Posts by Aurum

1) Message boards : Number crunching : Rosetta WU delivery out of control (Message 95806)
Posted 2 May 2020 by Aurum
Post:
It has nothing to do with WUs failing or the runtime estimate being wrong. I can crunch any Rosetta WU they send. Rosetta just simply does not respect BOINC settings and DLs far too many WUs. Client side fix is to set all computers to No New Work and abort a few thousand a day.
2) Message boards : Number crunching : L3 Cache is not a problem, CPU Performance Counter proves it (Message 94253)
Posted 12 Apr 2020 by Aurum
Post:
People said that Rosetta application is not optimized and needs 5MB of L3 cache for each instance running (though I don't really know where they got this number).
We observed it running MIP at WCG. When we dedicated all threads to MIP fewer results were completed in the same time. We verified this repeatedly for months. Then the project made a statement explaining:
"The short version is that Rosetta, the program being used by the MIP to fold the proteins on all of your computers, is pretty hungry when it comes to cache. A single instance of the program fits well in to a small cache. However, when you begin to run multiple instances there is more contention for that cache. This results in L3 cache misses and the CPU sits idle while we have to make a long trip to main memory to get the data we need. This behavior is common for programs that have larger memory requirements. It's also not something that we as developers often notice; we typically run on large clusters and use hundreds to thousands of cores in parallel on machines. Nothing seemed slower for us because we are always running in that regime."
https://www.worldcommunitygrid.org/forums/wcg/viewthread_thread,40374_offset,50#569786

I doubt it's been fixed unless Rosetta coders consciously went in and rewrote the offending code. Why won't Baker respond to this question???
3) Message boards : Number crunching : Rosetta WU delivery out of control (Message 94252)
Posted 12 Apr 2020 by Aurum
Post:
Last night Rosetta delivered 999 WUs to a quad-core computer with a total estimated time to complete of 265 days. Its queue was set to 0.5/0.1 days. This problem has gotten so bad I've had to set all my computers to accept no new work from Rosetta. Only when it runs out do I let it download way too much work requiring that most be aborted.
I'm tired of babysitting this project. From now on I'll just let it send me hundreds of times more than I can crunch and they can sit there until they expire.
4) Message boards : News : Rosetta's role in fighting coronavirus (Message 93252)
Posted 3 Apr 2020 by Aurum
Post:
...long list of target proteins we're interested in for COVID-19 (some of which are COVID-19 itself).

I'm so confused. No idea what proteins you're talking. Would love to hear more details.
Covid-19 is the disease.
SARS-CoV-2 is the virus.
NSP1, NSP2, NSP3, NSP4, NSP5, NSP6, NSP7, NSP8, NSP9, NSP10, NSP12 NSP13, NSP14, NSP15 ,NSP16, S, ORF3a, E, M, ORF6, ORF7a, ORF8, N & ORF10 are the proteins.
5) Message boards : News : Help in the fight against COVID-19! (Message 93248)
Posted 3 Apr 2020 by Aurum
Post:
Will you make any COVID-19 badges ? ;-)


That would be nice.
6) Message boards : News : Help in the fight against COVID-19! (Message 93202)
Posted 3 Apr 2020 by Aurum
Post:
Dear Brian Coventry et alia,
1. Your custom protein binding the Spike protein looks like it will put neutralizing IgGs to shame!!! Very impressive!!! {As an aside each dose of this protein therapeutic will have to be injected as it could never survive a transit of the stomach and upper GI tract with the low pH and proteolytic enzymes that would reduce it to amino acids.}

2. I'm laboring under the assumption that Rosetta code incorrectly programmed the use of the L3 cache and will bottleneck if too many Rosetta WUs are running simultaneously. I limited my use to one WU per 5 MB of L3 cache. Exceeding that limitation slows the entire CPU over 60%. This means that my fleet of Xeon E5s are running Rosetta at less than 20% capacity. Has this been fixed and is it safe to run full force now???

3. I believe that your statement implies that all WUs are contributing to Covid-19 research regardless of whether you include "Covid-19" in the WU name and that minirosetta is doing Covid-19 work as well. Is that what you mean???

4. I don't know if I'm running ARM64 Rosetta aps or not. I see from your Applications page there are numerous versions but when I look at the Properties of a given WU I'm running it says the version 3.76, 4.07, 4.08 or 4.12. What are the Linux-ARM platforms???

5. I suggest you tether these blocking proteins together like an IgM. This would more readily mark them for destruction. Also, it appears that your blocking protein is much larger than an IgG so when it binds a Spike it also blocks neighboring Spikes (like an umbrella) from binding ACE2 and being invaginated.

{Please use black text on white background for your forum to maximize contrast. This faded gray text really strains my old eyes so I never read your forums. I would not have seen this announcement except that you were nice enough to send out a BOINC Notice. Thank you in advance :-}
7) Message boards : News : Rosetta's role in fighting coronavirus (Message 92227)
Posted 24 Mar 2020 by Aurum
Post:
...delayed by all of the time required to physically isolate these things in the lab...
This is the first step. Isolate the virus and sequence its genome. Then model. Not sure what your point is.
...eventually perform x-ray spectroscopy...
Most proteins do not form single crystals and cannot have their structure determined by x-ray spectroscopy.
8) Message boards : Number crunching : How to get only Rosetta mini ? (Message 91521)
Posted 4 Jan 2020 by Aurum
Post:
Now that I realize the root cause of the problem is David Baker's flawed code I doubt it matters whether one runs rosetta or minirosetta WUs. I'll know in a day of testing.
9) Message boards : Number crunching : Preferences: Target CPU run time (Message 91520)
Posted 4 Jan 2020 by Aurum
Post:
Now I remember, David Baker uses defective code.
10) Message boards : Number crunching : How do you calculate points? (Message 91519)
Posted 4 Jan 2020 by Aurum
Post:
That's only part of the problem. The real problem is the code was programmed wrong and David Baker has known about the defective code for years. Baker should explain when he'll fix it. It's so cavalier to waste patrons money.
11) Message boards : Number crunching : How to get only Rosetta mini ? (Message 91514)
Posted 3 Jan 2020 by Aurum
Post:
This is still not available. I don't want any minis so I'll have to abort them.
12) Message boards : Number crunching : Preferences: Target CPU run time (Message 91513)
Posted 3 Jan 2020 by Aurum
Post:
I was looking through my Valid Tasks and it's obvious that there is no rhyme nor reason to the credit system on this project.
I was looking for a way test different length WUs.
13) Message boards : Number crunching : Preferences: Target CPU run time (Message 91512)
Posted 3 Jan 2020 by Aurum
Post:
Has anybody figured out if picking longer or shorter times is more efficient???
14) Message boards : Number crunching : Cryptocurrency Can Cover Electrical Costs? (Message 90199)
Posted 13 Jan 2019 by Aurum
Post:
Togo, Thanks, I've installed that BBP wallet and waiting for it to make connections and sync. It doesn't look like its network has any activity. Is it still working for you???
15) Message boards : Cafe Rosetta : Black & White, a Contrast in Visibility (Message 90154)
Posted 6 Jan 2019 by Aurum
Post:
This website is so hard for me to read. The color has no contrast. Black text on White background has the highest contrast and is the easiest to read.
16) Message boards : Number crunching : Larger Memory Models (Message 90149)
Posted 5 Jan 2019 by Aurum
Post:
...review your BOINC Manager preferences and the settings for the amount of memory, and virtual memory that BOINC Manager is allowed to use, when the system is both active, and idle.
Where is this memory control setting you're talking about???
17) Message boards : Number crunching : For the betterment of BOINC (Message 90148)
Posted 5 Jan 2019 by Aurum
Post:
1. Projects that run both CPU & GPU WUs must separate their controls.
It causes no end of problems trying to cope with a one-size-fits-all approach. Now one goes to Edit Project Preferences in My Account and you check the boxes of the kind of WUs you'll get. If you're running multiple computers then some will not have enough memory to run CPU projects but can run the GPU WUs no problem.
Or, it may be better to give more client-side control than presently available with cc_config & app_config. That's not in keeping with your request to make it easy for anybody to join and run BOINC without having to commit any time to learning the details.
At present I usually have no choice but to only run GPU WUs for one project and get my CPU WUs from other projects.

2. Every project should have a table spelling out what kind of WUs their project provides with their minimum requirements.
I tried to come back to Rosetta yesterday and The Beast of BOINC ate all 16 GB of my RAM and gave 3 of my computers LockJaw. I'm here looking for requirements and ideas for app_config to adapt to each computer I run Rosetta on, but, it should be as easy as clicking Requirements.
18) Questions and Answers : Wish list : Are you considering GPU usage? (Message 90147)
Posted 5 Jan 2019 by Aurum
Post:
Maybe a reasonable trade-off is to have two versions of Rosetta. One which use GPU and the other which use CPU only.
Is this feasible?
It's extremely important that they be separate. The controls for CPU projects has to be totally separate from those for GPU projects.
My compute node has a range of computer capabilities and one-size does NOT fit all in distributed computing.
Especially when The Beast of BOINC consumes all of your RAM.






©2020 University of Washington
https://www.bakerlab.org