Agreed, as you said they seem to have very limited resources to dedicate on the programming front. That along with apparently looking for a new niche to fill in the protein modeling world lead to the long project downtime after they completed the last major science run and caused a lot of people to 'wander' off.
Of course, the titanic memory requirements for part of the current simulation and the other problems makes it hard for folks with older hosts to participate as well.
Indeed -- but the project seems to have gotten back on track. Perhaps they should go looking for folks about to wander off from climate with their 5+ day outage and no status update since Wednesday.
Regarding the memory hit -- true enough -- though I don't put new computers online with less than 1G these days. But I suppose it might be hard to run a Windows 9x host with Predictor....
Quote:
Agreed, as you said they seem to have very limited resources to dedicate on the programming front. That along with apparently looking for a new niche to fill in the protein modeling world lead to the long project downtime after they completed the last major science run and caused a lot of people to 'wander' off.
Of course, the titanic memory requirements for part of the current simulation and the other problems makes it hard for folks with older hosts to participate as well.
I was able to track down some more information on the Climate Prediction status -- though this was not on their home page.
They have been working on the getting things back up and running since last week and this apparently involved new drives (doesn't it always). As has happened with that project a few times in the past, one of their primary support folks (who has another life it seems), has been trying to do the admin remotely from Philadelphia (the servers are in the UK).
A news update from Carl who is in Philadelphia:
Quote:
Sorry, I've been working 'round the clock' (remotely from Philly!) trying to get things going. Tolu & Milo put new big SCSI drives in the server, and I'm getting the database back on it. Probably Monday afternoon or evening. You can keep running; trickles will "catch up" once the database is back after a few days. Uploads will be OK as far as uploading the zip files goes, but the reporting status 'Done' to server (and new workunits of course) will be delayed. Thanks for bearing with us and our low-budget server (which is about 1/20th of what Einstein@home pays for their hardware -- and they still have downtime too! Wink
So the end is in sight. Many thanks to Carl for working through the weekend, and to all our crunchers for their patience and good humour on the forum.
If you look at the boincstats
)
If you look at the boincstats graph, Predictor stalled for almost a year and only appears to've recently started back up at any meaningful rate.
http://www.boincstats.com/stats/project_graph.php?pr=pah
Right, and in the process
)
Right, and in the process lost pretty much all of its previous contributor base.
Agreed, as you said they seem
)
Agreed, as you said they seem to have very limited resources to dedicate on the programming front. That along with apparently looking for a new niche to fill in the protein modeling world lead to the long project downtime after they completed the last major science run and caused a lot of people to 'wander' off.
Of course, the titanic memory requirements for part of the current simulation and the other problems makes it hard for folks with older hosts to participate as well.
Alinator
Indeed -- but the project
)
Indeed -- but the project seems to have gotten back on track. Perhaps they should go looking for folks about to wander off from climate with their 5+ day outage and no status update since Wednesday.
Regarding the memory hit -- true enough -- though I don't put new computers online with less than 1G these days. But I suppose it might be hard to run a Windows 9x host with Predictor....
I was able to track down some
)
I was able to track down some more information on the Climate Prediction status -- though this was not on their home page.
They have been working on the getting things back up and running since last week and this apparently involved new drives (doesn't it always). As has happened with that project a few times in the past, one of their primary support folks (who has another life it seems), has been trying to do the admin remotely from Philadelphia (the servers are in the UK).
A news update from Carl who is in Philadelphia:
Quote:
Sorry, I've been working 'round the clock' (remotely from Philly!) trying to get things going. Tolu & Milo put new big SCSI drives in the server, and I'm getting the database back on it. Probably Monday afternoon or evening. You can keep running; trickles will "catch up" once the database is back after a few days. Uploads will be OK as far as uploading the zip files goes, but the reporting status 'Done' to server (and new workunits of course) will be delayed. Thanks for bearing with us and our low-budget server (which is about 1/20th of what Einstein@home pays for their hardware -- and they still have downtime too! Wink
So the end is in sight. Many thanks to Carl for working through the weekend, and to all our crunchers for their patience and good humour on the forum.