so given the fact that each file is currently of the size of 20 Mb, if the factor 3 I recall is correct, we can only cluster 300 root files at a time (or 200 to be safe). ciao ric P.S. running on all data will take a lot more, but at least it will not fail, do I understand correctly? On Thu, 31 Oct 2002, Yury G. Kolomensky wrote: > ROOT should have no problem reading or writing files as long as they > are less than 2 GB. Making a few large files has many benefits -- > fewer system calls, fewer parallel jobs --> better CPU/wallclock > ratio. Larger files can also be directly backed up in mstore -- which > you probably want to do at some point. > > If you are going to run a skim job to produce large files -- I would > again encourage you to convert your trees to non-split mode. I see > that you now have 114 branches. This is very bad for I/O if your > typical job reads most of them for every event. > > Yury > > On Thu, Oct 31, 2002 at 10:54:29AM -0800, Riccardo Faccini wrote: > > Hi Alessio, > > each of the skimmed root files you would produce would be 300 times larger > > than the original one. Is this something we can handle? I do not know the > > answer... > > thanks > > ric > > > > On Thu, 31 Oct 2002, Alessio Sarti wrote: > > > > > I'm proposing to having skimmed rootfiles (just ~10) obtained running > > > against a splitted chain. > > > > > > Let say: > > > I start with a chain with 9000 rootfiles. > > > I split it up in 10 pieces. > > > I run 10 skimming jobs on 900 rootfiles each. > > > I'm producing just 10 output files. > > > And then you can safely just run against them! > > > what do you thyink about that? > > > > > > Alessio > > > > > > ______________________________________________________ > > > Alessio Sarti Universita' & I.N.F.N. Ferrara > > > tel +39-0532-781928 Ferrara > > > roma +39-06-49914338 > > > SLAC +001-650-926-2972 > > > > > > "... e a un Dio 'fatti il culo' non credere mai..." > > > (F. De Andre') > > > > > > "He was turning over in his mind an intresting new concept in > > > Thau-dimensional physics which unified time, space, magnetism, gravity > > > and, for some reason, broccoli". (T. Pratchett: "Pyramids") > > > > > > On Thu, 31 Oct 2002, Riccardo Faccini wrote: > > > > > > > Sorry Alessio, > > > > I am not sure I understand your proposal: you suggest to produce the > > > > reduced root files and make chains over them? But these root file are only > > > > 3 times smaller than the default ones, how can you claim we will need more > > > > than ten times less > > > > jobs? > > > > thanks > > > > ric > > > > > > > > > > > > On Thu, 31 Oct 2002, Alessio Sarti wrote: > > > > > > > > > Hi all, > > > > > I propose another workaround! > > > > > > > > > > I've produced the chains. > > > > > I can split them and run the skim job that produces in lees thatn 2 hours > > > > > 6-10 job contaiing ALL the generic / data/ cocktail info in such a way > > > > > that we donot longer rely on hundreds of job running..... > > > > > > > > > > What do you think about that? > > > > > Alessio > > > > > > > > > > ______________________________________________________ > > > > > Alessio Sarti Universita' & I.N.F.N. Ferrara > > > > > tel +39-0532-781928 Ferrara > > > > > roma +39-06-49914338 > > > > > SLAC +001-650-926-2972 > > > > > > > > > > "... e a un Dio 'fatti il culo' non credere mai..." > > > > > (F. De Andre') > > > > > > > > > > "He was turning over in his mind an intresting new concept in > > > > > Thau-dimensional physics which unified time, space, magnetism, gravity > > > > > and, for some reason, broccoli". (T. Pratchett: "Pyramids") > > > > > > > > > > On Thu, 31 Oct 2002, Daniele del Re wrote: > > > > > > > > > > > > > > > > > Hi Yury, > > > > > > > > > > > > one example is > > > > > > > > > > > > ~daniele/scra/newchains_1030/data-2 > > > > > > > > > > > > and the tipical message is > > > > > > > > > > > > Error in <TFile::TFile>: file /nfs/farm/babar/AWG18/ISL/sx-080702/data/2000/output/outputdir/AlleEvents_2000_on-1095.root does not exist > > > > > > > > > > > > on AWG8 this pathology happened just few times when there were >~300 jobs > > > > > > reading the same disk if I remember correctly. > > > > > > > > > > > > Do you know which is the difference between AWG8 and AWG18? > > > > > > > > > > > > My proposal is to split things on different disks, if possible. > > > > > > > > > > > > Thanks a lot, > > > > > > > > > > > > Daniele > > > > > > > > > > > > On Thu, 31 Oct 2002, Yury G. Kolomensky wrote: > > > > > > > > > > > > > Hi Daniele, > > > > > > > > > > > > > > do you have an example of a log file for these jobs ? I do not know > > > > > > > exactly what servers these disks have been installed on, but we > > > > > > > noticed in E158, where most of the data were sitting on one > > > > > > > (relatively slow) server, jobs were limited by I/O throughput to about > > > > > > > 2 MB/sec. This limit comes from the random access pattern that split > > > > > > > ROOT trees provide. If your job is sufficiently fast, you can saturate > > > > > > > I/O limit quite quickly -- with 2-3 jobs. If you submit too many jobs > > > > > > > (tens or even hundreds), the server will thrash to the point that the > > > > > > > clients will receive NFS timeouts. ROOT usually does not like that -- > > > > > > > you may see error messages in the log file about files not found (when > > > > > > > the files are actually on disk), or about problems uncompressing > > > > > > > branches. These are usually more severe on Linux clients, where the > > > > > > > NFS client implementation is not very robust.. > > > > > > > > > > > > > > There are several ways to cope with this problem: > > > > > > > > > > > > > > 1) Submit fewer jobs at one time. I would not submit more than 10 > > > > > > > I/O-limited jobs in parallel. > > > > > > > 2) Place your data on different servers. That means, different sulky > > > > > > > servers is best. Even if you are on the same sulky server but split > > > > > > > your data onto different partitions, you still get the benefit of > > > > > > > parallelizing disk access > > > > > > > 3) Re-write your jobs to first copy your data onto a local disk on the > > > > > > > batch worker (for instance, /tmp), then run on the local copy, then > > > > > > > delete the local copy. The benefit of that is that the cp command > > > > > > > will access the file in direct-access mode (with 10-20 MB/sec > > > > > > > throughput, depending on the network interface throughput). > > > > > > > 4) Make your ntuples non-split (very highly recommended). This usually > > > > > > > increases the throughput by a factor of 10-20. If your typical job > > > > > > > reads most of the branches of the tree, making tree split makes no > > > > > > > sense. Non-split trees provide direct access to disk, which is much > > > > > > > more optimal. > > > > > > > > > > > > > > Yury > > > > > > > > > > > > > > > > > > > > > On Thu, Oct 31, 2002 at 09:26:08AM -0800, Daniele del Re wrote: > > > > > > > > > > > > > > > > Hi all, > > > > > > > > > > > > > > > > in the last two days I tried to run on data and MC on the new disk AWG18. > > > > > > > > No way. I got problems in the 80% of the jobs. Someone crashed, most of > > > > > > > > them have did not read a large number of root files (actually there). > > > > > > > > > > > > > > > > This problem seems to be worse than ever. Do we have to contact > > > > > > > > computing people about this? > > > > > > > > > > > > > > > > Daniele > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > > >