The Economist is an English-language weekly news and international affairs publication owned by "The Economist Newspaper Ltd" and edited in London. It has been in continuous publication since James Wilson established it in September 1843. As at summer 2007, its average circulation tops 1.2 million copies a week, about half of which are sold in North America.[1] Consequently it is often seen as a transatlantic (as opposed to solely British) news source. (Wikipedia)

Here's the linky

Very interesting to note that Folding@Home and Sony's PS/3 get quite a bit of ink space, and Rosie gets, well, actually, none.

"Another development that is boosting volunteer computing is the use of devices other than PCs, in particular games consoles and the powerful processors they contain (see article). This has been demonstrated most spectacularly by a project called Folding@home, run by Vijay Pande and his team at Stanford University, which simulates protein folding and mis-folding—a cause of diseases such as Alzheimer's. In September the combined computing capacity of the project passed one petaflop—a quadrillion mathematical operations per second—something supercomputer designers have dreamed of for several years. With just over 40,000 PlayStation 3 volunteers, Folding@home entered the record books as the most powerful distributed-computing network on Earth."

The Economist is an English-language weekly news and international affairs publication owned by "The Economist Newspaper Ltd" and edited in London. It has been in continuous publication since James Wilson established it in September 1843. As at summer 2007, its average circulation tops 1.2 million copies a week, about half of which are sold in North America.[1] Consequently it is often seen as a transatlantic (as opposed to solely British) news source. (Wikipedia)

Here's the linky

Very interesting to note that Folding@Home and Sony's PS/3 get quite a bit of ink space, and Rosie gets, well, actually, none.

"Another development that is boosting volunteer computing is the use of devices other than PCs, in particular games consoles and the powerful processors they contain (see article). This has been demonstrated most spectacularly by a project called Folding@home, run by Vijay Pande and his team at Stanford University, which simulates protein folding and mis-folding—a cause of diseases such as Alzheimer's. In September the combined computing capacity of the project passed one petaflop—a quadrillion mathematical operations per second—something supercomputer designers have dreamed of for several years. With just over 40,000 PlayStation 3 volunteers, Folding@home entered the record books as the most powerful distributed-computing network on Earth."

R@H is at an interesting point because it clearly lacks the support required to compete with some of the other projects. When looking through the posts, it is clear that R@H does NOT support some of Intel's extended instructions (SSE1 - SSE3) and adding that support could provide an improvement in performance for all of the intel based crunchers. In addition, none of the optimized clients accelerate R@H because the project team will not release the source code. Finally, the project team decided NOT to pursue additional platforms like the PS3 with cell processors.

R@H could be a very powerful application if the project team would invest the resources to optimize the code for Intel and AMD processors and if they would port the code to other platforms like the PS3. The easiest solution is to publish the existing code and let someone else do the work. The smartest path would be to ask for the resources required to make these changes and get some press for R@H at the same time.

Getting more people to run R@H is an issue we can address. Efficiently using the resources we recruit and donate is an issue only the project team can address.

---

Thx!

Paul

...the project team will not release the source code.

Have you completed the non-disclosure agreement and requested a copy of the source for such purposes and been refused??

...the project team decided NOT to pursue additional platforms like the PS3 with cell processors.

Please document your source of that information.

---

Rosetta Moderator: Mod.Sense

Paul: You're preachin' to the choir !!!


Mod.Sense: You respond to Paul, yet ignore all my posts on the topic. Hmmm.

Even with my quadruple-digit RAC falling to double-digits, I’m still in the top 2% of Contributors to Rosie.

Just how may credits must one earn before they are accorded the professional courtesy of a response from the Project Team?

Perhaps there is some super-double-top-secret project to port Rosie code to the Cell BE / PS3, but where is the harm in answering some of the "general" questions I've been (repeatedly) asking?

Do you really want me to cite all my posts, which were met with deafening silence from the Project Team, lol ?!

We can start with this post to begin our discussion.

There’s no harm in having a discussion, is there?

And if the Project Team gets tired of answering this type of question over and over, how about putting a moderately detailed answer in the FAQ ?

So, let’s start with:


(3) "Main memory size. The PlayStation 3 is equipped with only 256 MB of main memory. This represents a serious limitation when combined with the slowness of the network interconnect."

Again, initial goal is not for clusters of PS/3'a via network interconnect, but a single PS/3.

That being said, the size of main memory IS a limitation, and it may in fact be fatal to any efforts to port Rosie over to the PS/3. However, I have yet to hear anyone officially affiliated with the Project make a definitive statement that this is a fact.


AND


(5) "Programming paradigm. One of the most attractive features of the CELL processor is its simple architecture and the fact that all its functionalities can be fully controlled by the programmer. In most other common processors performance can only be obtained with a good exploitation of cache memories whose behavior can be controlled by the programmer only indirectly."

This is likely a legitimate "limitation" for Rosie.

Unless there are additional "limitations" of the new 65nm Cell BE in a PS/3, then the out of the five "limitations" listed in the paper, only two are potentially applicable to Rosie.

If the memory footprint required by Rosie in greater than what the PS/3-CBEA can provide, there is little that can be done.

Alternatively, if Rosie can fit within the PS/3-CBEA memory footprint, then the "only" remaining limitation is the programming.

Note the quotes around "only".

I am aware of the effort that would be involved.

From Wikipedia:

In November 2006, David A. Bader at Georgia Tech was selected by Sony, Toshiba, and IBM from more than a dozen universities to direct the first STI Center of Competence for the Cell Processor. This partnership is designed to build a community of programmers and broaden industry support for the Cell processor. There is a Cell Programming tutorial video available.

All I am suggesting is that at some point in the Cost Benefit Analysis, there H-A-S to be a point where the expenditure of resources is justified in porting Rosie to PS/3.

What is the Project's answer to this question? 500 tflops? 1 pflop? 25 pflops?

I can't believe the Project would say the expenditure of resources could not be justified for a yottaflop !!!



AND DON'T FORGET THIS OTHER POST



For example, how was this possible?

"When David Baker, who also serves as a principal investigator for Howard Hughes Medical Institute, originally developed the code, it had to be run in serial - broken into manageable amounts of data, with each portion calculated in series, one after another.

Through a research collaboration, SDSC's expertise and supercomputing resources helped modify the Rosetta code to run in parallel on SDSC's massive supercomputers, dramatically speeding processing, and providing a testing ground for running the code on the world's fastest non-classified computer.

The groundbreaking demonstration, part of the biennial Critical Assessment of Structure Prediction (CASP) competition, used UW professor David Baker's Rosetta Code and ran on more than 40.000 central processing units (CPUs) of IBM's Blue Gene Watson Supercomputer, using the experience gained on the Blue Gene Data system installed at SDSC."




And why wouldn't such a strategy work on gaming consoles (PS/3, xBox360, Wii) ???

...the project team decided NOT to pursue additional platforms like the PS3 with cell processors.

Please document your source of that information.

---

Defeat Censorship! Wikileaks needs OUR help! Learn how you can help (d/l 'insurance' file), by clicking here. "Whoever would overthrow the liberty of a nation must begin by subduing the freeness of speech" B. Franklin

And why wouldn't such a strategy work on gaming consoles (PS/3, xBox360, Wii) ???

well there is a really simple reason for it, at least one that i can think off.
when im in NY with a dial in connection, and almost all the others are on dsl lines all over the world, then i'm a problem, since i also have to get all information, and send all my information. inside 1 system that is doable cause the cables are short and highspeed, but over longdistance with crappy cabels. this aint gonna work.

Penguin, my response, or lack thereof, has nothing to do with RAC, total credit, percentile on contributors list, etc. Simply a matter of you asking pointed and direct questions that I do not have answers to, nor comments to offer, and so I do not reply. I cannot offer an official response to your questions, and I have not responded to any of Paul's concerns either.

But Paul seems to have made statements speaking for the project team, and saying he is reporting decisions they have made. I'm not aware of any such decisions, so I'm asking he point me to the facts he seems to be siteing.

I am not involved in Rosetta's cost/benefit analysis and prioritization on development time, and so I will not offer any comments on those topics. The thread is here (and elsewhere as you point out) if anyone in such a position wishes to offer their comments.

---

Rosetta Moderator: Mod.Sense

Sorry, I am not following the concern you are trying to express.

If Rosetta code (which is currently sequential in nature) can be parallelized (as was done for its run on IBM supercomputers), then why can't the code be ported to run on the PS/3's six SPE's?

What does that have to do with dial-up connections versus dsl connections?

My understanding is that the issue of parallelization is NOT to link all contributing PS/3's into a single "supercomputer", BUT RATHER have all the individual PS/3's run parallel code on the six SPE's. THEN cumulatively, the PS/3's will provide the equivalent power of a supercomputer, ala F@H's 1 pflop.

And why wouldn't such a strategy work on gaming consoles (PS/3, xBox360, Wii) ???

well there is a really simple reason for it, at least one that i can think off.
when im in NY with a dial in connection, and almost all the others are on dsl lines all over the world, then i'm a problem, since i also have to get all information, and send all my information. inside 1 system that is doable cause the cables are short and highspeed, but over longdistance with crappy cabels. this aint gonna work.

Huh. Could swear that it says "Project Administrator" under your nickname. Is that meaningless fluff? Or just there so you can moderate AND ban people - the judge, jury and executioner thing.

Seems like a "Project Administrator" should be more in tune with what the project is (and isn't) doing.

Penguin, my response, or lack thereof, has nothing to do with RAC, total credit, percentile on contributors list, etc. Simply a matter of you asking pointed and direct questions that I do not have answers to, nor comments to offer, and so I do not reply. I cannot offer an official response to your questions, and I have not responded to any of Paul's concerns either.

But Paul seems to have made statements speaking for the project team, and saying he is reporting decisions they have made. I'm not aware of any such decisions, so I'm asking he point me to the facts he seems to be siteing.

I am not involved in Rosetta's cost/benefit analysis and prioritization on development time, and so I will not offer any comments on those topics. The thread is here (and elsewhere as you point out) if anyone in such a position wishes to offer their comments.

---

Proudly Banned from Predictator@Home and now Cosmology@home as well. Added SETI to the list today. Temporary ban only - so need to work harder :)



"You can't fix stupid" (Ron White)

Huh. Could swear that it says "Project Administrator" under your nickname. Is that meaningless fluff? Or just there so you can moderate AND ban people - the judge, jury and executioner thing.

Seems like a "Project Administrator" should be more in tune with what the project is (and isn't) doing.

BOINC defined the term, not Rosetta. Users expressed concern about the potential to hit the red X on an abusive post and have fairly immediate action be considered to remove it. When my ID was configured to receive the EMail generated when anyone click the abusive post indication, I became a "Project Administrator". Something like "Message Board Administrator" may have been more descriptive.

Perhaps now some of my posts about what I can and cannot offer comments on, and how I cannot speak for the Project Team nor access the project servers and databases make more sense. I'm just a volunteer cruncher with a few extra buttons available when I view the message boards.

---

Rosetta Moderator: Mod.Sense

Huh. Could swear that it says "Project Administrator" under your nickname. Is that meaningless fluff? Or just there so you can moderate AND ban people - the judge, jury and executioner thing.

Seems like a "Project Administrator" should be more in tune with what the project is (and isn't) doing.

BOINC defined the term, not Rosetta. Users expressed concern about the potential to hit the red X on an abusive post and have fairly immediate action be considered to remove it. When my ID was configured to receive the EMail generated when anyone click the abusive post indication, I became a "Project Administrator". Something like "Message Board Administrator" may have been more descriptive.

Perhaps now some of my posts about what I can and cannot offer comments on, and how I cannot speak for the Project Team nor access the project servers and databases make more sense. I'm just a volunteer cruncher with a few extra buttons available when I view the message boards.

It does make more sense now........maybe they should just lose the "project administrator" title and leave the "forum moderator."

Huh. Could swear that it says "Project Administrator" under your nickname. Is that meaningless fluff? Or just there so you can moderate AND ban people - the judge, jury and executioner thing.

Seems like a "Project Administrator" should be more in tune with what the project is (and isn't) doing.

BOINC defined the term, not Rosetta. Users expressed concern about the potential to hit the red X on an abusive post and have fairly immediate action be considered to remove it. When my ID was configured to receive the EMail generated when anyone click the abusive post indication, I became a "Project Administrator". Something like "Message Board Administrator" may have been more descriptive.

Perhaps now some of my posts about what I can and cannot offer comments on, and how I cannot speak for the Project Team nor access the project servers and databases make more sense. I'm just a volunteer cruncher with a few extra buttons available when I view the message boards.

It does make more sense now........maybe they should just lose the "project administrator" title and leave the "forum moderator."

have to agree on that one. it is more logical and in line with your duties.
you moderate/monitor the boards and have nothing to do with the actual "project" side of things. So they should remove or modify your project worded title to reflect your true duties.

---

Well, that's what I'm saying, it is BOINC that worded it that way, not Rosetta. Not that it would be impossible to change, but you don't want to add anything to your list of items you "must change upon every BOINC release". Often times the contact that people are intending to direct questions to would be given the "Project Scientist" or "Project Developer" titles. So anyway, that's what the titles mean.

---

Rosetta Moderator: Mod.Sense

...the project team will not release the source code.

Have you completed the non-disclosure agreement and requested a copy of the source for such purposes and been refused??

...the project team decided NOT to pursue additional platforms like the PS3 with cell processors.

Please document your source of that information.

Thank you for the response.

I don't want to get into an argument here so I will first concede that I have not requested the code nor have I looked at the non-disclosure agreement.

I have looked for optimized versions for R@H and found nothing. I have also noticed posts from others requesting optimized clients and they have received nothing. Perhaps you could point us in the right direction.

The lack of a PS3 client is the only evidence we have that the project team has decided not to pursue this platform. The PS3 has been available for over a year. Several threads have requests for such a client with no moderator response. Again, perhaps you can point us to the client.

I donate lots of CPU time to R@H and I believe in the science. I want to help find a cure so let's work together to find ways to use every CPU as efficiently as possible.

thx

Paul

Penguin, my response, or lack thereof, has nothing to do with RAC, total credit, percentile on contributors list, etc. Simply a matter of you asking pointed and direct questions that I do not have answers to, nor comments to offer, and so I do not reply. I cannot offer an official response to your questions, and I have not responded to any of Paul's concerns either.

But Paul seems to have made statements speaking for the project team, and saying he is reporting decisions they have made. I'm not aware of any such decisions, so I'm asking he point me to the facts he seems to be siteing.

I am not involved in Rosetta's cost/benefit analysis and prioritization on development time, and so I will not offer any comments on those topics. The thread is here (and elsewhere as you point out) if anyone in such a position wishes to offer their comments.

I apologize for any miscommunication. I am not part of the project team, I only donate CPU time to the project.

I am very disappointed in the fact that we do not have optimized versions of R@H for intel and AMD processors as the project would likely benefit from the efforts. I am also disappointed in the fact that no client exists for the PS3 or for Graphics Processors as they clearly represent a large number of idle CPU cycles.

Please keep crunching R@H.

---

Thx!

Paul

Agreed. I just purchased two ATI x1950pro gpu's, and will have my first (of many) PS/3's by year's end.

Would love to donate to Rosie, especially with CASP in the not-so-distant future.

But I guess all Rosie wants is non-optimized cpu cycles...

You all know the old saying: Waste not, want not !!!

I am very disappointed in the fact that we do not have optimized versions of R@H for intel and AMD processors as the project would likely benefit from the efforts. I am also disappointed in the fact that no client exists for the PS3 or for Graphics Processors as they clearly represent a large number of idle CPU cycles.

Please keep crunching R@H.

I would say you start by finding someone (or a team) able to make the porting, compilation, optimization and/or changes you are suggesting. It's one thing to post "you should...", "you should...", "how hard could it be?" and try to influence someone else's resource decisions. It is quite another to say "We will...", "We can..." and offer specific skills and experience to help the project.

Have them review the information on the Rosetta license. On that page is a contact EMail address, and a button at the top for "license".

Since your apparent intent would be to redistribute your resulting optimized program, it would appear you would have to further inquire about how that might be permissible beyond the standard agreement. You might send an EMail to that "scientific contact" EMail address and explain the resources you have available, state your intentions and ask how it might be possible. ...or if such things have already been investigated.

You might also post a follow-up message to this post from Dr. Baker and ask if there have been further developments with Mr. Hey since the time of that post.

...as for "Waste not, want not !!!"... the issue of whether simple compiler optimizations for processor-specific features would make any noticible difference has been discussed at considerable length previously. As has the question of whether the individual SPEs on a Cell processor would be capable of running something as large as Rosetta. I don't wish to further divert this thread, but suffice it to say that the question of whether the effort would be expected to bare fruit for improving Rosetta is an open question. My summary of the prior discussions would be that the more technical knowledge and experience with the environments being discussed the people posting appeared to have, the more doubt they seemed to have about the likely outcome and the scale of the undertaking.

---

Rosetta Moderator: Mod.Sense

With all due respect, thats kinda silly.


Which seems to make more sense:

(1) Get an answer from the Project as to whether or not the limited ram or porting to parallelized code is fatal to running Rosetta on a PS/3. If Project answers "fatal", then stop. If Project answers "not fatal", then begin to assemble a team and request a NDA.

or

(2)

I would say you start by finding someone (or a team) able to make the porting, compilation, optimization and/or changes you are suggesting.

And then after assembling a team, sit and wait and hope the Project will then decide to bestow enlightenment.

A team which was created with much effort, may or may not be enlightened by the Project, or may be enlightened with the knowledge that is fatal to run Rosie on a PS3.



I vote for, and am still waiting on, # 1.

I agree with the thought that it is best not to put in work that could easily prove wasted in the end.

But I point out that the answer from the Project Team will likely be "don't know". So then you are back to looking for someone with experience on Cell/PS3 etc. to take a stab at the question. And they will need to spend considerable time reviewing the source code to make such a determination.

---

Rosetta Moderator: Mod.Sense