Posts by FoldingSolutions

21) Message boards : Number crunching : Intel i7 CPU (Message 57621)
Posted 5 Dec 2008 by FoldingSolutions
Post:
We will most likely see the L2 cache size problem addressed in Westmere. AnandTech published an article where Intel recognized that the current cache situation is not ideal. A larger L2 size on the current process is hard to justify because that would decrease the effective size of the L3 cache, due to its inclusiveness as Mastergee pointed out. A larger L2 would have to come with a larger L3, which can be made possible with the extra transistors in 32 nm.

What would be interesting is that if HT for Rosetta is limited due to the shared use of the FPU, then would there be less of a performance drop when running an integer-heavy BOINC project on the other logical core. If this were true, then the ideal case is for BOINC or the OS to set process affinity such that each physical CPU runs an ideal mix of work.


That is perfectly true. If one math unit is being used by one application and then the other by another application. Then HT is being utilised to near 100% effectivenes (Probably never actually as good as 2 physical cores, but nearly) HT is mostly useful in enterprise applications (Data mining, 3D rendering) where all units are being used. However in scientific apps like Rosetta or Folding@home, I'm not too sure if there is much use of the interger (ALU) unit as opposed to the FPU.
22) Message boards : Number crunching : Intel i7 CPU (Message 57571)
Posted 3 Dec 2008 by FoldingSolutions
Post:
8 children + cookies = no cookies


Funny!

My point was simply that "it depends". ...if the kitchen is large enough, it can work more efficiently. But even if floor space (L2 cache) is not the problem, other key resources such as mixing bowel (bus), and oven (math coprocessor) may become bottlenecks.


OK I see your point, also if there isn't enough flour (data) to go round then some of those resouces are wasted. And Hyperthreading can provide a little extra flour :)
23) Message boards : Number crunching : Intel i7 CPU (Message 57559)
Posted 3 Dec 2008 by FoldingSolutions
Post:
Just think of it as 2/3 + 2/3 = 4/3. So even if each "hyperthreaded" core is only 2/3 as powerful as a full core


You can also think of it as a 10 year old being 2/3rds of a baker... now picture 8 of them in your kitchen and what will happen to your cookies? Faster? or slower?

Either outcome is possible. Time will tell.


Probably not the best analogy to compare a child to a piece of silicon. I think Intel designed HT to improve efficiency, not to create more of a weaker version of the previous model. The maths of 2 + 2 = 4 is very different to 8 children + cookies = no cookies ;)
24) Message boards : Number crunching : Intel i7 CPU (Message 57557)
Posted 3 Dec 2008 by FoldingSolutions
Post:
As previously, HT might help other threads keep out of the way of Rosetta, but for rosetta throughput the critical factor will be whether the increased use of the L3 (with its latency penalty) will be compensated for by the interleaving of the threads under HT

I agree that HT will have to share the L2 cache and that this could be detrimental. However, as the L2 in Core 2 Duos can be as much as 3MB/core, i dont think that 128KB going amiss in an i7 is going to affect it too much. And the quicker access to main memory should even out any short-falls. But you could be right that disabling HT could improve performance as Rosetta mostly uses the FPU so two threads using the same unit could get messy. But I did hear somewhere that Intel had looked at this and that the new HT is different to the old P4 version. Time will tell :)
25) Message boards : Number crunching : Intel i7 CPU (Message 57545)
Posted 3 Dec 2008 by FoldingSolutions
Post:
The i7 processor has an inclusive L3 cache. Which for those who don't know what that means is basically a small portion (1MB) of the 8MB L3 cache is reserved for the contents of all 4 256KB L2 caches so that if core 1 wants something from core 3 then rather than diving straight into core 3's L2 cache it can simply go to the L3 which prevents a core from "stalling".
Hyper-threading will not improve the credits per hour of individual cores, but certainly will for the CPU as a whole. Just think of it as 2/3 + 2/3 = 4/3. So even if each "hyperthreaded" core is only 2/3 as powerful as a full core. These two add up to equal greater than the sum of its parts.
It is also true that the small amount of L2 in i7 is a concern for those applications which "cache thrash" such as Rosetta. But this effect will be minimised by the large L3 cache and the new Quickpath interconnect which much reduces latency to the main memory. So the 8 cores will be waiting less for data. The large L2 caches we have been seeing on some of the latest Core 2 CPU's are to make up for the extreme latency problems of the ancient & inadequate front side bus.
The TDP of all released Core i7 CPU's is 130W, fortunately this isn't what Rosetta will use as this figure is assuming the CPU is fed perfectly with data to process and that all instruction units (MMX SSE SSE2 etc) are being used. So I would expect the CPU itself to be knocking out more like ~90W
We also have to remember that unfortunately Intel doesn't design processors for people like us who see them as tools for science, but rather for large corporations who don't like to wait for access to massive data-bases of who's been taking too many days off etc. To this end we may have to wait for a while for Rosetta to benefit significantly from any major changes in CPU architecture. Also the advent of green computing could see less major improvements too :(
26) Message boards : Number crunching : Why does this still happen. (Message 56515)
Posted 29 Oct 2008 by FoldingSolutions
Post:
Task ID - 202592794
Work unit ID - 185058479
Sent - 27 Oct 2008 20:17:33 UTC
Time reported or deadline - 29 Oct 2008 19:33:36 UTC
Server state - Over
Outcome - Client error
Client state - Compute error
CPU time (sec) - 70,590.59
Claimed credit - 329.12
Granted credit - ---

Shouldn't there be some kind of credit compensation for 20 hours of wasted CPU time??
27) Message boards : Number crunching : BOINC in Java code (Message 56422)
Posted 19 Oct 2008 by FoldingSolutions
Post:
I have downloaded this application but cannot seem to install it on my phone, how do we do this? I have a Sony Ericsson W850i


The developers have re-released the code using CVS, you should look at the SF page for details.
The code, though, is written in Java SE, so it will need to be converted to Java ME to run on your mobile phone (I'm testing on a Sony Ericsson K610i), but I have not yet been successful.

I certainly haven't given up though, there's at least three Sony Ericsson phones here that could be crunching WUs. :-)

Only SETI has been ported to Java at the moment anyway, you would need to obtain the source code, and port it manually, for each product if it is available for download.

Remember: the code is pre-alpha, moving WUs around needs to be done manually also.


Well its a good start, hope you get somewhere with it :)
28) Message boards : Number crunching : Cheap dual-core build on NewEgg for R@H (Message 55751)
Posted 14 Sep 2008 by FoldingSolutions
Post:
A X2 2.8ghz AMD only has a 2mb processor cache, the E8400 has a 6 mb processor cache. The bigger the processor cache, the more credit you are granted, especially for work units that are cache sensitive, like the Rosetta mini. The Intel core 2 Duo's are the way to go for building a crunching machine, and they use much less power and electricity than an AMD.


Though this theory works for Intel CPU's which use a memory controller on the motherboard, which introduces high latency for main memory accesses. AMD CPU's have an integrated memory controller which means memory accesses are much faster than Intel chips. Therefore reducing the need for a large cache, which
a) introduces latency itself if it is not neccesary, and
b) bumps up the cost of the CPU as cache memory is manufactured using SRAM which is very expensive.

Intel are moving away from this problem with their next gen Nehalam chips with the Quickpath design
29) Message boards : Number crunching : Which file holds the run time parameter ? (Message 54912)
Posted 4 Aug 2008 by FoldingSolutions
Post:
... and which directory is it located in ?

(I assume its value is in seconds, not hours)

Thanx!


I don't have rosetta as an app right now, but I do recall this...
There is an XML file named sfter the project, so in this case something like boinc.bakerlab.org/rosetta and in there there is a line which defines how many seconds to run for (3600 = 1 hour) so the default of 3 hours will be 10800.
I also remember trying to change it using an XML editor but the rosetta app takes no notice of it.
Hope this helps

Matt
30) Message boards : Cafe Rosetta : Mars: Red and Dead? (Message 54061)
Posted 29 Jun 2008 by FoldingSolutions
Post:
Mars Mission Petition

Here is a petition that you can all sign to encourage the people in the right places to get a manned mission going to Mars around 2019 rather than 2039, or even at all.
Depending upon how you view this matter, whether you feel charity should begin at home rather than spending billions (or trillions?) on this when we have a couple billion people living in poverty on Earth.
However, the manned missions to the moon cost the US a lot of money back in the late Sixties and early Seventies, but brought together a world at a very fragile time in it's history, maybe this could do the same for our time frame.
No personal Info is needed other than a name [can be general, like "Bob" ;) ] and an e-mail.
Many thanks to any one who takes time to do this.

Matt...
31) Message boards : Number crunching : new machine..help with win defender/software exp (Message 53604)
Posted 9 Jun 2008 by FoldingSolutions
Post:
I do not have Vista now, but have had it before, and once shared your frustration. I don't think the blocking is being done by defender, hence why you can't find an option for it. But rather by another sub-program of the OS. It is called user control or something along those lines and if you do a search on windows help i'm sure you can find it. I think the whole thing can be turned off so you don't get those silly nuiscance "Did you mean to press this button" or "I can't believe you just said that to me!" messages.
32) Message boards : Number crunching : BOINC in Java code (Message 53526)
Posted 3 Jun 2008 by FoldingSolutions
Post:
I have downloaded this application but cannot seem to install it on my phone, how do we do this? I have a Sony Ericsson W850i
33) Message boards : Number crunching : WU Checkpoint Issue (Message 52346)
Posted 9 Apr 2008 by FoldingSolutions
Post:
To hibernate, click start, then control panel, then performance and maintenance, then power options, then on the top tabs click "hibernate", then enable hibernation. Now a file which exactly corresponds to the size of your RAM will be created on C: drive. Though it is hidden therefore not easily found. But more importantly, when you click the shutdown tab, hold shift and the stand by option will turn into hibernate :)
34) Message boards : Number crunching : Genesis of Rosetta@home (Message 52318)
Posted 7 Apr 2008 by FoldingSolutions
Post:
A news article released when Rosetta@home was in it's early stages, a very interesting read...
35) Message boards : Number crunching : PC downtime advancing science (Message 52315)
Posted 7 Apr 2008 by FoldingSolutions
Post:

[/quote]
"Volunteer computing represents a trend, an age old tradition, which is citizen science," explained François Grey, head of IT communications at Cern.

"That tradition, which has produced some very good science, is being accelerated by the web into something where you can take a scientific problem and give it to people on the web and they can actively help to solve that problem.

"Many more people can take part in many more scientific problems and are amateur scientists contributing in a real way, not just watching science from the outside, but being part of the scientific process themselves."


[/quote]

Cool find, something maybe should be put up here more often ;)
36) Message boards : Number crunching : WU Checkpoint Issue (Message 52297)
Posted 6 Apr 2008 by FoldingSolutions
Post:
Easy solution to the checkpoint problem, just put your computer into hibernate rather than shutdown, this way the work is preserved at the state it is in when you press the hibernate button, and there is not the issue of possible data loss or corruption or continued power usage as there is with standby. And if you need to shut your computer down, for whatever reason, then just try to make sure the workunit has check-pointed fairly recently by using Ingleside's method of opening the check-point file :)
HTH
37) Message boards : Number crunching : Can Rosetta be run from a USB Flash Drive? (Message 50424)
Posted 7 Jan 2008 by FoldingSolutions
Post:
Just install a fresh client on another computer (i.e. your own computer) and then copy the entire "BOINC" folder from program files to a Flash drive. Then open the boincmgr.exe file from the flash drive every time you want to run it, it works perfectly fine it's what I do. It may be a good idea to get permission as BOINC has a tendancy to overheat laptops and it avoids problems when your employer says "How did this happen?". "Ummm..."
38) Message boards : Number crunching : RAM instability yet no obvious reason (Message 50218)
Posted 31 Dec 2007 by FoldingSolutions
Post:
Is your memory voltage set correctly, what memory is it (512MB stick of 400MHz DDR PC3200) but what exactly is it , Corsair , OCZ , Crucial etc. A part number would be good and also to know what your BIOS memory voltage is set to.

Had similar problems when memory voltage was set -0.2V below spec of OCZ DDR2 DIMM - PC would run but give it anything "heavy" to do and ...... POP ! BSOD or lock up


That is an amazingly brilliant idea!!! I will try...thanks a lot!!
39) Message boards : Number crunching : RAM instability yet no obvious reason (Message 50170)
Posted 29 Dec 2007 by FoldingSolutions
Post:
Hey thx for your ideas their both things i hadn't thought of. I don't think it's the psu though because the problem only occurs when new RAM is inserted. The lower speeds do make it more stable, but only when reduced down to 200MHz which make the system overall slower and more unresponsive than when 512MB of 400MHz of RAM is in. It's a rather annoying predicament because it's the second brand of RAM (first being Corsair) that hasn't worked. So leads me to beleive I'm stuck with a measly half pint.
40) Message boards : Number crunching : RAM instability yet no obvious reason (Message 50147)
Posted 28 Dec 2007 by FoldingSolutions
Post:
I have just bought a 512MB stick of 400MHz DDR PC3200 RAM for my comp (which already has a stick the same as just purchased) and whenever I begin a RAM intensive task like a game or 'sandra lite' the system restarts. The RAM modules are situated in a 3 slot configuration (different to the 2x2 normally found) and the new module made by PNY and the old one by Infineon (previously Siemans). Both have a CAS latency of 3.0 when running at full speed (400MHz). And I have run memory stability tests on them and they come out clean! There is absolutely no reason I can think of why they cause this problem but I'm hoping at least some of you guys will have come across this problem before and know how to fix it :) thanks in advance for any help


Previous 20 · Next 20



©2024 University of Washington
https://www.bakerlab.org