Servers are back up, that's fine!

BUT:
In the meantime i was over at Einstein crunching some GPU-WUs.
Man, there you get a lot of more credits for the time... :/

---

Aloha, Uli

I got my fair share for CPU - about 50 vlar tasks in progress.

I tried the other day to crunch a vlar with GPU. It (AR 0.012) took about 4500 seconds to crunch on my 780. I tested it with lunatics benchmark when no other tasks were running either on CPU or GPU.

Haven't tested with 3 vlars at a time on NV GPU. :)

If you only crunch 1 w/u at a time those VLARs arn't to bad, but it's when a GPU is set to crunch more than 1 is when the problems with them start (on nvidia cards anyway), though it would be good to see how those new cards react to them when multiple VLARs are crunched even if it's just 1-2 of them along with a normal 1's.

Cheers.

Ah, we're back.
The site went AWOL for a while there.

---

Grant
Darwin NT

That it certainly did, but I suspect that it was just muarae1 taking a dive.

Cheers.

That it certainly did, but I suspect that it was just muarae1 taking a dive.

Cheers.

Especially as uploads, downloads and scheduler services were still working fine.

---

My only beef with the servers running out of work is that my i7 will then go to Einstein and download a huge pile of work, so then even when the Seti servers recover, it won't ask for new CPU work for the next week or so.

---

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

My only beef with the servers running out of work is that my i7 will then go to Einstein and download a huge pile of work, so then even when the Seti servers recover, it won't ask for new CPU work for the next week or so.

If you haven't already, under Project preferences in your Account panel for each project, set SETI@Home's Resource share to 100 and Einstein@Home's to 0 (zero.) Then you'll get a minimal amount of work from Einstein for your CPU and/or GPU only when there's none for that platform from SAH.

---

My only beef with the servers running out of work is that my i7 will then go to Einstein and download a huge pile of work, so then even when the Seti servers recover, it won't ask for new CPU work for the next week or so.

If you haven't already, under Project preferences in your Account panel for each project, set SETI@Home's Resource share to 100 and Einstein@Home's to 0 (zero.) Then you'll get a minimal amount of work from Einstein for your CPU and/or GPU only when there's none for that platform from SAH.

I like to get a little bit of work from Einstein and have my resource shares set where I want them. Einstein's share is less than 1/4 of what Seti's is.

Would it help if I upgraded Boinc? I'm still running 6.10.something.

---

David
Sitting on my butt while others boldly go,
Waiting for a message from a small furry creature from Alpha Centauri.

Before I fetch work from a secondary project, I lower the cache setting to .5 or 1 day. I still get 2 or 3 days work but it helps. The zero resource share thing will work, but you may get only one task per device at a time and not keep the host maxed out.

I don't think the BOINC version will impact the work fetch, but a recent version that supports app_config.xml will allow you to use the max_concurrent setting which I find useful in enforcing resource share. The newer versions use a different definition of resource share, and it can cause new problems.

---

Another Fred
Support SETI@home when you search the Web with GoodSearch or shop online with GoodShop.

The cricket graph has taken a dive and the SSP seems to have frozen and uploads are not working.

Backup projects are ready to take up the slack again if this lasts much longer.

Cheers.

The cricket graph has taken a dive and the SSP seems to have frozen and uploads are not working.

Yep, we're going down. All my uploads are failing.

---

Another Fred
Support SETI@home when you search the Web with GoodSearch or shop online with GoodShop.

Same here. Many uploads in progress but none completing.

---

Not only uploads not happening, but when I came here to find out what was going on, this site was down as well.
As well as no uploads, graphs show that prior to the outage AP WUs waiting for assimilation were backing up, and the splitters once again were choking & the ready-to-send buffer was quickly shrinking.

---

Grant
Darwin NT

Not only uploads not happening, but when I came here to find out what was going on, this site was down as well.
As well as no uploads, graphs show that prior to the outage AP WUs waiting for assimilation were backing up, and the splitters once again were choking & the ready-to-send buffer was quickly shrinking.

Same here. noticed I have many uploads on my machines. So I wanted to see what was up. Could'nt get to the site at all.

---

[/quote]

Old James

How long before the uploads start again ? how do i find out on the site ?
Thanks

---

How long before the uploads start again ? how do i find out on the site ?
Thanks

The Sever Status page here, http://setiathome.berkeley.edu/sah_status.html, which is currently frozen in time.

And the Cricket Graphs here, http://fragment1.berkeley.edu/newcricket/grapher.cgi?target=%2Frouter-interfaces%2Finr-211%2Fgigabitethernet6_17;ranges=d%3Aw%3Am%3Ay;view=Octets, which is run independently and gives a better real time view.

Cheers.

I added Einstein and set preferences as suggested.
How do I find out how fast my GPU is ? I bench-marked my CPU but cant find anything for GPU ? Thanks

I'm not sure about that, but I can tell you that your GPU is 1 of the slowest out there.

Cheers.

Now to being back on topic, my 660's are now onto a backup project and if things don't come back in 2hrs my 550Ti's will be doing the same.

[edit] Well Synergy is talking to me now, but Bruno is still in hiding. [/edit]

Cheers.

I wonder whats happend of late. After the colo move we were running pretty smooth. But the last few weeks has seen one thing after another. Is it hardware or software problems.
I do miss Matts weekly tech posts.

---

[/quote]

Old James

How long before the uploads start again ? how do i find out on the site ?
Thanks

The Sever Status page here, http://setiathome.berkeley.edu/sah_status.html, which is currently frozen in time.

And the Cricket Graphs here, http://fragment1.berkeley.edu/newcricket/grapher.cgi?target=%2Frouter-interfaces%2Finr-211%2Fgigabitethernet6_17;ranges=d%3Aw%3Am%3Ay;view=Octets, which is run independently and gives a better real time view.

Cheers.

Still frozen at almost 2330 PST. Looks like it will get looked at in the morning during the weekly outage.

---

Donald
Infernal Optimist / Submariner, retired