Advanced search

Forums : General Topics : What's the status of the project?
Message board moderation

To post messages, you must log in.

1 · 2 · Next

AuthorMessage
Ryan Kipp

Send message
Joined: 12 Apr 09
Posts: 8
Credit: 292,500
RAC: 0
Message 8506 - Posted: 4 Aug 2009, 4:26:29 UTC

How far along is cosmology@home in terms of its goals? I took a look at server status, but I don't think I quite understood it...
ID: 8506 · Report as offensive     Reply Quote
Rapture
Avatar

Send message
Joined: 27 Oct 07
Posts: 85
Credit: 661,330
RAC: 0
Message 8519 - Posted: 12 Aug 2009, 21:55:59 UTC - in response to Message 8506.  

The goal of this project is to find the data that best fits with current cosmology models. Since we do not know what the results with be, we can not predict how long this will take.

This is also an exciting time in cosmology! In fact, this project is an official partner with the ongoing Planck mission which will attempt to gather more accurate data on cosmological parameters regarding the early universe.

Cosmology@home is my favorite project!


Be cool and have fun here! :)
ID: 8519 · Report as offensive     Reply Quote
zpm

Send message
Joined: 13 Mar 09
Posts: 3
Credit: 8,400
RAC: 0
Message 8521 - Posted: 14 Aug 2009, 17:51:54 UTC - in response to Message 8519.  

it would be a project i like more if it was "more popular; more active" no offense...
ID: 8521 · Report as offensive     Reply Quote
Profile Dingo
Avatar

Send message
Joined: 28 Jun 07
Posts: 9
Credit: 1,290,234
RAC: 0
Message 8524 - Posted: 16 Aug 2009, 19:52:11 UTC

So is the project dead ???


ID: 8524 · Report as offensive     Reply Quote
Brian Silvers

Send message
Joined: 11 Dec 07
Posts: 420
Credit: 270,580
RAC: 0
Message 8528 - Posted: 17 Aug 2009, 2:21:17 UTC - in response to Message 8524.  

So is the project dead ???


No, the project is very much alive, it's just the project management is very much dead. The Planck spacecraft is at the Sun-Earth L2 Lagrangian point in our solar system and is doing the research it was sent to do. Not exactly sure what data we are processing now... Hopefully it is something that Planck is sending back...


ID: 8528 · Report as offensive     Reply Quote
Rapture
Avatar

Send message
Joined: 27 Oct 07
Posts: 85
Credit: 661,330
RAC: 0
Message 8532 - Posted: 18 Aug 2009, 0:11:15 UTC - in response to Message 8528.  

The project management is not dead even though there has been no project news posted here in awhile.

Ben Wandelt has told me privately a few months ago that there is a new infusion of funding - this project is now officially part of the outreach program for the Planck mission. So project management will bring some important and long overdue improvements to the project soon and Ben plans to start a couple of new research initiatives as well.

As you can see, I am excited about what the future holds for this project! As we wait for more news from Ben here, I continue to crunch happily! :)
ID: 8532 · Report as offensive     Reply Quote
Brian Silvers

Send message
Joined: 11 Dec 07
Posts: 420
Credit: 270,580
RAC: 0
Message 8534 - Posted: 18 Aug 2009, 7:14:19 UTC - in response to Message 8532.  

The project management is not dead even though there has been no project news posted here in awhile.


You and I differ on things, obviously.

Anshul said that he would work on getting things straightened out here after May 17th. The only posts from him since then contained a dump of the process information from the server to say that everything was up and running, and a few posts about how Ben wanted the memory consumption up as high as it is, and possibly higher. Beyond that, nothing from him...

As for Ben, he posts updates here and there about Planck, as well as confirmed that the memory requirements are going to stay as they are or even increase.

Ben clearly does not understand that a very large portion of the donated resources are still Pentium 4 / Athlon XP / Athlon 64 class systems that range between 512MB and 1GB of memory. Those machines have either been pushed out of participation (512MB) or are barely able to participate due to major performance impacts if the system is being used for anything else (1GB systems). You also have those of us with 2GB who just participate in other projects because the 800MB or so of memory is even difficult for us as well if we are doing anything major on our systems. As good as Core2 and higher may be, Ben would be well served by not seeming to be so aloof in regards to just how demanding it is on systems to run this project. The issue that we all share with getting download errors (file not found) is because people are bailing out after they see how long these tasks are going to take and how much of a performance drain it is on their systems.


Ben Wandelt has told me privately a few months ago that there is a new infusion of funding - this project is now officially part of the outreach program for the Planck mission. So project management will bring some important and long overdue improvements to the project soon and Ben plans to start a couple of new research initiatives as well.


Yes, those would be the things that would take even more memory and run even longer. It is the wrong path to go down. Instead effort should be made into seeing if the tasks could be done on a GPU.


As you can see, I am excited about what the future holds for this project! As we wait for more news from Ben here, I continue to crunch happily! :)


I'd be excited if they'd show that they can keep their word and work on some things to fix some really silly problems that have been around for well over a year now. I'd also like to know what the 2 tasks I'm processing now is doing for "the science" of the project. Is the data something that Planck sent back, or is it just a simulation? Is the 40-60 hours of time on my system actually "worth it" scientifically?

ID: 8534 · Report as offensive     Reply Quote
Rapture
Avatar

Send message
Joined: 27 Oct 07
Posts: 85
Credit: 661,330
RAC: 0
Message 8539 - Posted: 18 Aug 2009, 21:24:43 UTC - in response to Message 8534.  

As I recall, Ben said that the current workunits are doing real science. We are not actually crunching data from the Planck mission but rather to test our results here with observational data from Planck.
ID: 8539 · Report as offensive     Reply Quote
steveplanetary

Send message
Joined: 19 May 09
Posts: 18
Credit: 24,360
RAC: 0
Message 8540 - Posted: 21 Aug 2009, 1:43:43 UTC - in response to Message 8519.  

It sounds like you feel the same as I do about cosmology, Rapture. It is exiting! But speeking as a scientist (and I mean no offense) I would say that the goal of this project is to find the cosmology model that best fits the data.

Cheers!
Steve
ID: 8540 · Report as offensive     Reply Quote
steveplanetary

Send message
Joined: 19 May 09
Posts: 18
Credit: 24,360
RAC: 0
Message 8541 - Posted: 21 Aug 2009, 1:47:20 UTC - in response to Message 8540.  

Sorry about the misspellings...I need new glasses.

Steve
ID: 8541 · Report as offensive     Reply Quote
Emanuel

Send message
Joined: 28 Oct 07
Posts: 31
Credit: 316,100
RAC: 0
Message 8542 - Posted: 21 Aug 2009, 8:50:27 UTC - in response to Message 8541.  
Last modified: 21 Aug 2009, 8:55:34 UTC

Steve, I think a lot of us are excited about the goals of this project and the chance to be involved in reaching them. That's why it's all the more frustrating when management doesn't allow us to get involved at all. I'm not saying they have to go all the way, i.e. open source and all that, but it would be such a help to at least get periodic status updates, and maybe get some more technical information about why exactly the WUs can't be split up any more without harming the scientific value of the project.

In my opinion the whole point of running projects on BOINC is so you can get as many computers as possible running your project - but I firmly believe they are scaring off at least two thirds of potential users by making the requirements this high - it'll get worse if they increase them even more. I think it would be wise to invest in some server-side code to recombine subsets of finished calculations so complexity can be increased at will without increasing the size of WUs. Computers might get a little less work done that way because of the overhead involved in WU turnaround time, but I think they'd attract enough new users that the overall result would be better for all of us. Not to mention all the users who currently have to turn off BOINC when they want to do a bit of gaming because it makes their computers too sluggish. (considering most games only use one CPU core, that's up to 75% of CPU time wasted during many hours of the day)

I don't personally stand to gain from this as I recently upgraded my computer to 8GiB of RAM which allows me to run Cosmo almost 24/7 - my frustration is purely for the sake of the project.
ID: 8542 · Report as offensive     Reply Quote
Profile Ivor Cogdell

Send message
Joined: 19 Jan 10
Posts: 3
Credit: 420
RAC: 0
Message 8861 - Posted: 11 Feb 2010, 0:58:38 UTC

Hi folks,
As a newcomer to the project, I find the lack of information from the "Science big-wigs" off-putting. I get the impression that those that do post are trying to cope under a great deal of pressure, but are not getting support that they need.
The News section needs more information - even if its just a general situation report. If there are problems, say so. At least we would understand what was going on.

Good project, I hope it does not fold through apathy at management level.

Ivor
ID: 8861 · Report as offensive     Reply Quote
Profile Benjamin Wandelt
Volunteer moderator
Project administrator
Project scientist
Avatar

Send message
Joined: 24 Jun 07
Posts: 188
Credit: 15,273
RAC: 0
Message 8862 - Posted: 11 Feb 2010, 1:38:57 UTC

Hi guys -

Thank you so much for your thoughts. It is clear that you all want this project to succeed as much as I do.

I admit that we have been somewhat scarce. In fact we have talked about shutting the project down for a while until we find someone we can hire to help us maintain and develop the project at the level which it and you deserve.

There are currently several issues:

1) The mysterious intermittent download errors.

2) Many work units are long and memory consuming. There is a great deal of variation between work units.

3) The project has been under attack by spam robots, and our only practicable way of removing offensive material from the website is to remove those accounts affected.

4) There may be additional underlying problems I am not aware of which we need to find and fix.

The good news is that the work units you are crunching are all good units, which are used in accelerating cosmological parameter analysis from Planck data and other data which are analyzed jointly.


Scott, who did a marvelous job in getting everything set up has embarked on his own research program which simply does not leave him enough time to commit to C@H.

I had my first child in August and almost simultaneously had an accident which left me without the use of my arms for about 2 months. I have since recovered almost 100%, but these two events obviously took a great deal of my attention.

In addition, I accepted a new position in November as Professor and International Chair at the Paris Institut for Astrophysics (IAP) and Pierre and Marie Curie University in Paris (Paris 6) which began in January. I am still not done with the transatlantic move. Most of my family's worldly belongings are currently somewhere on the Atlantic Ocean.

I continue to hold an adjunct appointment at the University of Illinois to complete the research projects I started there. So Cosmology@Home continues and once the hire will go forward it will hopefully flourish, possibly branching out into new areas of cosmology - I have described some of my ideas along those lines elsewhere in these forums.

The reason the hire has not happened yet, is because it is difficult to find someone with the required skill set given our budgetary constraints. We do have a yearly budget but it would only pay for a fraction of a professional computer engineer. The goal is to find someone like Scott - a talented computer engineering or computer science student with a passion for astronomy/cosmology who can sink his/her teeth into this project.

As some of you have noted our last attempt to do so did not work out and possibly had a net negative effect. So we have been looking and have not made a hire for a long time since then. One would think that with the recession and the excitement we all have about the potential for cosmology@home there would be a line around the block for this position and the opportunity it represents for some students CV but that has not happened.

There it is. The project is alive, and productive, though not well. As I redirect more of my energy towards it I am hopeful that we can return it to its former glory and beyond.

Best regards,

Ben

Creator of Cosmology@Home
ID: 8862 · Report as offensive     Reply Quote
Emanuel

Send message
Joined: 28 Oct 07
Posts: 31
Credit: 316,100
RAC: 0
Message 8863 - Posted: 11 Feb 2010, 7:45:59 UTC - in response to Message 8862.  

Thank you for the update, Ben. Considering the realities of your budget, have you considered open sourcing the code? You would need tests to validate new applications (of course) and make sure to only accept versions you've personally checked out, but the core development could be done by enthousiasts. I'm not saying they will, though - that's always the danger of open source - but if you're willing to deal with the risks, the benefit -could- be great: I myself am a former CS student (now a student of interdisciplinary sciences) with a strong interest in fundamental physics and cosmology, but I doubt I have the experience to contribute significantly at this time. As my own education continues, that might change.

If you're interested, the first choice you have to make in making your code open source is choosing a license. This is a tough decision that deserves to be looked at carefully, not least because you have to do your best to get the permission of everyone involved in writing the code in the past. The legal validity of licenses has only rarely been tested, but on the internet, everything is based on trust, and violating it is something to be avoided at all costs. This is also, of course, why this project badly needs more public outreach and communication.

I'm looking forward to hearing your take on this. Good luck in your current endeavours (including fatherhood!), and I hope your arms recover fully :)
ID: 8863 · Report as offensive     Reply Quote
Brian Silvers

Send message
Joined: 11 Dec 07
Posts: 420
Credit: 270,580
RAC: 0
Message 8864 - Posted: 11 Feb 2010, 12:01:30 UTC - in response to Message 8862.  
Last modified: 11 Feb 2010, 12:05:08 UTC


1) The mysterious intermittent download errors.


I've been mentioning for months that this happens when people reset the project because of seeing your number 2 item:


2) Many work units are long and memory consuming. There is a great deal of variation between work units.


What seems to being going on is you're not keeping the task input files on the download server until the result comes back as a success, or not restoring them when a timeout happens. All you likely need to do is ask other project administrators what the proper settings should be.

In addition, I accepted a new position in November as Professor and International Chair at the Paris Institut for Astrophysics (IAP) and Pierre and Marie Curie University in Paris (Paris 6) which began in January. I am still not done with the transatlantic move. Most of my family's worldly belongings are currently somewhere on the Atlantic Ocean.



While I do congratulate you, the proper thing to have done would've been to come here back in November and mention that you were going to be doing moves. I am sure you would expect a student of yours to provide you with a reasonable explanation as to why they went missing if that student then showed up asking for extensions on projects / papers. While that is a reasonable explanation, the time missing was known, not an unknown, so I don't know how you would make a judgement call on a situation with a student coming to you like this, but I'd only give partial credit...


The reason the hire has not happened yet, is because it is difficult to find someone with the required skill set given our budgetary constraints. We do have a yearly budget but it would only pay for a fraction of a professional computer engineer. The goal is to find someone like Scott - a talented computer engineering or computer science student with a passion for astronomy/cosmology who can sink his/her teeth into this project.


I've mentioned elsewhere that I think you could sweeten the deal by going up to $20/hr from $15. I could probably do it if I was there, but as you said, I'd be wanting much more than $20/hr (likely in the 60-70k salary range). $15/hr is too low for all that you are asking for...

-Brian
ID: 8864 · Report as offensive     Reply Quote
Profile Ivor Cogdell

Send message
Joined: 19 Jan 10
Posts: 3
Credit: 420
RAC: 0
Message 8866 - Posted: 12 Feb 2010, 2:44:48 UTC - in response to Message 8862.  

Many thanks for the information. I'm sure we all wish you a speedy recovery and that you can get your life back on track in short order.
ID: 8866 · Report as offensive     Reply Quote
Profile Ivor Cogdell

Send message
Joined: 19 Jan 10
Posts: 3
Credit: 420
RAC: 0
Message 8867 - Posted: 12 Feb 2010, 3:00:50 UTC - in response to Message 8528.  

Hey folks,
Here is a link to the ESA Project Website.


http://www.esa.int/esaSC/120398_index_0_m.html

ID: 8867 · Report as offensive     Reply Quote
Rapture
Avatar

Send message
Joined: 27 Oct 07
Posts: 85
Credit: 661,330
RAC: 0
Message 8868 - Posted: 12 Feb 2010, 13:51:21 UTC - in response to Message 8862.  

Hi, Ben! Glad to hear from you in a long time! Thanks for the detailed update! Despite the recent setbacks, I am still pleased with the way this project is going. Even though there is room for improvements, I still continue to be active here. I look forward to bigger and better things with this great project! If you decide to shut down this project temporarily, please let us know ahead of time so we can be prepared.
ID: 8868 · Report as offensive     Reply Quote
Profile Saenger
Volunteer tester
Avatar

Send message
Joined: 22 May 07
Posts: 110
Credit: 282,157
RAC: 1
Message 8875 - Posted: 14 Feb 2010, 11:29:17 UTC - in response to Message 8862.  
Last modified: 14 Feb 2010, 11:30:09 UTC

3) The project has been under attack by spam robots, and our only practicable way of removing offensive material from the website is to remove those accounts affected.

Unfortunately this seems to have been done in a very unprofessional manner as you can see in this threads:
deleted Account
URGENT PROBLEMS THREAD (2009 and after)

At least 3 accounts of harmless users have been deleted without any reason, at least without any thinkable reason. And nothing has been done to undo this severe error within the last two weeks.
Grüße vom Sänger
ID: 8875 · Report as offensive     Reply Quote
Brian Silvers

Send message
Joined: 11 Dec 07
Posts: 420
Credit: 270,580
RAC: 0
Message 8876 - Posted: 15 Feb 2010, 12:46:22 UTC - in response to Message 8875.  
Last modified: 15 Feb 2010, 12:48:09 UTC


At least 3 accounts of harmless users have been deleted without any reason, at least without any thinkable reason. And nothing has been done to undo this severe error within the last two weeks.


Using past history as a guide for future events, I'd plan on a longer wait... You might try to find Ben's phone number and call it...
ID: 8876 · Report as offensive     Reply Quote
1 · 2 · Next

Forums : General Topics : What's the status of the project?