Santosh, Similar performance issue was reported on this list in the past (see http://hdf-forum.184993.n3.nabble.com/Variable-length-types-slowness-td195101.html#a195101)
The fix is scheduled for 1.8.7 (May 2011), but we will let everyone on this list know when it becomes available in a snapshot (hopefully before Thanksgiving if things go according to the plan ;-) Thank you! Elena ~~~~~~~~~~~~~~~~~~~~~ Elena Pourmal The HDF Group [email protected] ~~~~~~~~~~~~~~~~~~~~~ On Oct 11, 2010, at 12:04 AM, santoshdarekar wrote: > > Hi, > > I have fucntion in my code which writes variable length data to HDF5 file. > > I have a dataset created as, > > fileType = H5Tvlen_create(H5T_NATIVE_UCHAR) > > 'DataSpace' with Rank: 1 and Dims: 1; > > And chunking enabled for the same. > > A function writing a data to this dataset, is being called in loop. > > Before writing to a dataset I extend the dimensions of the dataset as well. > > Till here, everything goes correct. i.e. I'm able to extend dataset with > this data type successfully. > > But, when I call, > > H5Dwrite (dset, memtype, memspace, filespace, H5P_DEFAULT,wdata); > > CPU usage goes 100%. > > Is it a normal behaviour that the H5DWrite API take such a huge part of CPU > to write variable length of data? > > > > > > ----- > Best Regards, > Santosh > -- > View this message in context: > http://hdf-forum.184993.n3.nabble.com/Write-Variable-length-record-to-HDF5-file-tp1678291p1678291.html > Sent from the hdf-forum mailing list archive at Nabble.com. > > _______________________________________________ > Hdf-forum is for HDF software users discussion. > [email protected] > http://mail.hdfgroup.org/mailman/listinfo/hdf-forum_hdfgroup.org
_______________________________________________ Hdf-forum is for HDF software users discussion. [email protected] http://mail.hdfgroup.org/mailman/listinfo/hdf-forum_hdfgroup.org
