Message boards :
Graphics cards (GPUs) :
Initial Replication = 2
Message board moderation
| Author | Message |
|---|---|
|
Send message Joined: 11 Jul 09 Posts: 1639 Credit: 10,159,968,649 RAC: 351 Level ![]() Scientific publications ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]()
|
but minimum quorum = 1. Is this an implicit recognition that too many tasks are failing? |
|
Send message Joined: 5 Jan 09 Posts: 670 Credit: 2,498,095,550 RAC: 0 Level ![]() Scientific publications ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]()
|
I think Richard deserves an answer. I also, would like to know why you're sending these twice. Radio Caroline, the world's most famous offshore pirate radio station. Great music since April 1964. Support Radio Caroline Team - Radio Caroline |
GDFSend message Joined: 14 Mar 07 Posts: 1958 Credit: 629,356 RAC: 0 Level ![]() Scientific publications ![]() ![]() ![]() ![]() ![]() |
Every job is sent originally with initial replication 1. However, if it does not get back within some time, replication is set to two and sent to fast host. We have been testing this for just few days now. gdf |
|
Send message Joined: 11 Jul 09 Posts: 1639 Credit: 10,159,968,649 RAC: 351 Level ![]() Scientific publications ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]() ![]()
|
WU 1087235 is interesting. #1 sent 11 Jan 2010 21:08:58 UTC, deadline 16 Jan 2010 21:08:58 UTC #2 sent 15 Jan 2010 22:36:09 UTC (before #1 timed out). Presumably this is the new test 'Replication = 2' #3 sent 16 Jan 2010 21:11:34 UTC (clearly a routine replacement for #1, which had just timed out). In the end, all three came back, although #1 was late - yet still got the 'fast return' bonus set by #2 - which I'm pleased to say was mine. If you've already sent #2, do you need #3 as well? |
GDFSend message Joined: 14 Mar 07 Posts: 1958 Credit: 629,356 RAC: 0 Level ![]() Scientific publications ![]() ![]() ![]() ![]() ![]() |
no, we will add a check for this. gdf |
©2025 Universitat Pompeu Fabra