Welcome to MilkyWay@home

Posts by Wrend

1) Message boards : News : Nbody WU Flush (Message 72808)
Posted 15 Apr 2022 by Profile Wrend
Post:
It had been suggested in another forum thread that having more systems crunching Nbody WUs would help. I just have the one computer of course, but wouldn't mind switching over from Einstein@H on the CPU to MW@H once the queue is empty (I've set it to no new WUs), if that's the case. Thanks.

Usually I just crunch on the GPUs for MW@H, because my Titan Black cards can be DP/FP64 optimized beyond most other Nvidia cards, figuring that they could do more good here, while helping out other projects on the CPU threads.
2) Message boards : Number crunching : Validation inconclusive (Message 72769)
Posted 13 Apr 2022 by Profile Wrend
Post:
If there's some conciseness that this'd help, I'm fine with giving it a go after my queue of Einstein WUs run dry. I have it set to no new WUs now, but it might take a a few days.
3) Message boards : Number crunching : Validation inconclusive (Message 72761)
Posted 13 Apr 2022 by Profile Wrend
Post:
For me Separation have been very hit or miss, mostly miss, having to repeatedly request updates to get WUs and then only occasionally.
I'm doing this for the science, not the credits. Credits are a nice milestone to help keep track of things though, so naturally the less artificially manipulated or inflated they are is preferable to me. I do Separation specifically because my two Titan Black cards are more competent at crunching them due to their DP/FP64 capabilities. I'm doing Einstein@Home on the CPU.

You can see the batches of GPU tasks stuttering along here in my RAC listing in the BOINC manager UI. https://i.imgur.com/5gSp9WY.png

The GPUs have been actively crunching maybe 1/3 to 1/5 of the time on average within that RAC upward trajectory.
4) Message boards : Number crunching : PLEASE send more separation tasks ... (Message 72726)
Posted 12 Apr 2022 by Profile Wrend
Post:
Yeah, it's a little goofy. I'm back to having to do update requests in the hopes of getting some work units every once in a while, when I can. Also now there seems to be one stuck in "in progress" for a couple days that isn't on my computer anywhere that I can tell, even after resetting the project, certainly not running. https://milkyway.cs.rpi.edu/milkyway/workunit.php?wuid=417189391

Funny enough though, my average credit rates still seem to be going up (for now) since setting my computer up to work on MW@H again, and since then now there are enough work units still in "validation pending" and "validation inconclusive" that when they are validated it's still at a faster credit rate even though my computer is often idle in terms of MW@H GPU work units.
5) Message boards : News : Server Downtime March 28, 2022 (12 hours starting 00:00 UTC) (Message 72699)
Posted 11 Apr 2022 by Profile Wrend
Post:
...
Not sure how that works, but I do know even the cheapest ones have spare capacity you don't see, which is inserted when some of what you're using wears out. I assume after there's no spare left, the drive just shrinks? I'll know soon, one of mine is at "40% remaining life" and I'm thrashing it on virtualbox stuff. From a Google search, it appears they're not that clever, they just switch to read only mode so you can copy the data off them. How hard can it be to just use the unused space on the disk? Why does nobody think when designing things?
...

Yeah, I guess in theory you could increase the over-provisioning capacity and shrink usable partition sizes to only good cells. I would assume they prioritize having capacity and data integrity for as long as they can, and then selling more SSDs when they can't. My own personal general and loose rule of thumb has been to ideally use up to half the available capacity of an SSD if possible and then to upgrade or replace them when using more than three-fourths, putting the older SSDs on laptops or something which see less continuous use, but that's just me.

It seems that by the time I need higher capacities they're on the market, so so far it's worked out for me well enough, excluding some form-factors and interfaces that are harder to find.
6) Message boards : News : Server Downtime March 28, 2022 (12 hours starting 00:00 UTC) (Message 72682)
Posted 11 Apr 2022 by Profile Wrend
Post:
Yeah, that's what I was meaning, with SSDs using less power so more power efficient on power supplies and by extension making the power supplies a little more reliable in a drive enclosure for a NAS or whatever.

In a RAID you're typically not getting any usable data off of a single disk by itself, but regarding failing HDDs that aren't in a RAID, that data is often recoverable depending how far you want to delve into it and of course they have SMART specifications as well. HDDs are at least capable of recovering data where SSDs generally aren't. But yeah, SSDs have come a long way and aren't likely to fail on you. Still not quite what I would hope for in capacity regarding SLC and MLC, but I think their internal controllers these days can often make up for it with how they do wear leveling and so on. But then too HDDs are now sometimes incorporating some of these technologies with internal controllers and an internal SSD as cache, basically hybrid drives of a sort.

But anyway... We've gotten off on a tangent. I'm not sure what MW@H is using nor what specific challenges they may be facing with it.
7) Message boards : News : Server Downtime March 28, 2022 (12 hours starting 00:00 UTC) (Message 72679)
Posted 11 Apr 2022 by Profile Wrend
Post:
Yeah, my main system drive on my PC is a 4TB SSD Samsung Pro and the secondary system drive an older 1TB. MY 64GB of RAM makes for a decent cache too, or even RAM drive if I feel like it. HDDs are fine for storage drives still where speed is less of a concern. I think HDDs still have a viable place in some use scenarios (for now), but in general it's hard not to recommend SSDs. Lower power usage too, which can mean more power supply efficiency and reliability and so on.
8) Message boards : News : Server Downtime March 28, 2022 (12 hours starting 00:00 UTC) (Message 72677)
Posted 11 Apr 2022 by Profile Wrend
Post:
I'm not sure offhand where Western Digital makes them. They're "enterprise grade" HDDs made for server RAIDs and similar – fairly high-end as far as conventional HDDs go, not something you'd typically find in a PC or the like. They're made for capacity and reliability for continuous use over several years. Overall system speed is determined by the RAID setup, so typically not limited to individual disk speeds. That's another matter when rebuilding a RAID disk though, if the RAID is in use, and so on.

Anyway, it can potentially take quite a while and there are other factors to consider. It really just depends on the setup and use case scenario.
9) Message boards : News : Server Downtime March 28, 2022 (12 hours starting 00:00 UTC) (Message 72674)
Posted 11 Apr 2022 by Profile Wrend
Post:
It depends on the capacity and the technology being used – HDD vs SSD, etc. Can take up to maybe a day or two.
No disk of any technology takes that long to write to every sector of itself. 3 hours max.

Took about a full day for a 16TB WD Gold HDD which has a max sequential write speed of about 250MB/s. I don't recall exactly how long it was, but it was definitely more than 3 hours. As conventional HDDs go, it's a decently fast one, so I figured there are likely some HDDs out there that would take a fair bit longer.
10) Message boards : News : Server Downtime March 28, 2022 (12 hours starting 00:00 UTC) (Message 72668)
Posted 10 Apr 2022 by Profile Wrend
Post:
It depends on the capacity and the technology being used – HDD vs SSD, etc. Can take up to maybe a day or two.
11) Message boards : News : Server Trouble (Message 72627)
Posted 9 Apr 2022 by Profile Wrend
Post:
Thanks for the gpu longevity answers, I'll stick with more worry free CPU's only. That's one thing I like about World Community Grid, it's CPU only.

Sorry, I wasn't meaning to scare you off. I thought you were maybe wanting to optimize their total work potential. I'm running 4 MW@H tasks per GPU and could run up to 5 or 6 if I wanted to fully load them up, but fewer tasks per GPU shouldn't load them up too much to worry about either way beyond needing to clean dust off a bit more often, if you have your computer on and are using it anyway. I can generally run 2 tasks per GPU and play games at the same time without issue, since I use V-sync or frame rate limits in the video games for 60 FPS. 1 task per GPU wouldn't be much above idle for me, expect for keeping the clock rates up.

As Peter said above, GPUs are often much more powerful at crunching than CPUs for the work units that can make use of them. You can pretty easily tell from my computer's BOINC stats in my signature which projects I've used the GPUs or CPU on. Currently I'm running 8 MW@H tasks total on the GPUs, and 8 Einstein task on the CPU. https://i.imgur.com/YgGl6Zs.png
12) Message boards : News : Server Trouble (Message 72624)
Posted 9 Apr 2022 by Profile Wrend
Post:
I've been able to get a few more batches requesting updates here and there during the day, but now more recently, within the past couple hours or so, seem to be getting them more consistently automatically.
13) Message boards : News : Server Trouble (Message 72606)
Posted 8 Apr 2022 by Profile Wrend
Post:
I haven't used them nonstop for crunching the whole time, but my Titan Black cards have been in use for about... I guess 7 years now. In addition to the occasional dust removal, so far I've only had to replace the thermal paste on the GPUs once when they started overheating and throttling themselves. That did the trick and they're almost as good as new, dropping over 10°C under load.

Bearing that in mind, you probably want to keep a close eye on loads, voltage levels, temps, and fan speeds so you know what to expect from your cards and what you're willing to ask of them.

For now I've settled for having my cards run at about 80% load and 75% target power while dynamically (power load based) clocked down to 862MHz and a fan speed ranging between 50 to 60%.

Of course your mileage may vary, but all things in moderation, as they say.
14) Message boards : News : Server Trouble (Message 72601)
Posted 8 Apr 2022 by Profile Wrend
Post:
Likewise for me, having double precision optimized Titan Black GPUs, I prefer utilizing them with MW@H where they can do some good, so currently set up for 8 tasks in parallel on the 2 GPUs (4 each), when I can get them at least, and 8 Einstein tasks on the i7-3930K CPU. My computer is getting a bit long in the tooth, but still does a decent job of it.
15) Message boards : News : Server Trouble (Message 72588)
Posted 8 Apr 2022 by Profile Wrend
Post:
Similarly for me, though I haven't gotten any work units in the past day or so, but before then would often have to request an update to get a batch of them.

...

Edit: Just got some now after an update request. Thanks.

Edit 2: And it's run dry for me again. Well, a handful today so far at least.
16) Message boards : News : Server Trouble (Message 72586)
Posted 8 Apr 2022 by Profile Wrend
Post:
Thanks Tom. Yes, for me at least they don't seem to be downloading with BOINC, just deferred a few times when requesting an update for the project in the UI.
17) Message boards : Number crunching : Run Multiple WU's on Your GPU (Message 72583)
Posted 8 Apr 2022 by Profile Wrend
Post:
You can drop the milkyway_separation_modified_fit section. It's just milkway for the Separation tasks now.

Thanks.
18) Message boards : Number crunching : Run Multiple WU's on Your GPU (Message 72579)
Posted 8 Apr 2022 by Profile Wrend
Post:
I'm wondering what the current name syntax for work units types/apps all are now for the purposes of this file. I think it's been years since I've updated them. I currently have this to run 4 tasks per GPU for 2 SLIed Titan Black cards:

...
<app>
<name>milkyway</name>
<max_concurrent>0</max_concurrent>
<gpu_versions>
<gpu_usage>0.25</gpu_usage>
<cpu_usage>0.10</cpu_usage>
</gpu_versions>
</app>

<app>
<name>milkyway_nbody</name>
<max_concurrent>0</max_concurrent>
<gpu_versions>
<gpu_usage>0.25</gpu_usage>
<cpu_usage>0.10</cpu_usage>
</gpu_versions>
</app>

<app>
<name>milkyway_separation__modified_fit</name>
<max_concurrent>0</max_concurrent>
<gpu_versions>
<gpu_usage>0.25</gpu_usage>
<cpu_usage>0.10</cpu_usage>
</gpu_versions>
</app>
...


If someone could let me know, I'd appreciate it. Thanks.
19) Message boards : News : Server Trouble (Message 72578)
Posted 8 Apr 2022 by Profile Wrend
Post:
I believe Tom was working on something and called it a night last night to look into it more today, something to do with trying to flush stuck tasks or something. https://milkyway.cs.rpi.edu/milkyway/forum_thread.php?id=4863&postid=72565#72565 I haven't had work units for about a day now, maybe just under a day. My computer had been working on the Separation tasks though. But either way, hopefully we'll get more work units today at some point or a status update.
20) Message boards : Number crunching : Validation inconclusive (Message 72528)
Posted 6 Apr 2022 by Profile Wrend
Post:
My main concern regarding this matter is that my computer has had invalid work unites for this project in the past when my GPUs were overheating and throttling themselves, and I'm currently using a dynamic clock rate on them based on load to help keep them cooler overall. But so far nothing seems to have had an error nor been listed as invalid, so I guess I'm in the clear.


Next 20

©2024 Astroinformatics Group