so far will be finished and credited. I'm not sure that pushing the last ones through is necessary, so I probably won't put time into this.
There is no (intentional) change in the crediting, so wrt the credit it shouldn't matter whether you run S5R4 or S5R5 workunits. The S5R5 ones will run a bit shorter (design goal was 50%, but I'm afraid with the adjustments we had to make afterwards we missed it by about 10%).
BM
BM
RE: RE: the relative
)
Actually it's more predictable than ever. Thanks to the work of Bikeman the granted credit will more accurately match the runtime variation than ever before. Also too as the total Task runtime will be halved, even a 30% variation will be smaller in absolute time than before. Finally I think we got the floating point estimation better than ever and update the progress counter more frequently in the S5R5 App, so the client should be able to estimate the runtime more accurately.
BM
BM
RE: good question, 40 hour
)
If the currents S5R4 are 40h for you, S5R5 should be 20h on average.
We're running some final tests, S5R5 should start today or tomorrow if we don't find any oddities.
BM
BM
The S5R4 workunit generator
)
The S5R4 workunit generator (WUG) has been stopped and the S5R5 one been started instead. S5R5 has officially been launched.
The server status page has been updated. BTW ABP1 (currently under "previous searches") lists the workunits in the active database from the Arecibo Binary Pulsar search; currently the ones remaining from the test in December. The time estimations for S5R5 look a bit absurd, but that usually settles for some reasonable value after the WUG ran for a few days.
The runtimes of S5R5 tasks should be half the one you know from S5R4, but don't judge this after a few tasks - the runtime variation between different tasks of the same frequency is larger in S5R5.
We intend to keep the credit level (credits per CPU hour) the same as in S5R4, but due to the rather short internal testing we relied on some educated guessing. We think we got it pretty close, but might need to make some adjustment to that in a week or so.
BM
BM
Yes, the S5R4 workunits will
)
Yes, the S5R4 workunits will remain the majority for the next few days. We won't cancel them so people will get credit for them; this will also help our servers to cope with the transition.
BM
BM