On Thu, May 28, 2009 at 10:38:51AM +0900, Namshin Kim wrote: -> I changed MEM_USG_AMT down to 1kb, and it looks like the loading time is -> almost same, 1sec. On the contrary, I can open seqdb.SequenceFileDB -> instantly. -> For the cache issue, I agree with you. Usually large sequence database such -> as solexa dataset, the fastq file is more than a few gigabytes for single -> experiments and there could be tens of these dataset. We cannot load all -> sequences into memory, thus 500MB would not help us reduce query time. -> However, we need to open ScreedSequenceDB very fast to get individual record -> if we are planning to use it on the fly (web or...)
Hey Namshin, it took me a bit of a struggle to make sure that everything recompiled & reinstalled from scratch after changing that setting; I suspect the dependencies aren't entirely right in the cmake setup. Could you double check (delete the built and installed .so and any .o files, rebuild, reinstall) to make sure that the Python module "got" the .h change? Alternatively you can look at memory usage -- it should stay well under 100 mb. I saw a decline from 7s to < 1s in loading time, once I got everything working. cheers, --titus --~--~---------~--~----~------------~-------~--~----~ You received this message because you are subscribed to the Google Groups "pygr-dev" group. To post to this group, send email to pygr-dev@googlegroups.com To unsubscribe from this group, send email to pygr-dev+unsubscr...@googlegroups.com For more options, visit this group at http://groups.google.com/group/pygr-dev?hl=en -~----------~----~----~----~------~----~------~--~---
