I sent him a pm 11 hours ago or so. no response

I sent him a pm 11 hours ago or so. no response

Did you write in English or Spanish ? http://en.wikipedia.org/wiki/Chile#Language

Back to the main topic. hostid=4000000 may arrive today, but I think tomorrow is more likely.


3994999 Created 14 Nov 2007 9:56:11 UTC
3995000 Created 14 Nov 2007 9:57:39 UTC
3995001 Created 14 Nov 2007 9:58:48 UTC

3995010 Created 14 Nov 2007 10:06:34 UTC

3995050 Created 14 Nov 2007 10:36:21 UTC

3995090 Created 14 Nov 2007 11:05:34 UTC

3995099 Created 14 Nov 2007 11:15:39 UTC
3995100 Created 14 Nov 2007 11:15:46 UTC

3995111 Created 14 Nov 2007 11:28:12 UTC

---

Sir Arthur C Clarke 1917-2008

[quote]This user http://setiathome.berkeley.edu//hosts_user.php?userid=8481345 seems to have a lot of computers with only 1 WU each. ...

There has to be a problem with the project surely?

But the majority of them with no tasks or credit. Surely all these non-productive hosts are taking up space on the servers!

I will agree with the guess that those old Katmai P-III CPUs are part of a computer lab where all the machines get a full new 'reinstall' on each reboot.

I would also guess that the s@h WUs never get a chance to finish during one teaching session before the machines get wiped again...

Perhaps another Boinc enhancement could be to offer short duration WUs for such machines? Or even to firstly offer short WUs until a machine has demonstrated that it can return WUs and then steadily increase the WU size for whatever is most useful?


As for the database loading, I would expect those few hosts to be a small drop in the bits'n'bytes when compared to the extra loading due to people keeping a large WU cache of days...

Happy crunchin',
Martin

---

See new freedom: Mageia Linux
Take a look for yourself: Linux Format
The Future is what We all make IT (GPLv3)

The one I noticed with multiple id's where each host has 0 WUs is less of a problem than one where each id downloads one or more WUs then trashes it. I'm not sure if that is what is happening or not.

Update on id generation, I looked back a little to try to improve my prediction:

3980000 Created 7 Nov 2007 14:14:29 UTC

3981000 Created 7 Nov 2007 23:22:32 UTC

3982000 Created 8 Nov 2007 9:52:30 UTC

3983000 Created 8 Nov 2007 19:05:59 UTC

3984000 Created 9 Nov 2007 6:09:54 UTC

3985000 Created 9 Nov 2007 15:46:47 UTC

3986000 Created 10 Nov 2007 0:43:14 UTC

3987000 Created 10 Nov 2007 14:28:54 UTC

3988000 Created 11 Nov 2007 2:05:11 UTC

3989000 Created 11 Nov 2007 16:38:22 UTC

3990000 Created 12 Nov 2007 4:25:13 UTC

3991000 Created 12 Nov 2007 16:04:37 UTC

3992000 Created 12 Nov 2007 22:49:47 UTC

3993000 Created 13 Nov 2007 8:54:57 UTC

3994000 Created 13 Nov 2007 21:39:19 UTC

3995000 Created 14 Nov 2007 9:57:39 UTC

...

3995300 Created 14 Nov 2007 13:35:21 UTC

3995339 Created 14 Nov 2007 13:59:18 UTC
3995340 Created 14 Nov 2007 14:00:22 UTC

At the current rate of id creation, I think # 4000000 will arrive late on 16 Nov (UTC) or early on 17 Nov (UTC). I expect someone with better math skills than me could make a better prediction than that.

3995555 Created 14 Nov 2007 16:10:21 UTC

...
I will agree with the guess that those old Katmai P-III CPUs are part of a computer lab where all the machines get a full new 'reinstall' on each reboot.

I would also guess that the s@h WUs never get a chance to finish during one teaching session before the machines get wiped again
...

I had a reply to the PM that I sent to Wayne State College, thanking me for the heads-up. They have cleaned-up most of their duplicate hostid's.

Update on the CountUp to # 400000.

3995000 Created 14 Nov 2007 9:57:39 UTC

3995500 Created 14 Nov 2007 15:42:52 UTC

3996000 Created 14 Nov 2007 19:22:03 UTC

3996500 Created 15 Nov 2007 0:36:50 UTC

3997000 Created 15 Nov 2007 8:32:25 UTC
...
3997250 Created 15 Nov 2007 12:14:47 UTC

---

Sir Arthur C Clarke 1917-2008

Just found another one that has some sort of problem. It looks like they are not holding on to the tasks for long, but it must be creating extra unnecessary load on the servers. http://setiathome.berkeley.edu//hosts_user.php?userid=8515613.

Update on host creation:

3997500 Created 15 Nov 2007 15:03:26 UTC

Just found another one that has some sort of problem. It looks like they are not holding on to the tasks for long, but it must be creating extra unnecessary load on the servers. http://setiathome.berkeley.edu//hosts_user.php?userid=8515613.

Update on host creation:

3997500 Created 15 Nov 2007 15:03:26 UTC

More Updates:

3998000 Created 15 Nov 2007 20:12:19 UTC

3998500 Created 16 Nov 2007 2:31:09 UTC

------

Just spotted another user with multiple id's http://setiathome.berkeley.edu//hosts_user.php?userid=8737510.
138 hostid's created since 1 Nov with 1 or 2 tasks each. 0.00 credits for each host! Any Chinese speakers want to send this one a PM ?

# 4000000 should arrive today (UTC).

3998999 Created 16 Nov 2007 10:38:17 UTC
3999000 Created 16 Nov 2007 10:39:27 UTC

3999100 Created 16 Nov 2007 11:47:45 UTC

3999121 Created 16 Nov 2007 11:59:33 UTC
3999122 Created 16 Nov 2007 12:02:26 UTC

[EDIT]
3999200 Created 16 Nov 2007 12:59:51 UTC
3999201 Created 16 Nov 2007 13:00:27 UTC

Looks like about 8 hours to go at current rate.

---

Sir Arthur C Clarke 1917-2008

# 4000000 should arrive today (UTC).

3998999 Created 16 Nov 2007 10:38:17 UTC
3999000 Created 16 Nov 2007 10:39:27 UTC

3999100 Created 16 Nov 2007 11:47:45 UTC

3999121 Created 16 Nov 2007 11:59:33 UTC
3999122 Created 16 Nov 2007 12:02:26 UTC

[EDIT]
3999200 Created 16 Nov 2007 12:59:51 UTC
3999201 Created 16 Nov 2007 13:00:27 UTC

Looks like about 8 hours to go at current rate.

3999300 Created 16 Nov 2007 13:51:49 UTC

3999400 Created 16 Nov 2007 14:46:38 UTC

3999500 Created 16 Nov 2007 15:49:31 UTC

I'd expected a post with the link to #4000000.

Ah well here it is: 4000000

---

I'd expected a post with the link to #4000000.

Ah well here it is: 4000000

owned by H.K, it's an AMD64 X2 3800 and has 13 results. 9 in progress, and 4 computation errors. LOL

He's using Boinc version.

<core_client_version>5.5.0</core_client_version>

I'd expected a post with the link to #4000000.

Ah well here it is: 4000000

owned by H.K, it's an AMD64 X2 3800 and has 13 results. 9 in progress, and 4 computation errors. LOL

He's using Boinc version.

<core_client_version>5.5.0</core_client_version>

And he/she is using science app

Version info: Windows SSE2 32-bit V5.15 'Chicken Good!' (R-1.2|+fversion|xW)

From memory, that's about 15 months out-of-date for both BOINC and app?

Tony, you probably know how good the SSE2 is on an AMD64 X2 3800?

His/her other two machines both use the v5.5.0 R-1.2 combination, and have got the validate errors to match.

Unfortunately I was away from my computer yesterday evening. I missed the arrival of hostid=4000000.

It arrived at 16 Nov 2007 20:34:22 UTC

hostid 4000000
Owner H.K
Created 16 Nov 2007 20:34:22 UTC
Total Credit 0
Recent average credit 0.00
CPU type AuthenticAMD
AMD Athlon(tm) 64 X2 Processor 3800+
Number of CPUs 2
Operating System Microsoft Windows XP
Home Edition, Service Pack 2, (05.01.2600.00)

hostid 3999999
Owner dietcoke1975
Created 16 Nov 2007 20:34:18 UTC
Total Credit 0
Recent average credit 0.00
CPU type GenuineIntel
Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz [x86 Family 6 Model 15 Stepping 11]
Number of CPUs 4
Operating System Microsoft Windows XP
Professional Edition, Service Pack 2, (05.01.2600.00)

Unfortunately I was away from my computer yesterday evening. I missed the arrival of hostid=4000000.

It arrived at 16 Nov 2007 20:34:22 UTC

hostid 4000000
Owner H.K
Created 16 Nov 2007 20:34:22 UTC
Total Credit 0
Recent average credit 0.00
CPU type AuthenticAMD
AMD Athlon(tm) 64 X2 Processor 3800+
Number of CPUs 2
Operating System Microsoft Windows XP
Home Edition, Service Pack 2, (05.01.2600.00)

hostid 3999999
Owner dietcoke1975
Created 16 Nov 2007 20:34:18 UTC
Total Credit 0
Recent average credit 0.00
CPU type GenuineIntel
Intel(R) Core(TM)2 Quad CPU Q6600 @ 2.40GHz [x86 Family 6 Model 15 Stepping 11]
Number of CPUs 4
Operating System Microsoft Windows XP
Professional Edition, Service Pack 2, (05.01.2600.00)

Well, at least dietcoke is using up-to-date stock everything, so he/she has some potentially valid pendings. But may only be using two cores of that quaddy. Or maybe the other cores are doing other work - just look at that project list!

As other have noticed # 4000000 has so far only returned 4 Compute error tasks. - exit code -1073741819 (0xc0000005)

The owner is in Japan.

3999999 is in Italy, and has got it's first granted credit 63.98

---

Sir Arthur C Clarke 1917-2008