SETI@Home Wow!-Event 2018

[StefanR5R]

@lane42, keep in mind though that there is bunkering going on around us at a large scale.

 

[lane42]

I see the #1 guy on [H] has been quiet for a few days.

 

[Assimilator1]

Cool

Hi Lane42 , btw you're sig is 4 yrs out of date

 

[lane42]

Hi Mark, good to see you posting. I guess iam getting lazy in my old age

 

[ao_ika_red]

Points per 4 hours, 3rd - 5th team:

Something's going on with GC.

 

[Markfw]

My 2990WX and 1080TI are running for now. I have to go down for the water cooling install Tues or Wed, its only running 3550 mhz now

 

[StefanR5R]

Something's going on with GC.

Do you mean [H]?
It's reasonable to assume that those who stopped uploading will upload eventually, and that those who have uploaded only dribbles yet have (or had) a marginally leaky bunkering procedure going on and will upload on one of the last days. Besides, late registrations remain possible.

Edit,
[H] were overtaken by us at August 19 22:00UTC and by GC at August 20 00:00UTC. But don't read too much into that.

 

[ao_ika_red]

Do you mean [H]?

Oh, yeah [H]! I was looking at yellow bar but my brain said it was GC instead of [H].

 

[StefanR5R]

Days 3...5, top sixteen teams:

Points per 4 hours, 3rd...5th team:

Note, TeAm AnandTech's color is now yellow, not Noctua-brown. I am too lazy to keep stable color codes per team. And remember, [H]ard|OCP's graphs are not to be used to derive a trend.

 

[Assimilator1]

But I can read into our trend? Cool

Lane42
I'd be lazier & just say crunchin' since '99 ....... or crunchin' since the last century

 

[StefanR5R]

I am still undecided myself how I will manage Maintenance Tuesday during the Wow!-Event. Let the hosts sit idle? Probably not, but maybe if weather is hot. Set up multiple clients per host, as Tony suggested? If so, do I run them in parallel or one after the other? I am not sure, but I believe I had several clients running in parallel during maintenance downtime at last year's Wow!-Event. Or do I take it easier with SETI this time and run a 0 % backup project instead? ... Decisions, decisions.

As it happens, my hosts are now set up to keep on SETIing for about 1/2 day without server contact. On the CPU hosts, it's a straightforward solution with several client instances running in parallel. On the GPU hosts, a few client instances will be run in series, but it is a makeshift solution which I may turn into an automated ring buffer like setup eventually.

For situations like this, it is convenient to have separate hosts or clients for CPU work and GPU work.

 

[Markfw]

So I was trying to figure out why my 2990WX was only at 69%load, when I clearly set it for 2 1/2 days to queue tasks, but when I try to check my stats, its down for maintenance ! In the middle of this event ?

 

[StefanR5R]

So I was trying to figure out why my 2990WX was only at 69%load, when I clearly set it for 2 1/2 days to queue tasks, but when I try to check my stats, its down for maintenance ! In the middle of this event ?

SETI@home limits us to 100 CPU tasks in progress per host, and 100 GPU tasks in progress per GPU. I suspect a 2990WX goes through 100 tasks in a matter of hours.

And yes, SETI@home shuts down every Tuesday morning (Californian morning) for several hours... or longer. From what I heard, the duration is not predictable, and the main purpose is to create a consistent database dump for backup.

--------

Days 4...6, top fifteen teams:

Points per 4 hours, 3rd...5th team:

The last update is already diminished by SETI@home's Tuesday downtime.
[H]ard|OCP continue to play submarine.
Gridcoin gained a productive user yesterday and almost match our output now.

 

[Markfw]

Well, its doing the CPU tasks@about 1:50 each, and 63 at a time, so 100 tasks in about 3 hours.

 

[lane42]

That limit is way to small imo. It might of been ok years ago with slower cpu and
gpu's, but not today.

 

[StefanR5R]

That limit is way to small imo.

It's about only topped by GPUGrid (2 in progress per GPU; they want quick returns like F@H) and YAFU (2 in progress per host). Ah, and king of low quota is F@H: ≈1 task in progress per GPU, plus ≈1 per CPU slot. ("≈" because the client starts fetching a new task a few seconds before the one in progress is actually finished.)

On the other hand, SETI@home is among the projects with longest task deadlines. This means that effective participation in the Wow!-Event happens in July, not in August. (Edit: That's what SETI.Germany's top users do. Their #1 for example has a live production of 7,000...8,000 per hour, the rest comes from buffers which were completed before the start of the event.)

--------

A certain xii5ku's stats:

11th per total credits
10th by daily output 2 days ago
7th by yesterday's output
5th by output today before project downtime​

This is a lot harder than PrimeGrid challenges.

 

[Thebobo]

Hmm I'm getting "no work available to process" this the server outage talked about?

 

[biodoc]

Hmm I'm getting "no work available to process" this the server outage talked about?

Yes, the site is down for maintenance. 4+ hours and counting. Their goal is 5 hours downtime.

 

[biodoc]

Just as I posted Seti came back up.

 

[Markfw]

I am still down for water cooling conversion. ETA sometime tomorrow.

 

[StefanR5R]

Half-time!

Days 5...7, top fifteen teams:

Points per 4 hours, 3rd...5th team:

 

[StefanR5R]

And now... some unexpected and totally unfair TeAm AnandTech Subteam stats.
Taken at August 22, 20:00UTC.

Taurus: 3,778,481

(10esseeTony, Tom_Servo, xii5ku)​

Libra: 2,482,950

(CFrench, crashtech, emoga, Ingleside, Orange Kid, petrusbroder)​

Virgo: 1,404,304

(Howdy2u2)​

Gemini: 1,186,542

(Lane42)​

Aries: 896,579

(biodoc, MarcelliusIkaP)​

Cancer: 677,041

(GLeeM, Ken_g6 - Team Anandtech, QuietDad)​

Aquarius: 607,486

(Tejas)​

Scorpio: 136,440

([TA]Assimilator1, kiska)​

Capricornus: 51,698

(Mark-4790k)​

Pisces: 25,097

(curiousmike)​

Leo: 14,175

(ericlp)​

@TennesseeTony, you booked a risky flight again.

 

[Markfw]

And now... some unexpected and totally unfair TeAm AnandTech Subteam stats.
Taken at August 22, 20:00UTC.

Taurus: 3,778,481

(10esseeTony, Tom_Servo, xii5ku)​

Libra: 2,482,950

(CFrench, crashtech, emoga, Ingleside, Orange Kid, petrusbroder)​

Virgo: 1,404,304

(Howdy2u2)​

Gemini: 1,186,542

(Lane42)​

Aries: 896,579

(biodoc, MarcelliusIkaP)​

Cancer: 677,041

(GLeeM, Ken_g6 - Team Anandtech, QuietDad)​

Aquarius: 607,486

(Tejas)​

Scorpio: 136,440

([TA]Assimilator1, kiska)​

Pisces: 25,097

(curiousmike)​

Leo: 14,175

(ericlp)​

@TennesseeTony, you booked a risky flight again.

My capricorn is not there ?

 

[StefanR5R]

@Markfw, sorry, I had this in my spreadsheet but missed to copy it into the post. #-|
I edited Capricornus in now.

 

[crashtech]

I'm having a strange problem with one of my machines (Ryzen 1600 and an HD7970), where the first client instance detects two GPUs, and the second won't detect any. It's affecting my ability to continue crunching Seti while also working yoyo for the Sprint. Has anyone seen this happen?

8/23/2018 11:37:26 AM |  | Starting BOINC client version 7.12.1 for windows_x86_64
8/23/2018 11:37:26 AM |  | log flags: file_xfer, sched_ops, task
8/23/2018 11:37:26 AM |  | Libraries: libcurl/7.47.1 OpenSSL/1.0.2g zlib/1.2.8
8/23/2018 11:37:26 AM |  | Data directory: C:\ProgramData\BOINC2
8/23/2018 11:37:26 AM |  | Running under account Artisan
8/23/2018 11:37:26 AM |  | No usable GPUs found
8/23/2018 11:37:26 AM |  | Host name: Artisan2
8/23/2018 11:37:26 AM |  | Processor: 12 AuthenticAMD AMD Ryzen 5 1600 Six-Core Processor [Family 23 Model 1 Stepping 1]
8/23/2018 11:37:26 AM |  | Processor features: fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 htt pni ssse3 fma cx16 sse4_1 sse4_2 movebe popcnt aes f16c rdrandsyscall nx lm avx avx2 svm sse4a osvw skinit wdt tce topx page1gb rdtscp fsgsbase bmi1 smep
8/23/2018 11:37:26 AM |  | OS: Microsoft Windows 10: Core x64 Edition, (10.00.17134.00)
8/23/2018 11:37:26 AM |  | Memory: 15.94 GB physical, 16.13 GB virtual
8/23/2018 11:37:26 AM |  | Disk: 418.23 GB total, 254.54 GB free
8/23/2018 11:37:26 AM |  | Local time is UTC -6 hours
8/23/2018 11:37:26 AM |  | No WSL found.
8/23/2018 11:37:26 AM |  | VirtualBox version: 5.1.30
8/23/2018 11:37:26 AM |  | Config: allow multiple clients
8/23/2018 11:37:26 AM |  | Config: use all coprocessors
8/23/2018 11:37:26 AM | SETI@home | URL http://setiathome.berkeley.edu/; Computer ID 8566948; resource share 100000
8/23/2018 11:37:26 AM | SETI@home | Your current settings do not allow tasks from this project.  To fix this, you can change Project Preferences on the project's web site.
8/23/2018 11:37:31 AM |  | General prefs: from http://www.worldcommunitygrid.org/ (last modified 17-Jul-2018 22:05:32)
8/23/2018 11:37:31 AM |  | Host location: none
8/23/2018 11:37:31 AM |  | General prefs: using your defaults
8/23/2018 11:37:31 AM |  | Preferences:
8/23/2018 11:37:31 AM |  | max memory usage when active: 14686.97 MB
8/23/2018 11:37:31 AM |  | max memory usage when idle: 14686.97 MB
8/23/2018 11:37:31 AM |  | max disk usage: 100.00 GB
8/23/2018 11:37:31 AM |  | (to change preferences, visit a project web site or select Preferences in the Manager)
8/23/2018 11:37:31 AM |  | Setting up project and slot directories
8/23/2018 11:37:31 AM |  | Checking active tasks
8/23/2018 11:37:31 AM |  | Setting up GUI RPC socket
8/23/2018 11:37:31 AM |  | Checking presence of 4 project files