Index  | Recent Threads  | Unanswered Threads  | Who's Active  | Guidelines  | Search
 

Quick Go »
No member browsing this thread
Thread Status: Active
Thread Type: Sticky Thread
Total posts in this thread: 1840
Posts: 1840   Pages: 184   [ Previous Page | 22 23 24 25 26 27 28 29 30 31 | Next Page ]
[ Jump to Last Post ]
Post new Thread
Author
Previous Thread This topic has been viewed 1600885 times and has 1839 replies Next Thread
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Research Log: Updates from the Harvard Team

We had out monthly phone-conference with the IBM WCG team yesterday. CEP2 seems to be running without significant problems - that was before the server issue this morning biggrin.
[Mar 17, 2011 8:56:14 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Research Log: Updates from the Harvard Team

You never know Sgt Joe. right now, I have a project, I can either spend 20 hours on a CPU to grind out, or let it run for 48 minutes on a GPU to get done. A year or so ago that was not even an option. Look at where we are now compared to the 486 era? Now they are talking the possibility of a 1000 core or more processor. Lets go back ten years to the Pentium III era, how long would one of these projects take on one of those chips compared to now?

Never say never.

You might as well say the clean energy thing will never happen, because nobody has done it before, it's too far fetched, it's unrealistic.

Mention terabyte drives 5 years ago and you probably would have been laughed at.

Just some food for thought.

Aaron
----------------------------------------
[Edit 1 times, last edit by Former Member at Mar 18, 2011 2:28:45 AM]
[Mar 18, 2011 2:28:11 AM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Research Log: Updates from the Harvard Team

Viktors from the IBM WCG team will visited us in Cambridge next week.
[Mar 18, 2011 4:29:43 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Research Log: Updates from the Harvard Team

The usual scientific controversies...
http://xkcd.com/867/
[Mar 19, 2011 3:38:18 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Research Log: Updates from the Harvard Team

It was pointed out to us that the linked comic contains crude language and we apologize. The content is funny though, in particular when one knows real life clashes along those lines biggrin.
[Mar 21, 2011 4:02:16 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Research Log: Updates from the Harvard Team

We had mentioned a while ago that we were planning to build our own 90TB storage array 'Jabba'. This effort was spearheaded by Roberto and the built actually turned into an interesting exercise. We are now reasonably confident that Jabba works, tests so far have been successful, and we’ll commission it in the next few days. We’ll soon post a series of pix of the construction stages. Here is the design we followed (courtesy of Backblaze).
http://blog.backblaze.com/2009/09/01/petabyte...uild-cheap-cloud-storage/
[Mar 21, 2011 4:05:08 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Research Log: Updates from the Harvard Team

Will this thing work with different drive sizes in it, as long as you keep the 'chunks' the same? In other words, as drives get cheaper, and / or start failing, you can replace the 1.5tb with a 2 (well probably a few 2's to keep the raid physical attributes mirrored) and eventually up to 3, 4, 10 .. whatever we are up to in drive size at that time.

With luck you can get one or two of these things, and hopefully for a while just put bigger drives in as technology and your wallet allow you to. Or am I totally clueless on the abilities of the storage array?

Oh one other thing, when you roll out the new one, are you going to give us some beta's to test on it so those of us who are slackers can get a shot at some beta time towards our badges :D

Aaron
[Mar 21, 2011 8:50:44 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Research Log: Updates from the Harvard Team

A number of members of the Aspuru-Guzik group are traveling to Dallas for the American Physical Society (APS) March meeting. Alan and Johannes bumped into each other at the gate at Logan airport yesterday unaware that they’d be on the same flight... As the APS meeting is quite a big event, that flight was packed with physicists - like on a field-trip. http://www.aps.org/meetings/march/
[Mar 21, 2011 10:53:29 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Rickjb
Veteran Cruncher
Australia
Joined: Sep 17, 2006
Post Count: 666
Status: Offline
Project Badges:
Reply to this Post  Reply with Quote 
Re: Research Log: Updates from the Harvard Team

Will the Backblaze pods run CEP2, and have you managed to persuade Backblaze to donate the spare CPU cycles in their cloud(s) to you yet? wink
[Mar 22, 2011 1:49:44 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Former Member
Cruncher
Joined: May 22, 2018
Post Count: 0
Status: Offline
Reply to this Post  Reply with Quote 
Re: Research Log: Updates from the Harvard Team

A Backblaze pod is an array of 45 hard disks. It stores the results from CEP2.

Lawrence
[Mar 22, 2011 3:19:23 PM]   Link   Report threatening or abusive post: please login first  Go to top 
Posts: 1840   Pages: 184   [ Previous Page | 22 23 24 25 26 27 28 29 30 31 | Next Page ]
[ Jump to Last Post ]
Post new Thread