Hi Rob, So probably HDF5 library uses path resolution whenever dataset read process executed? But in my code, for no hierarchy there is no need to do that process by the library since the dataset directly can be accessed from "root" group, which means that the process must be shorter to open and read dataset. What do you think?
Thanks for replying. Regards, Elisa --- On Wed, 9/15/10, Rob Latham <[email protected]> wrote: From: Rob Latham <[email protected]> Subject: Re: [Hdf-forum] Store data in hierarchy, does it impact the performance? To: "HDF Users Discussion List" <[email protected]> Date: Wednesday, September 15, 2010, 10:20 AM On Wed, Sep 15, 2010 at 12:23:47AM +0200, Quincey Koziol wrote: > Hi Elisa, > > On Sep 15, 2010, at 12:09 AM, elisa sibarani wrote: > > > Hi Quincey, > > > > Here I attached the example of HDF5 file with hierarchy, file with no > > hierarchy inside, and the benchmark result. > > Actually, I meant source code for the benchmark. Can you send that? the graph is interesting, though: writing either way is essentially the same. it's the read case that is taking significantly longer. could there be anything in the way HDF5 resolves groups? In UNIX land, path resolution happens directory by directory. -- Rob Latham Mathematics and Computer Science Division Argonne National Lab, IL USA _______________________________________________ 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
