UD Stats Update

[AtomicMoose]

Dear members,

We forgot to mention that we continue to work diligently at repairing the Team Stats figures. The problem appears to be related to a corrupted table, which just so happens to host the team statistics; the data is ok - it's just the table that seems to be problematic. To fix, we are going through a process of 1) exporting all team stats data from the corrupted table, 2) recreating a new team stats table, 3) importing team stats data back into new table.

The explanation above sounds rather simplistic, but it's very intensive due to the enormity of the data contained in this table. Just the exercise of moving the data around via the export/import is very time consuming - so we beg for your continued patience. Thank you very much.

From GRID.org

 

[Carnival Forces]

good golly, Miss Molly!

thanks for the update, moose.

 

[Darth Bagel]

I was wondering what the hell happened to the stats. My personal stats haven't updated since the 5th, also. Do they not update on weekends or something?

 

[Kvothe]

Grid updates the UD stats daily, and Hardfolding updates based off of the daily updates. It seems this is just the continuation of all the other hardware/software issues they've been having, so hopefully it will iron itself out soon.

The good thing is that all stats are still in tact, so once things settle out, all the information should be there.

 

[Flying Fox]

To all that are still hanging in there for UD, the Cancer (LigandFit) project is still waiting for data so they are letting us work on duplicates (may be they can use it double-validate the stuff). If you would like to work on something more "useful", you need to change to the Human Proteome (Rosetta) project. Beware you may not be able to change all of your boxen though, as Rosetta's memory and CPU requirements are significantly heavier than the Cancer project.

 

[Darth Bagel]

I was wondering why all my systems have switched over to Rosetta.

 

[Kvothe]

It appears that Grid.org is getting a little closer to having their stats back up, however there is still no word on when we can expect to see more LigandFit Cancer units.

 

[Flying Fox]

And with the latest update there has been great confusion whether they are going to start rejecting results that take more than 3 days to compute. If that's the case, my Pentium 200 may have to retire and my Celeron 1.1@1.47GHz may run into trouble with one of the big Rosettas (last one I had took 90 hours!)...

Seriously, may be we need to re-evaluate our position in this project?

 

[Carnival Forces]

And with the latest update there has been great confusion whether they are going to start rejecting results that take more than 3 days to compute. If that's the case, my Pentium 200 may have to retire and my Celeron 1.1@1.47GHz may run into trouble with one of the big Rosettas (last one I had took 90 hours!)...

Seriously, may be we need to re-evaluate our position in this project?

i already pulled out my (brief, competition-inspired) investment in the project

down time = lost time.

 

[Rustedimpala]

And with the latest update there has been great confusion whether they are going to start rejecting results that take more than 3 days to compute. If that's the case, my Pentium 200 may have to retire and my Celeron 1.1@1.47GHz may run into trouble with one of the big Rosettas (last one I had took 90 hours!)...

Seriously, may be we need to re-evaluate our position in this project?

Man.... this sucks. I run UD on my laptop, it takes weeks to finish because its not always on. Does this mean there is no point to me doing UD on my laptop?

 

[Carnival Forces]

Man.... this sucks. I run UD on my laptop, it takes weeks to finish because its not always on. Does this mean there is no point to me doing UD on my laptop?

if i were you i'd
A) Start leaving my laptop on all the time (i already do)
and
B) start running FAH on it on timless tinkers if you can't

 

[Kvothe]

Yes, it certainly looks like they've gone to crediting for everything to more of a round robin approach, if you're lucky enough to get a rosetta unit first, and get it turned in first, you get the credit. Anyone else that gets the duplicate of that unit gets nothing unless they fall within the redundant quota.

Now I know my P3 can't hold a candle to just about any rig, but it still gets done with stuff in 2 days. I can only hope that once they get the new cancer data loaded and spread out, this will not be an issue.

 

[Flying Fox]

Discussions on the latest news had someone suggested a seemingly more reasonable idea. For most people the Cancer WUs should be ok, assuming everyone is on the latest WUs. All the old WUs have been "disabled" and will not receive credit. For Rosetta it will be another story, but then again you wouldn't be crunching Rosettas on your P200 right? I know I don't.

The optimistic side of me also agrees with fingerle:

Also note that once we move to the latest version of the grid software, this will no longer be an issue. When we have all of the results, the job will be marked "suspended" which means no more workunits will be dispatched, but we will still credit any returned workunits. Unfortunately, that same behavior just does not exist in the version we are currently running

So we just have to hang in there until the new version comes, if I understand this correctly.

Feel free to correct me.