Here are my results
Asus P5B-E, Core2Duo E6600, WinXP pro, Geforce GT 430 (Zotac PCI card):

WU : PG0009_v7.wu
Lunatics_x41zc_win32_cuda50.exe -verb -device 1 :
Elapsed 683.266 secs
CPU 7.766 secs
Lunatics_x41zc_win32_cuda50_commode.exe -verb -device 1 :
Elapsed 683.891 secs
CPU 7.391 secs
(...)

All in all it's a +/- 0 = draw.

You used the cuda50 app, why not the cuda42 for your GT430?
AFAIK, 4xx/5xx series cuda42 and 6xx/7xx series cuda50.

---

Got home to find uploads all backed up again, and network traffic reflects that. Inbound is less than 4MB/s instead of 12-16MB/s.

---

Grant
Darwin NT

Yep all uploads are stuck again. If I run out it backup time.

---

[/quote]

Old James

Yep all uploads are stuck again. If I run out it backup time.

...and with all the shorties being fed to us that's another problem on top !

---

Well my main rig's GPU's have now hit up their backup project.

Cheers.

You used the cuda50 app, why not the cuda42 for your GT430?
AFAIK, 4xx/5xx series cuda42 and 6xx/7xx series cuda50.

Yes, i thought the same, but tested the following:
Device 0 is GT 640, device 1 is GT 430

Quick timetable

WU : PG0009_v7.wu
setiathome_7.00_windows_intelx86.exe -verb -nog :
Elapsed 863.750 secs
CPU 861.203 secs
Lunatics_x41zc_win32_cuda42.exe -device 0 :
Elapsed 331.859 secs, speedup: 61.58% ratio: 2.60x
CPU 8.188 secs, speedup: 99.05% ratio: 105.18x
Lunatics_x41zc_win32_cuda42.exe -device 1 :
Elapsed 756.609 secs, speedup: 12.40% ratio: 1.14x
CPU 6.563 secs, speedup: 99.24% ratio: 131.22x
Lunatics_x41zc_win32_cuda50.exe -device 0 :
Elapsed 347.734 secs, speedup: 59.74% ratio: 2.48x
CPU 6.734 secs, speedup: 99.22% ratio: 127.89x
Lunatics_x41zc_win32_cuda50.exe -device 1 :
Elapsed 750.656 secs, speedup: 13.09% ratio: 1.15x
CPU 6.563 secs, speedup: 99.24% ratio: 131.22x

WU : PG0395_v7.wu
setiathome_7.00_windows_intelx86.exe -verb -nog :
Elapsed 956.781 secs
CPU 954.391 secs
Lunatics_x41zc_win32_cuda42.exe -device 0 :
Elapsed 135.781 secs, speedup: 85.81% ratio: 7.05x
CPU 8.844 secs, speedup: 99.07% ratio: 107.91x
Lunatics_x41zc_win32_cuda42.exe -device 1 :
Elapsed 320.703 secs, speedup: 66.48% ratio: 2.98x
CPU 9.594 secs, speedup: 98.99% ratio: 99.48x
Lunatics_x41zc_win32_cuda50.exe -device 0 :
Elapsed 134.969 secs, speedup: 85.89% ratio: 7.09x
CPU 7.547 secs, speedup: 99.21% ratio: 126.46x
Lunatics_x41zc_win32_cuda50.exe -device 1 :
Elapsed 315.797 secs, speedup: 66.99% ratio: 3.03x
CPU 10.297 secs, speedup: 98.92% ratio: 92.69x

WU : PG0444_v7.wu
setiathome_7.00_windows_intelx86.exe -verb -nog :
Elapsed 848.063 secs
CPU 845.781 secs
Lunatics_x41zc_win32_cuda42.exe -device 0 :
Elapsed 120.031 secs, speedup: 85.85% ratio: 7.07x
CPU 7.969 secs, speedup: 99.06% ratio: 106.13x
Lunatics_x41zc_win32_cuda42.exe -device 1 :
Elapsed 287.641 secs, speedup: 66.08% ratio: 2.95x
CPU 8.188 secs, speedup: 99.03% ratio: 103.30x
Lunatics_x41zc_win32_cuda50.exe -device 0 :
Elapsed 118.859 secs, speedup: 85.98% ratio: 7.14x
CPU 7.281 secs, speedup: 99.14% ratio: 116.16x
Lunatics_x41zc_win32_cuda50.exe -device 1 :
Elapsed 282.297 secs, speedup: 66.71% ratio: 3.00x
CPU 9.938 secs, speedup: 98.82% ratio: 85.11x

WU : PG1327_v7.wu
setiathome_7.00_windows_intelx86.exe -verb -nog :
Elapsed 594.438 secs
CPU 591.563 secs
Lunatics_x41zc_win32_cuda42.exe -device 0 :
Elapsed 158.219 secs, speedup: 73.38% ratio: 3.76x
CPU 14.438 secs, speedup: 97.56% ratio: 40.97x
Lunatics_x41zc_win32_cuda42.exe -device 1 :
Elapsed 400.563 secs, speedup: 32.61% ratio: 1.48x
CPU 11.953 secs, speedup: 97.98% ratio: 49.49x
Lunatics_x41zc_win32_cuda50.exe -device 0 :
Elapsed 154.016 secs, speedup: 74.09% ratio: 3.86x
CPU 12.563 secs, speedup: 97.88% ratio: 47.09x
Lunatics_x41zc_win32_cuda50.exe -device 1 :
Elapsed 383.141 secs, speedup: 35.55% ratio: 1.55x
CPU 16.281 secs, speedup: 97.25% ratio: 36.33x

As you can see, the cuda50 executable wins over the cuda42 executable, regardless of GT 640 or GT 430. That's why i use cuda50 on both devices.

---

Aloha, Uli

Is this the same problem....

2/27/2014 7:23:37 AM | SETI@home | Started upload of 16au13ac.14495.14382.438086664205.12.174_0_0
2/27/2014 7:23:51 AM | SETI@home | Started upload of 16au13ad.21976.11110.438086664203.12.124_1_0
2/27/2014 7:24:00 AM | SETI@home | Temporarily failed upload of 16au13ac.14495.14382.438086664205.12.174_0_0: connect() failed
2/27/2014 7:24:00 AM | SETI@home | Backing off 2 min 56 sec on upload of 16au13ac.14495.14382.438086664205.12.174_0_0
2/27/2014 7:24:13 AM | SETI@home | Temporarily failed upload of 16au13ad.21976.11110.438086664203.12.124_1_0: connect() failed
2/27/2014 7:24:13 AM | SETI@home | Backing off 3 min 44 sec on upload of 16au13ad.21976.11110.438086664203.12.124_1_0
2/27/2014 7:24:18 AM | | Project communication failed: attempting access to reference site
2/27/2014 7:24:21 AM | | Internet access OK - project servers may be temporarily down.

---

I wonder if the 1-2 weeks until a server swap out is
due to manpower/budget restraints?

Sounds more like 1-2 weeks for the hardware to arrive, and then get an OS installed, get everything configured, and then load the last back-up of the database. Everything after the hardware arrives and gets installed is a day or two. It's just the sitting around, twiddling of thumbs waiting for a box to arrive.

I was thinking that they would have to schedule time with the colo management for the hardware swap. With their schedules an open slot would have to line up with their free time.

I'm just glad they had the time to post a notice so we know what is occurring.

---

SETI@home classic workunits: 93,865 CPU time: 863,447 hours
Join the [url=http://tinyurl.com/8y46zvu]BP6/VP6 User Group[

Perhaps some of that time is required for the SETI staff to get the software loaded and the server and its hardware set up to run correctly, tested, etc.

If we think about it, it is quite understandable. With only several staff members available and capable to replace the server, and with their time split over many tasks, it could easily take 1 to 2 weeks to get everything ready, installed and tested before going online.

Kirk: "Scotty, how long does it take to repair the warp engine?"
Scotty: "Captain, i'm sorry, the dilithium has disengaged, so at least a day or two!"
Kirk: "Scotty, i give you 8 hours!"
Scotty: "Alright sir, it'll be working again in two..."

;)

---

Aloha, Uli

LOL - just don´t forget our friend Murphy, seems like he was visiting the COLO few times in the last weeks, normaly in the middle of the night in CA.

BTW Fix a warp engine is a lot easier than fix the SETI servers. :(

---

Is this the same problem....

2/27/2014 7:23:37 AM | SETI@home | Started upload of 16au13ac.14495.14382.438086664205.12.174_0_0
2/27/2014 7:23:51 AM | SETI@home | Started upload of 16au13ad.21976.11110.438086664203.12.124_1_0
2/27/2014 7:24:00 AM | SETI@home | Temporarily failed upload of 16au13ac.14495.14382.438086664205.12.174_0_0: connect() failed
2/27/2014 7:24:00 AM | SETI@home | Backing off 2 min 56 sec on upload of 16au13ac.14495.14382.438086664205.12.174_0_0
2/27/2014 7:24:13 AM | SETI@home | Temporarily failed upload of 16au13ad.21976.11110.438086664203.12.124_1_0: connect() failed
2/27/2014 7:24:13 AM | SETI@home | Backing off 3 min 44 sec on upload of 16au13ad.21976.11110.438086664203.12.124_1_0
2/27/2014 7:24:18 AM | | Project communication failed: attempting access to reference site
2/27/2014 7:24:21 AM | | Internet access OK - project servers may be temporarily down.

Yes, that's the problem... the upload server needs a nudge, which should happen shortly. (in an hour or so, when the staff gets in - it's 0720 Berkeley time......)

---

.

Hello, from Albany, CA!...

Still hasn't happened and it's 9am PT.

---

:(

http://fragment1.berkeley.edu/newcricket/grapher.cgi?target=%2Frouter-interfaces%2Finr-211%2Fgigabitethernet6_17;ranges=d;view=Octets

---

Noli tangere circulos meos...

OK, everything on the Server Status Page is off (near enough).

I hope that means the boyz are in the lab and preparing for an orderly restart.

OK, everything on the Server Status Page is off (near enough).

I hope that means the boyz are in the lab and preparing for an orderly restart.

Thanks Richard for the news
Fingers crossed X

---

......and uploads coughing into life.......

---

Bob Smith
Member of Seti PIPPS (Pluto is a Planet Protest Society)
Somewhere in the (un)known Universe?

......and uploads coughing into life.......

Looks like the splitters kept running without inhibition while the SSP was frozen, as they've done before. Thank goodness it didn't last the whole weekend.

......and uploads coughing into life.......

Still unable to upload.
Many of them get to 100%, then just sit there waiting for the final acknowledgment, then time out.

---

Grant
Darwin NT

Works fine here now :) Upload is OK !

---