Posts by Keith Myers

21) Message boards : Number crunching : SETI orphans (Message 2096011)
Posted 17 Mar 2022 by Profile Keith Myers Special Project $250 donor
Post:
Both the Einstein GW tasks along with the GPUGrid tasks are sensitive to memory bandwidth.

The 1650 only has 128 bit memory width versus the 192 bit memory width on the 1060.

So even though the 1650 has more memory than the 3GB 1060, the 1060 still outclasses the 1650 on these applications.
22) Message boards : Number crunching : SETI orphans (Message 2095988)
Posted 17 Mar 2022 by Profile Keith Myers Special Project $250 donor
Post:
Bernd at Einstein said that they
He also said they are developing a new Radio Pulsar search to make use of the high powered AMD and Nvidia cards.

Interesting information. I wonder what is classed as high-end

He didn't say exactly "high-end" This is what he said.

FGRPB1G will eventually run out of work as well, but we are working on a new Radio-Pulsar search that will run on the larger AMD and NVidia GPUs then


In this context it could mean newer cards with greater compute capability and likely more memory. So possibly not able to run on ten year old gpus anymore.

I'm not worrying about that as this issue is still far off in the future.
23) Message boards : Number crunching : SETI orphans (Message 2095859)
Posted 15 Mar 2022 by Profile Keith Myers Special Project $250 donor
Post:
Bernd at Einstein said that they recently received a ton of new Arecibo data.

He also stated that the FGRP5 search has exhausted any new "promising targets" = scientifically relevant "work" than the current one"

That is the reason they have deprecated search resources for the FGRP5 apps and are now concentrating on the new Arecibo BRP4 beams.

He also said they are developing a new Radio Pulsar search to make use of the high powered AMD and Nvidia cards.

So we might see a return of the old BRP4G application or a newer, updated version for this new data eventually.
24) Message boards : Number crunching : SETI orphans (Message 2095455)
Posted 7 Mar 2022 by Profile Keith Myers Special Project $250 donor
Post:
I haven't been able to snag much of any. Only 9 tasks on one host, nothing on the others. Just endless nothing to send request reports.
25) Message boards : Number crunching : Don't know where it should go? Stick it here! (Message 2095401)
Posted 5 Mar 2022 by Profile Keith Myers Special Project $250 donor
Post:
Curious to know where people putting their processing power in regards to Covid research? While there is no WCG. I am aware of Folding at Home (Non-boinc can use GPU & CPU) and sidock (Boinc currently CPU only) I am also aware there is a non-boinc version as well

I would assume the approximate conjugate project would be SiDock.
26) Message boards : Number crunching : Don't know where it should go? Stick it here! (Message 2095306)
Posted 3 Mar 2022 by Profile Keith Myers Special Project $250 donor
Post:
Very interesting. All the players are heading the same direction with MCM.
27) Message boards : Number crunching : SETI orphans (Message 2095143)
Posted 28 Feb 2022 by Profile Keith Myers Special Project $250 donor
Post:
I continue on with Einstein predominately for gpus and for Universe for cpus.
Our GPUUG team continues to put as much effort into improving the science apps at Einstein as we did with Seti.
Making great progress.
First iteration for the Separation app was at the same equivalent stage as the Lunatics apps adopted by Seti.
Einstein admins adopted those quickly to launch the 1.28 version.
Current iteration is at the Seti "special sauce" app stage. Basically a rewrite of the original application I believe based on the stderr.txt output.
Milkyway's boards are very infrequently used. Strangely quiet compared to most projects boards. No idea why.
28) Message boards : Number crunching : Ryzen and Threadripper (Message 2095018)
Posted 26 Feb 2022 by Profile Keith Myers Special Project $250 donor
Post:
I just copied the value from the story. Didn't check their math.
29) Message boards : Number crunching : Ryzen and Threadripper (Message 2094979)
Posted 26 Feb 2022 by Profile Keith Myers Special Project $250 donor
Post:
New AMD Milan-X cpus are up for ordering now at ShopBLT
https://www.shopblt.com/search/s_all=EPYC&s_max=25&s_lastpos=4688760&order_id=219337371
But a big 34% increase in price for the 7473X 24C over my existing 7443P 24C cpu. Think I will skip it and wait for Zen 4.
30) Message boards : Number crunching : Boinc manager disconnected (Message 2094960)
Posted 25 Feb 2022 by Profile Keith Myers Special Project $250 donor
Post:
Probably a permissions issue.
Why don't you start the client and the Manager in terminals so you can see the output of both for better troubleshooting information.
31) Message boards : Number crunching : SETI orphans (Message 2094016)
Posted 13 Feb 2022 by Profile Keith Myers Special Project $250 donor
Post:
Howsa yourself, Stephen?
Still plugging away in BOINC.
Cursing the 80° F. days in North American winter and having to use air conditioning already.
32) Message boards : Number crunching : OpenCL and Mint 20.2 (Message 2093997)
Posted 12 Feb 2022 by Profile Keith Myers Special Project $250 donor
Post:
https://rocmdocs.amd.com/en/latest/Installation_Guide/Installation_new.html#rocm-stack-uninstallation
33) Message boards : Number crunching : OpenCL and Mint 20.2 (Message 2093374)
Posted 2 Feb 2022 by Profile Keith Myers Special Project $250 donor
Post:
Oh, I get it. You have installed a nvme drive in a USB carrier. I thought it was installed in a motherboard M.2 slot.
Have you looked in the USB hosted nvme drive via Disks to see if it has a FAT UEFI partition on it?
You may be having conflicts with two UEFI partitions hosting two different grub configs.
You should have only one in play.
Whatever Mint installation is your normal default one probably should support the grub file that boots both versions of Mint.
If you can get both Mint versions visible to your default OS, I would run os-prober and see if you can get it to write a correct grub file for both Mint entries.
Barring getting that to work, I would do some editing in the default /boot/grub/grub.cfg file to add the second Mint version. You could just copy the block for the normal Mint OS entry behind the original one and just change the UUID entry for the second Mint version.
Then go into your /etc/default/grub file and change the GRUB_DEFAULT= line to point at the preferred Mint OS entry. Remember to do a update-grub afterwards.
34) Message boards : Number crunching : OpenCL and Mint 20.2 (Message 2093358)
Posted 1 Feb 2022 by Profile Keith Myers Special Project $250 donor
Post:
I think it is a bit much to have to open the computer and pull out the nvme drive from the slot every time you want to boot the other Mint version.

The drive names shouldn't be changing for the nvme drives. You mention a USB drive though and I am confused. USB drive names will change if you plug them into a different port.

The drive names shouldn't be used for the grub entry or the fstab entries. That is why I suggested you define a drive by its UUID. The UUID will not change. It is like a fingerprint identifying the drive no matter where it is installed.

You didn't by chance clone the second nvme drive from the 1st did you. That would make both drives have the same UUID and I can see why you are having issues. That would be a big no-no.
35) Message boards : Number crunching : Ryzen and Threadripper (Message 2093085)
Posted 28 Jan 2022 by Profile Keith Myers Special Project $250 donor
Post:
I think the implementation of AVX-512 is a case of marketing guys getting the upper hand over the design team in forcing AVX-512 to better match Intel capabilities. Every review you see comparing Ryzen/Threadripper/Epyc to Intel HEDT/server parts has the lack of AVX-512 support number one in the disadvantage lists.

The marketing people want to knock that disadvantage from the list right now.

I agree the amount of applications that actually implement AVX-512 is very limited. I also agree it is a waste of die space for BOTH Intel and AMD.
36) Message boards : Number crunching : Ryzen and Threadripper (Message 2093002)
Posted 27 Jan 2022 by Profile Keith Myers Special Project $250 donor
Post:
For those interested in the high-level architectural changes coming in AM5 processors, the Gigabyte leak and the analysis from Chips and Cheese provides great information.
https://chipsandcheese.com/2021/08/23/details-on-the-gigabyte-leak/
37) Message boards : Number crunching : OpenCL and Mint 20.2 (Message 2092108)
Posted 13 Jan 2022 by Profile Keith Myers Special Project $250 donor
Post:
Check your /boot/grub/grub.cfg file and /etc/fstab file to make sure the grub entry for each Mint version is pointing to the correct M.2 drive UUID.
Run sudo blkid to find the UUID of each M.2 drive.
38) Message boards : Number crunching : IBM steps away from WCG. Moving it to Krembil Research Institute. (Message 2091886)
Posted 10 Jan 2022 by Profile Keith Myers Special Project $250 donor
Post:
I've received and processed 2 1/2 pages of OPNG tasks today. 5 1/2 pages yesterday.
The tasks are available if you ask for them regularly.
39) Message boards : Number crunching : Ryzen and Threadripper (Message 2091681)
Posted 7 Jan 2022 by Profile Keith Myers Special Project $250 donor
Post:
AMD Zen4, AM5 platform Ryzen 9 7950X ES sample being used at Milkyway on the cpu N-body MT tasks. Bonkers BOINC integer benchmarks.
https://milkyway.cs.rpi.edu/milkyway/show_host_detail.php?hostid=911833
Still haven't had any wingman complete the paired tasks to compare the processing times. Never ran the N-body tasks before so I don't know what representative times for current hardware is.
The floating point benchmarks are nothing out of the ordinary compared to my Ryzen 9 5950X though. Just a bit better.
40) Message boards : Number crunching : Don't know where it should go? Stick it here! (Message 2091638)
Posted 7 Jan 2022 by Profile Keith Myers Special Project $250 donor
Post:
Best way to figure an app_name out is to look in the client_state file. Or put in some bogus name in the app_config and BOINC should report it is an unrecognized name and offer the known ones it knows about.


Previous 20 · Next 20


 
©2024 University of California
 
SETI@home and Astropulse are funded by grants from the National Science Foundation, NASA, and donations from SETI@home volunteers. AstroPulse is funded in part by the NSF through grant AST-0307956.