@
Berkeley Open Infrastructure
for Network Computing
|
|
|
Berkeley Open Infrastructure
for Network Computing
|
| Home | Help | Status | Forums | Glossary | Account
|
| log in |
Message boards : Wish list : Longer communication deferrals while in port
| Author | Message |
|---|---|
 OdysseusVolunteer tester  Send message Joined: 20 Aug 06
Canada
Team Canada Credit: 1,777.8 RAC: 0.00
Joined: Aug 20, 2006
Verified: Sep 30, 2009 Dubloons: 3 Pieces of Eight: 5 Punishment: Aztec curse | |
|
I’ve noticed that me hosts generate frequent traffic tryin’ t’ get work, often wit’ only a few minutes between requests. No big deal, but it clutters up th’ logs. Moreover th’ crew are beginnin’ t’ resent constant watch rotations while they might as well be on shore-leave … | |
ID: 8115 | Rating: 0 | rate:
 /   Reply
Quote
| |
|
Pooh Bear 27 Volunteer tester  Send message Joined: 7 Sep 08
United States
The Knights Who Say Ni! Credit: 7,851.4 RAC: 0.00
Joined: Sep 7, 2008
Verified: Jul 4, 2011 Dubloons: 3 Pieces of Eight: 6 Punishment: Misfit | |
|
If I understand it correctly, it's an algorithm built into the BOINC software, and the projects cannot influence it. | |
| ID: 8117 | Rating: 0 | rate:
/ Reply
Quote
| |
|
Ageless Chief Petty Officer Volunteer tester  Send message Joined: 20 Jul 04
Netherlands
Machinae Supremacy Credit: 1,295.9 RAC: 0.00
Joined: Jul 20, 2004
Verified: Jul 9, 2011 Dubloons: 3 Pieces of Eight: 7 Punishment: Cat o' Nine Tails | |
|
It's both. | |
| ID: 8119 | Rating: 0 | rate:
/ Reply
Quote
| |
|
Wormholio Captain  Send message Joined: 6 Jun 04
United States
Away Credit: 4,009.8 RAC: 0.00
Joined: Jun 6, 2004
Verified: Mar 13, 2008 Dubloons: 3 Pieces of Eight: 10 Punishment: Aztec curse | |
|
Ageless wrote:
Except that I think the meanings are slightly different. What the Mad Toms are talking about is a signal that they don't have to come back for a while. A lower bound on the delay. Setting next_rpc_delay tells the client an upper bound on the delay -- you must try to be back for more work within this time span. It's to allow canceling jobs if needed, and could also be used for "low latency" tasks. So I think we are talking about two different things. But I've set next_rpc_delay to 1 hour just to try it out. That may make it worse for Tom, not better, but we'll see what happens. ____________ -- Eric Myers "Education is not the filling of a pail, but the lighting of a fire." -- William Butler Yeats | |
| ID: 8120 | Rating: 0 | rate:
/ Reply
Quote
| |
|
Pooh Bear 27 Volunteer tester Send message Joined: 7 Sep 08
United States
The Knights Who Say Ni! Credit: 7,851.4 RAC: 0.00
Joined: Sep 7, 2008
Verified: Jul 4, 2011 Dubloons: 3 Pieces of Eight: 6 Punishment: Misfit | |
|
Just testing this, with a request for work, it then took me to 1 minute, 2 minutes, 15 minutes. So I think the setting you set will only allow for a maximum of 1 hour, as I had one that was on a 4 hour one that I started testing this with. | |
| ID: 8121 | Rating: 0 | rate:
/ Reply
Quote
| |
|
Odysseus Volunteer tester Send message Joined: 20 Aug 06
Canada
Team Canada Credit: 1,777.8 RAC: 0.00
Joined: Aug 20, 2006
Verified: Sep 30, 2009 Dubloons: 3 Pieces of Eight: 5 Punishment: Aztec curse | |
|
I’ve seen 24-hour deferrals from other projects without work for me host. BOINC v5.10.x doesn’t log deferral times th’ way some older versions did, so I can nah check back, but I’ve certainly observed it recently from Docking@home (which has no current PPC app—since ’tis only once a day I let me G5 nag instead o’ suspendin’ it) ’n from SETI@home Beta (fer most o’ this week I couldn’t get AstroPulse work). I dunno wha’ circumstances trigger this behaviour. I ’ave also observed that repeated connection failures can prompt one-week deferrals. | |
| ID: 8122 | Rating: 0 | rate:
/ Reply
Quote
| |
|
Ageless Chief Petty Officer Volunteer tester Send message Joined: 20 Jul 04
Netherlands
Machinae Supremacy Credit: 1,295.9 RAC: 0.00
Joined: Jul 20, 2004
Verified: Jul 9, 2011 Dubloons: 3 Pieces of Eight: 7 Punishment: Cat o' Nine Tails | |
|
Mine did: | |
| ID: 8123 | Rating: 0 | rate:
/ Reply
Quote
| |
code { dwarf } Send message Joined: 20 Jan 07
Poland
Credit: 709.5 RAC: 0.00
Joined: Jan 20, 2007
Verified: NEVER Dubloons: 3 Pieces of Eight: 2 | |
|
Ageless wrote:
Thanks. This is what I suggested, next_rpc_delay is a maximum delay. But I didn't know that we now had a minimum, request_delay. This is what Tom asked for, a "minimum" for the delay interval before the next contact. Let's try it out. I'm setting it to 24 hours (and the upper bound at 10 days). | |
| ID: 8124 | Rating: 0 | rate:
/ Reply
Quote
| |
|
Pooh Bear 27 Volunteer tester Send message Joined: 7 Sep 08
United States
The Knights Who Say Ni! Credit: 7,851.4 RAC: 0.00
Joined: Sep 7, 2008
Verified: Jul 4, 2011 Dubloons: 3 Pieces of Eight: 6 Punishment: Misfit | |
Let's try it out. I'm setting it to 24 hours (and the upper bound at 10 days). Yup, looks like you got it. I am getting larger hour push backs than ever before. Remember to change it back once you get new work in. ____________ | |
| ID: 8126 | Rating: 0 | rate:
/ Reply
Quote
| |
|
Ageless Chief Petty Officer Volunteer tester Send message Joined: 20 Jul 04
Netherlands
Machinae Supremacy Credit: 1,295.9 RAC: 0.00
Joined: Jul 20, 2004
Verified: Jul 9, 2011 Dubloons: 3 Pieces of Eight: 7 Punishment: Cat o' Nine Tails | |
|
After today's initial 1 minute deferrals growing slightly, eventually reaching 3 hour+ deferrals, it looks like it's going through a loop: | |
| ID: 8128 | Rating: 0 | rate:
/ Reply
Quote
| |
Message boards : Wish list : Longer communication deferrals while in port
| Home | Help | Status | Forums | Glossary | Account
|
