Progress metric

log in

Advanced search

Message boards : Science : Progress metric

Author Message
Profile cIclops
Send message
Joined: 19 Feb 05
Posts: 26
Credit: 450
RAC: 0
Message 4857 - Posted: 23 Feb 2005, 19:11:28 UTC

What would be the best metric to measure the project's progress towards the goal of detecting a pulsar signature in the LSC data?
____________
--
searching for gravitational waves since 2005

Profile cIclops
Send message
Joined: 19 Feb 05
Posts: 26
Credit: 450
RAC: 0
Message 5366 - Posted: 25 Feb 2005, 15:11:32 UTC - in response to Message 4857.

> What would be the best metric to measure the project's progress towards the
> goal of detecting a pulsar signature in the LSC data?
>

As nobody has responded to this perhaps the question needs clarification.

Given that the aim of the Einstein@Home project is to detect a pulsar signature in the LSC data, how can the progress towards this be made visible?

My best guess is to display the percentage analysis complete of the total data set(s) and the number of detected candidate signals.

____________
--
searching for gravitational waves since 2005

Profile S@NL - Marleen
Avatar
Send message
Joined: 18 Jan 05
Posts: 25
Credit: 784,196
RAC: 1,173
Message 6001 - Posted: 27 Feb 2005, 22:19:01 UTC

You mean something like Seti Classic has on its Current progress summary page?
Of course it will need to be adjusted, eg the "Check data integrity" step is not needed because BOINC already takes care of that.

But it would be nice to have something like:
1. Amount of data gathered (eg x hours of LIGO data) and number of WU's that can be generated from this
2. Number of WU's finished (validated WU's)
3. Number of candidate signals found

Of course there could be other and/or extra steps, that depends on how the data analysis is done.


____________

Profile debugas
Avatar
Send message
Joined: 11 Nov 04
Posts: 171
Credit: 77,331
RAC: 0
Message 6091 - Posted: 28 Feb 2005, 7:53:48 UTC - in response to Message 6001.
Last modified: 28 Feb 2005, 7:56:31 UTC

This thread more belongs to Wish List section and not here to Science :)

As to the project progress - imagine that you gather sound signals from all over but leave to analyse it to someone else later. How can you estimate the progress ? Only formally by counting your WUs processed or the percentage of sky listened during given period of time

Profile Iron Sun 254
Avatar
Send message
Joined: 25 Feb 05
Posts: 38
Credit: 55,455
RAC: 0
Message 6138 - Posted: 28 Feb 2005, 14:40:36 UTC

While it may be possible to print out a list of potential pulsars found, I'm sure it's going to be a while before those signals can be verified through other means. It will probably be a while before any true numbers are available.
____________
------------------------------------
There's a thin line between Genius and Insanity. That's where I live, baby!

THE SPACEPORT - The Other Side of Space

epros
Send message
Joined: 21 Jun 05
Posts: 10
Credit: 23,438
RAC: 0
Message 14228 - Posted: 22 Jun 2005, 16:47:35 UTC

Having read this thread, I thought that the metric of progress would be really useful thing if its definition were sensible - not just about "how much hours wasted".

AFAIK the project tries to answer some question about the gravity waves. ANY definite answer ("yes" or "no") would be the project result. Presumably, we'l never get definite answer. But it's possible to estimate, how much credits we need to get 90% proof. It would be the real basis to measure current progress. Wouldn't it?
____________

Profile Ben Owen
Send message
Joined: 21 Dec 04
Posts: 117
Credit: 6,617,751
RAC: 12,993
Message 15266 - Posted: 10 Jul 2005, 0:34:34 UTC

Folks, and those on other threads asking similar questions:

There has been a bit of silence on this for a while for several reasons.

(1) The labor shortage behind the scenes is getting more intense.

(2) There have been several scientific conferences lately, and many people are living in airports.

(3) Most importantly, we are trying to figure out various progress metrics for various audiences and timescales. The best sign of progress will be a technical paper for scientists accompanied by press releases for the public, but that's a while off. A shorter-term thing would be the "End of May" report, which will be like an informal conference paper; and you've probably figured out from the name that it's more work than we thought. Things like the "percentage of data analyzed" are trickier than they look - apparently the "Server Status" numbers that I pointed out in another thread don't necessarily translate to that, because due to server constraints a lot of things are done on the fly.

I don't really have a good answer yet, but I wanted to say that we haven't forgotten about this and we are paying attention to your suggestions. Can't promise on any of them because what is doable does not always map onto what is desirable, but more things are coming.

Hope this helps,
Ben
____________

Profile Czar Brent
Avatar
Send message
Joined: 20 Feb 05
Posts: 26
Credit: 860,674
RAC: 0
Message 15486 - Posted: 14 Jul 2005, 7:42:29 UTC - in response to Message 15266.


(1) The labor shortage behind the scenes is getting more intense.

Things like the "percentage of data analyzed" are trickier than they look - apparently the "Server Status" numbers that I pointed out in another thread don't necessarily translate to that, because due to server constraints a lot of things are done on the fly.


If you need processor time to create the results I think I can spare some :)
____________
<p>WARNING! <font>D</font>i<font>H</font>ydrogen
<font>M</font>on<font>O</font>xide
kills!</p>

Message boards : Science : Progress metric


Home · Your account · Message boards

This material is based upon work supported by the National Science Foundation (NSF) under Grants PHY-1104902, PHY-1104617 and PHY-1105572 and by the Max Planck Gesellschaft (MPG). Any opinions, findings, and conclusions or recommendations expressed in this material are those of the investigators and do not necessarily reflect the views of the NSF or the MPG.

Copyright © 2016 Bruce Allen