Out of the fire and into the pit of sulfuric acid. (Feb 19, 2010)


log in

Advanced search

Message boards : Technical News : Out of the fire and into the pit of sulfuric acid. (Feb 19, 2010)

Previous · 1 . . . 4 · 5 · 6 · 7 · 8 · 9 · 10 . . . 15 · Next
Author Message
Profile Bill Walker
Avatar
Send message
Joined: 4 Sep 99
Posts: 3330
Credit: 1,976,649
RAC: 2,295
Canada
Message 972466 - Posted: 20 Feb 2010, 21:59:53 UTC - in response to Message 972462.

I was just going to post this on my Website and everywhere else I could think of, but, what does the staff think?
(Probably not on a weekend)


This might pass under the new Canadian libel laws, but you should get it checked against the US laws.
____________

Profile Link
Avatar
Send message
Joined: 18 Sep 03
Posts: 824
Credit: 1,545,480
RAC: 290
Germany
Message 972474 - Posted: 20 Feb 2010, 22:08:44 UTC
Last modified: 20 Feb 2010, 22:22:05 UTC

And here one from Berlin, Germany:


-----------------------------------------------------------------------------
Routenverfolgung zu boinc2.ssl.berkeley.edu [208.68.240.18]

Über maximal 30 Abschnitte:

0 LinkLap.htw-berlin.de [141.45.217.239]
1 141.45.216.4
2 141.45.2.33
3 141.45.2.17
4 xr-adh1-ge8-4.x-win.dfn.de [188.1.232.177]
5 xr-zib1-te1-1.x-win.dfn.de [188.1.144.21]
6 zr-pot1-te0-0-0-3.x-win.dfn.de [188.1.144.30]
7 de-cix.he.net [80.81.192.172]
8 10gigabitethernet1-2.core1.par1.he.net [72.52.92.89]
9 10gigabitethernet1-3.core1.lon1.he.net [72.52.92.33]
10 10gigabitethernet2-3.core1.nyc4.he.net [72.52.92.77]
11 10gigabitethernet3-1.core1.sjc2.he.net [72.52.92.25]
12 10gigabitethernet3-2.core1.pao1.he.net [72.52.92.69]
13 64.71.140.42
14 * 208.68.243.254
15 boinc2.ssl.berkeley.edu [208.68.240.18]



Berechnung der Statistiken dauert ca. 375 Sekunden...

Quelle zum Abs. Knoten/Verbindung

Abs. Zeit Verl./Ges.= % Verl./Ges.= % Adresse
0 LinkLap.htw-berlin.de [141.45.217.239]
0/ 100 = 0% |
1 10ms 0/ 100 = 0% 0/ 100 = 0% 141.45.216.4
0/ 100 = 0% |
2 27ms 0/ 100 = 0% 0/ 100 = 0% 141.45.2.33
0/ 100 = 0% |
3 22ms 0/ 100 = 0% 0/ 100 = 0% 141.45.2.17
0/ 100 = 0% |
4 17ms 0/ 100 = 0% 0/ 100 = 0% xr-adh1-ge8-4.x-win.dfn.de [188.1.232.177]
0/ 100 = 0% |
5 36ms 0/ 100 = 0% 0/ 100 = 0% xr-zib1-te1-1.x-win.dfn.de [188.1.144.21]
0/ 100 = 0% |
6 30ms 0/ 100 = 0% 0/ 100 = 0% zr-pot1-te0-0-0-3.x-win.dfn.de [188.1.144.30]
0/ 100 = 0% |
7 39ms 0/ 100 = 0% 0/ 100 = 0% de-cix.he.net [80.81.192.172]
0/ 100 = 0% |
8 63ms 0/ 100 = 0% 0/ 100 = 0% 10gigabitethernet1-2.core1.par1.he.net [72.52.92.89]
0/ 100 = 0% |
9 84ms 0/ 100 = 0% 0/ 100 = 0% 10gigabitethernet1-3.core1.lon1.he.net [72.52.92.33]
0/ 100 = 0% |
10 139ms 0/ 100 = 0% 0/ 100 = 0% 10gigabitethernet2-3.core1.nyc4.he.net [72.52.92.77]
0/ 100 = 0% |
11 228ms 0/ 100 = 0% 0/ 100 = 0% 10gigabitethernet3-1.core1.sjc2.he.net [72.52.92.25]
0/ 100 = 0% |
12 222ms 0/ 100 = 0% 0/ 100 = 0% 10gigabitethernet3-2.core1.pao1.he.net [72.52.92.69]
0/ 100 = 0% |
13 216ms 0/ 100 = 0% 0/ 100 = 0% 64.71.140.42
8/ 100 = 8% |
14 210ms 8/ 100 = 8% 0/ 100 = 0% 208.68.243.254
0/ 100 = 0% |
15 203ms 8/ 100 = 8% 0/ 100 = 0% boinc2.ssl.berkeley.edu [208.68.240.18]


Ablaufverfolgung beendet.
-----------------------------------------------------------------------------

____________
.

Kaylie
Send message
Joined: 26 Jul 08
Posts: 39
Credit: 332,100
RAC: 0
United States
Message 972483 - Posted: 20 Feb 2010, 22:17:52 UTC - in response to Message 972466.

I was just going to post this on my Website and everywhere else I could think of, but, what does the staff think?
(Probably not on a weekend)


This might pass under the new Canadian libel laws, but you should get it checked against the US laws.


Your left wing is in that acid.

Profile Gary Charpentier
Volunteer tester
Avatar
Send message
Joined: 25 Dec 00
Posts: 12156
Credit: 6,442,440
RAC: 8,271
United States
Message 972485 - Posted: 20 Feb 2010, 22:18:37 UTC

I wonder if everyone stopped pinging the system if there would be more bandwidth available?
I wonder if everyone quit hitting update if there would be more bandwidth available?

The sooner there is bandwidth the sooner it is fixed.


____________

Profile arkayn
Volunteer tester
Avatar
Send message
Joined: 14 May 99
Posts: 3597
Credit: 47,421,684
RAC: 7,323
United States
Message 972496 - Posted: 20 Feb 2010, 22:30:48 UTC

208.68.243.254 - Whois Information

OrgName: SETIATHOME
OrgID: SETIA
Address: 7 Gauss Way
Address: Room 325A
City: Berkeley
StateProv: CA
PostalCode: 94720
Country: US

NetRange: 208.68.240.0 - 208.68.243.255
CIDR: 208.68.240.0/22
NetName: SETIATHOME
NetHandle: NET-208-68-240-0-1
Parent: NET-208-0-0-0-0
NetType: Direct Assignment
NameServer: ADNS1.BERKELEY.EDU
NameServer: ADNS2.BERKELEY.EDU
Comment:
RegDate: 2006-07-19
Updated: 2006-11-14


Ping has started…

PING 208.68.240.16 (208.68.240.16): 56 data bytes
64 bytes from 208.68.240.16: icmp_seq=0 ttl=49 time=57.414 ms
64 bytes from 208.68.240.16: icmp_seq=1 ttl=49 time=48.683 ms
64 bytes from 208.68.240.16: icmp_seq=2 ttl=49 time=56.586 ms
64 bytes from 208.68.240.16: icmp_seq=3 ttl=49 time=58.014 ms
64 bytes from 208.68.240.16: icmp_seq=4 ttl=49 time=65.436 ms
64 bytes from 208.68.240.16: icmp_seq=5 ttl=49 time=51.773 ms
64 bytes from 208.68.240.16: icmp_seq=6 ttl=49 time=53.666 ms
64 bytes from 208.68.240.16: icmp_seq=7 ttl=49 time=58.524 ms
64 bytes from 208.68.240.16: icmp_seq=8 ttl=49 time=53.451 ms
64 bytes from 208.68.240.16: icmp_seq=9 ttl=49 time=55.335 ms

--- 208.68.240.16 ping statistics ---
10 packets transmitted, 10 packets received, 0.0% packet loss
round-trip min/avg/max/stddev = 48.683/55.888/65.436/4.307 ms


traceroute to 208.68.240.16 (208.68.240.16), 64 hops max, 52 byte packets
1 192.168.1.1 (192.168.1.1) 7.215 ms 4.343 ms 1.556 ms
2 10.247.0.1 (10.247.0.1) 10.844 ms 7.933 ms 8.702 ms
3 24-156-52-17.unassigned.npgco.com (24.156.52.17) 8.953 ms 22.674 ms 9.227 ms
4 24-156-52-1.unassigned.npgco.com (24.156.52.1) 17.261 ms 11.749 ms 13.370 ms
5 24.121.60.4 (24.121.60.4) 26.934 ms 26.141 ms 28.010 ms
6 24.121.123.13 (24.121.123.13) 43.326 ms 38.144 ms 38.889 ms
7 70.96.35.245 (70.96.35.245) 35.876 ms 43.601 ms 47.450 ms
8 tg9-1.cr02.phnzaz87.integra.net (209.63.113.97) 65.158 ms 57.173 ms 40.556 ms
9 tg13-1.cr02.lsancarc.integra.net (209.63.114.33) 44.428 ms 37.632 ms 41.421 ms
10 tg4-1.br01.lsancarc.integra.net (209.63.113.30) 37.671 ms 63.694 ms 41.011 ms
11 he.net.coresite.com (206.223.143.122) 44.517 ms 35.491 ms 47.587 ms
12 10gigabitethernet2-1.core1.lax1.he.net (72.52.92.121) 50.478 ms 44.214 ms 41.888 ms
13 10gigabitethernet1-3.core1.pao1.he.net (72.52.92.21) 54.094 ms 49.056 ms 49.135 ms
14 64.71.140.42 (64.71.140.42) 47.875 ms 52.300 ms 72.056 ms
15 208.68.243.254 (208.68.243.254) 60.072 ms 71.909 ms 76.468 ms
16 * * *
17 * * *
18 * *


It cannot get any further than that particular router and just spins it's wheels.

____________

Kaylie
Send message
Joined: 26 Jul 08
Posts: 39
Credit: 332,100
RAC: 0
United States
Message 972499 - Posted: 20 Feb 2010, 22:32:57 UTC - in response to Message 972224.

Boinc won't ask for work from a project, if the the number of uploads for that project exceeds twice the number of CPU cores,
so until you're down to a handfull of uploads, you'll get no downloads from Seti,
But you can still get work from other projects.

Claggy


If you’re looking for an interesting task to sick your GPU’s on, check out Collatz Conjecture at http://boinc.thesonntags.com/collatz/ They’re called Wondrous Numbers (really).

For the CPU, there are projects to help improve the human condition at World Community Grid. http://www.worldcommunitygrid.org/


This is a unique event. Everyone is, surely, doing everything they have done before. Either this project has suddenly become too popular for its own good, or some external or internal force has dampened its progress. What has happened to jam this up for so long?
I am a proponent of seeking other projects for free resources. I have done so. I wonder what is going on here?

Kaylie
Send message
Joined: 26 Jul 08
Posts: 39
Credit: 332,100
RAC: 0
United States
Message 972505 - Posted: 20 Feb 2010, 22:39:36 UTC - in response to Message 972496.
Last modified: 20 Feb 2010, 22:57:17 UTC

HMM... Funny, everything but the transgers or updates goes through. (Right me a script for that)

Grant (SSSF)
Send message
Joined: 19 Aug 99
Posts: 5701
Credit: 56,527,567
RAC: 48,536
Australia
Message 972513 - Posted: 20 Feb 2010, 22:49:13 UTC - in response to Message 972407.

Simple failure - immediate backoff/retry:


(Direct link)

More complicated failure: connected, upload progress bar to 100%, then retry.


(Direct link)

Yesterday most of my failures were the partial upload ones- sometimes 16%, somtimes 60%, often 100%- all usually after about a minute or so. There were a few that died within a second or so.
Today almost all of the failures occur straight away; Scarecrow's graphs & the network traffic would indicate it's this way for most people. Even less work is getting back than a couple of days ago.
____________
Grant
Darwin NT.

JimDz
Send message
Joined: 18 May 99
Posts: 1
Credit: 8,141,094
RAC: 3,668
United States
Message 972527 - Posted: 20 Feb 2010, 23:36:08 UTC

People.
I have been having problems for about 4 days now on 4 PCs and some ran out of work (prefs set for 3 days of work). Yesterday, WUs are downloading, but not uploading. So at least work is being done. That's an improvement. Patience.
They. Will. Fix. It.

____________

Profile RFP
Avatar
Send message
Joined: 21 Jan 10
Posts: 44
Credit: 29,197
RAC: 0
United States
Message 972540 - Posted: 21 Feb 2010, 0:15:57 UTC

Looks like some of you can at least download. I can not download or upload. I get a message the the inet connection is good, but the servers no workie.
TTFN

Richard Haselgrove
Volunteer tester
Send message
Joined: 4 Jul 99
Posts: 8378
Credit: 46,827,697
RAC: 24,484
United Kingdom
Message 972541 - Posted: 21 Feb 2010, 0:23:43 UTC - in response to Message 972505.

HMM... Funny, everything but the transgers or updates goes through. (write me a script for that)

Comms and ISPs are good.

Two individual servers - Anakin and Bruno - are sick and require attention.

1mp0£173
Volunteer tester
Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 972544 - Posted: 21 Feb 2010, 0:34:27 UTC - in response to Message 972499.

This is a unique event. Everyone is, surely, doing everything they have done before. Either this project has suddenly become too popular for its own good, or some external or internal force has dampened its progress. What has happened to jam this up for so long?
I am a proponent of seeking other projects for free resources. I have done so. I wonder what is going on here?

Actually, it's not that unique.

The basic idea behind BOINC is that projects that would have required an expensive investment in computing can be run using relatively simple hardware, BOINC, and a stack of computers "in the wild" whose excess time is donated.

The problem is that most BOINC projects are not well funded. There is a lot less redundancy and a lot less excess capacity, and sometimes, things break.

Some of those who have been around longer remember some pretty huge outages.

They're unfortunate. If you believe the work should be done, but you can't find the money to fund it very well, this is the result. The outage problem is offset by caching in the BOINC client on your machine.

It's also offset by the fact that everyone can assign 90% of their resources to SETI and 10% to Einstein, or any one of several dozen worthy projects.

____________

Profile RFP
Avatar
Send message
Joined: 21 Jan 10
Posts: 44
Credit: 29,197
RAC: 0
United States
Message 972549 - Posted: 21 Feb 2010, 0:39:13 UTC

If things are working, and I guess we are using our government's deff of the word working, then what is the menaing of these messages I keep getting?

2/20/2010 4:27:55 PM SETI@home update requested by user
2/20/2010 4:27:57 PM SETI@home Sending scheduler request: Requested by user.
2/20/2010 4:27:57 PM SETI@home Requesting new tasks for GPU
2/20/2010 4:27:58 PM Project communication failed: attempting access to reference site
2/20/2010 4:27:59 PM Internet access OK - project servers may be temporarily down.
2/20/2010 4:28:02 PM SETI@home Scheduler request failed: Server returned nothing (no headers, no data)
2/20/2010 4:29:03 PM SETI@home Fetching scheduler list
2/20/2010 4:29:08 PM SETI@home Master file download succeeded
2/20/2010 4:29:13 PM SETI@home Sending scheduler request: To fetch work.
2/20/2010 4:29:13 PM SETI@home Requesting new tasks for GPU
2/20/2010 4:29:14 PM Project communication failed: attempting access to reference site
2/20/2010 4:29:15 PM Internet access OK - project servers may be temporarily down.
2/20/2010 4:29:18 PM SETI@home Scheduler request failed: Server returned nothing (no headers, no data)
2/20/2010 4:30:19 PM SETI@home Sending scheduler request: To fetch work.
2/20/2010 4:30:19 PM SETI@home Requesting new tasks for GPU
2/20/2010 4:30:21 PM Project communication failed: attempting access to reference site
2/20/2010 4:30:22 PM Internet access OK - project servers may be temporarily down.
2/20/2010 4:30:24 PM SETI@home Scheduler request failed: Failure when receiving data from the peer
2/20/2010 4:31:25 PM SETI@home Sending scheduler request: To fetch work.
2/20/2010 4:31:25 PM SETI@home Requesting new tasks for GPU
2/20/2010 4:31:26 PM Project communication failed: attempting access to reference site
2/20/2010 4:31:27 PM Internet access OK - project servers may be temporarily down.
2/20/2010 4:31:30 PM SETI@home Scheduler request failed: Server returned nothing (no headers, no data)
2/20/2010 4:32:31 PM SETI@home Sending scheduler request: To fetch work.
2/20/2010 4:32:31 PM SETI@home Requesting new tasks for GPU
2/20/2010 4:32:32 PM Project communication failed: attempting access to reference site
2/20/2010 4:32:33 PM Internet access OK - project servers may be temporarily down.
2/20/2010 4:32:36 PM SETI@home Scheduler request failed: Server returned nothing (no headers, no data)

madasczik
Avatar
Send message
Joined: 13 May 09
Posts: 12
Credit: 1,693,704
RAC: 0
United States
Message 972575 - Posted: 21 Feb 2010, 1:55:36 UTC

It's kinda working...I got 17,119 through yesterday but not all...

Didn't know about PathPing, it's pretty sweet. Found this convenient gadget: http://whitepapers.zdnet.com/abstract.aspx?docid=1164559

Tracing route to setiboincdata.ssl.berkeley.edu [208.68.240.16]
over a maximum of 30 hops:
0 WOPR-III [192.168.101.245]
1 192.168.101.1
2 10.50.96.1
3 ge3-2-njmnyhubp-rtr1.nj.rr.com [24.168.128.146]
4 cpe-24-29-150-94.nyc.res.rr.com [24.29.150.94]
5 tenge-0-1-0-nwrknjmd-rtr.nyc.rr.com [24.29.119.150]
6 ae-4-0.cr0.nyc30.tbone.rr.com [66.109.6.78]
7 ae-0-0.cr0.nyc20.tbone.rr.com [66.109.6.27]
8 66.109.6.10
9 ae-1-0.pr0.sjc10.tbone.rr.com [66.109.6.137]
10 gige-g5-6.core1.sjc2.he.net [216.218.135.225]
11 10gigabitethernet3-2.core1.pao1.he.net [72.52.92.69]
12 64.71.140.42
13 208.68.243.254
14 setiboincdata.ssl.berkeley.edu [208.68.240.16]

Computing statistics for 350 seconds...
Source to Here This Node/Link
Hop RTT Lost/Sent = Pct Lost/Sent = Pct Address
0 WOPR-III [192.168.101.245]
0/ 100 = 0% |
1 0ms 0/ 100 = 0% 0/ 100 = 0% 192.168.101.1
0/ 100 = 0% |
2 --- 100/ 100 =100% 100/ 100 =100% 10.50.96.1
0/ 100 = 0% |
3 12ms 0/ 100 = 0% 0/ 100 = 0% ge3-2-njmnyhubp-rtr1.nj.rr.com [24.168.128.146]
0/ 100 = 0% |
4 14ms 0/ 100 = 0% 0/ 100 = 0% cpe-24-29-150-94.nyc.res.rr.com [24.29.150.94]
0/ 100 = 0% |
5 17ms 0/ 100 = 0% 0/ 100 = 0% tenge-0-1-0-nwrknjmd-rtr.nyc.rr.com [24.29.119.150]
0/ 100 = 0% |
6 19ms 0/ 100 = 0% 0/ 100 = 0% ae-4-0.cr0.nyc30.tbone.rr.com [66.109.6.78]
0/ 100 = 0% |
7 19ms 0/ 100 = 0% 0/ 100 = 0% ae-0-0.cr0.nyc20.tbone.rr.com [66.109.6.27]
0/ 100 = 0% |
8 95ms 0/ 100 = 0% 0/ 100 = 0% 66.109.6.10
0/ 100 = 0% |
9 98ms 0/ 100 = 0% 0/ 100 = 0% ae-1-0.pr0.sjc10.tbone.rr.com [66.109.6.137]
0/ 100 = 0% |
10 108ms 0/ 100 = 0% 0/ 100 = 0% gige-g5-6.core1.sjc2.he.net [216.218.135.225]
0/ 100 = 0% |
11 108ms 0/ 100 = 0% 0/ 100 = 0% 10gigabitethernet3-2.core1.pao1.he.net [72.52.92.69]
0/ 100 = 0% |
12 108ms 0/ 100 = 0% 0/ 100 = 0% 64.71.140.42
10/ 100 = 10% |
13 111ms 10/ 100 = 10% 0/ 100 = 0% 208.68.243.254
3/ 100 = 3% |
14 110ms 13/ 100 = 13% 0/ 100 = 0% setiboincdata.ssl.berkeley.edu [208.68.240.16]

Trace complete.

Kaylie
Send message
Joined: 26 Jul 08
Posts: 39
Credit: 332,100
RAC: 0
United States
Message 972580 - Posted: 21 Feb 2010, 2:07:41 UTC - in response to Message 972544.

This is a unique event. Everyone is, surely, doing everything they have done before. Either this project has suddenly become too popular for its own good, or some external or internal force has dampened its progress. What has happened to jam this up for so long?
I am a proponent of seeking other projects for free resources. I have done so. I wonder what is going on here?

Actually, it's not that unique.

The basic idea behind BOINC is that projects that would have required an expensive investment in computing can be run using relatively simple hardware, BOINC, and a stack of computers "in the wild" whose excess time is donated.

The problem is that most BOINC projects are not well funded. There is a lot less redundancy and a lot less excess capacity, and sometimes, things break.

Some of those who have been around longer remember some pretty huge outages.

They're unfortunate. If you believe the work should be done, but you can't find the money to fund it very well, this is the result. The outage problem is offset by caching in the BOINC client on your machine.

It's also offset by the fact that everyone can assign 90% of their resources to SETI and 10% to Einstein, or any one of several dozen worthy projects.


If you are saying that this project is too popular I agree. Distributed computing should be distributed. I have always agreed. People should find their red buttons and compute on them. Health, , numbers, forms of matter, where that matter is, where it might go, how the earth evolves, what stars do, where they are (I think everything but safety[and hopefully that somehow]): these are other projects that can use your idle CPU and GPU time.

Certainly there will always be network and hardware issues, caused by both political and financial drives. The crunching must go on: one way or another!

madasczik
Avatar
Send message
Joined: 13 May 09
Posts: 12
Credit: 1,693,704
RAC: 0
United States
Message 972584 - Posted: 21 Feb 2010, 2:19:24 UTC - in response to Message 972580.

This is a unique event. Everyone is, surely, doing everything they have done before. Either this project has suddenly become too popular for its own good, or some external or internal force has dampened its progress. What has happened to jam this up for so long?
I am a proponent of seeking other projects for free resources. I have done so. I wonder what is going on here?

Actually, it's not that unique.

The basic idea behind BOINC is that projects that would have required an expensive investment in computing can be run using relatively simple hardware, BOINC, and a stack of computers "in the wild" whose excess time is donated.

The problem is that most BOINC projects are not well funded. There is a lot less redundancy and a lot less excess capacity, and sometimes, things break.

Some of those who have been around longer remember some pretty huge outages.

They're unfortunate. If you believe the work should be done, but you can't find the money to fund it very well, this is the result. The outage problem is offset by caching in the BOINC client on your machine.

It's also offset by the fact that everyone can assign 90% of their resources to SETI and 10% to Einstein, or any one of several dozen worthy projects.


If you are saying that this project is too popular I agree. Distributed computing should be distributed. I have always agreed. People should find their red buttons and compute on them. Health, , numbers, forms of matter, where that matter is, where it might go, how the earth evolves, what stars do, where they are (I think everything but safety[and hopefully that somehow]): these are other projects that can use your idle CPU and GPU time.

Certainly there will always be network and hardware issues, caused by both political and financial drives. The crunching must go on: one way or another!


Too true, I jumped on GPUGRID today for lack of SETI CUDA jobs. I'm still crunching SETI on CPU, GPUs are cooking some molecules now... :)

1mp0£173
Volunteer tester
Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 972610 - Posted: 21 Feb 2010, 3:21:45 UTC - in response to Message 972580.


If you are saying that this project is too popular I agree.

No, I wasn't saying that at all.

What I said was that the normal way of doing an internet based "x" (whatever it is) involves buying rack space in three or four large data centers, chosen to be relatively near a major exchange point.

You put together a package that will handle 2 to 3 times the anticipated load (database servers, switches and routers, etc.) and you put a complete set in each of the four data centers.

... and you write big checks every single month.

Either that, or buy time on a supercomputer cluster somewhere, and not have to deal with all of the messy volunteer issues.

To do large computing when you don't have that kind of budget, you have to scale something back.

So SETI runs on what they can afford. They don't have spares in depth, or excess capacity. There are lots of places where just one failure can take the server component down, or cause it to run badly, for an extended period.

It has happened before, and it will happen again.

As a 40-year veteran of the computer industry (more than 20 in datacomm and networking) SETI@Home (and BOINC) isn't a study in reliability, it is a study in fault tolerance and resiliency.

... and if we can stop worrying about the moment-to-moment performance, it does an amazingly good job.

If you ask me, the problem isn't that SETI@Home is too popular, it is that the average computer user is comparing it to eBay or Amazon.com. If they went down it'd make the nightly news -- but if SETI@Home is down, the clients keep crunching and work gets reported when things come back.
____________

zoom314
Avatar
Send message
Joined: 30 Nov 03
Posts: 45811
Credit: 36,431,489
RAC: 6,677
Message 972630 - Posted: 21 Feb 2010, 4:39:03 UTC - in response to Message 972610.

Yeah, I agree Ned, If Seti@Home itself went down, Except for the users, The media would probably not even notice or care. :(
____________

1mp0£173
Volunteer tester
Send message
Joined: 3 Apr 99
Posts: 8423
Credit: 356,897
RAC: 0
United States
Message 972637 - Posted: 21 Feb 2010, 5:03:56 UTC - in response to Message 972630.
Last modified: 21 Feb 2010, 5:18:31 UTC

Yeah, I agree Ned, If Seti@Home itself went down, Except for the users, The media would probably not even notice or care. :(

What I meant was that Amazon.com being down would be a notable event, that would have repercussions beyond the inability to see a web site.

It would lower revenue at Amazon, and certainly orders would be up at BN.com.

A lot of money would not go to Amazon (and their merchant processor) but Barnes&Noble (and their bankers) would be happy. Some customers would switch.

It would be newsworthy.

In contrast, if SETI@Home is down, computers that are turned on anyway are sending a few hundred bytes of data toward servers that are not up, and if it stays down, those same computers will retry until they come back up.

... a few people at Berkeley would work very hard, and a few of the volunteers would waste a whole lot of angst over something they cannot control. [Edit: and a lot of processing would still get done during the outage]

That is not newsworthy.
____________

Rick
Avatar
Send message
Joined: 3 Dec 99
Posts: 79
Credit: 11,486,227
RAC: 0
United States
Message 972643 - Posted: 21 Feb 2010, 5:22:50 UTC - in response to Message 972637.

That is not newsworthy.


Translated - Nobody ever lost a dime because SETI wasn't available.
____________

Previous · 1 . . . 4 · 5 · 6 · 7 · 8 · 9 · 10 . . . 15 · Next

Message boards : Technical News : Out of the fire and into the pit of sulfuric acid. (Feb 19, 2010)

Copyright © 2014 University of California