A Friday 28 May 2010 22:54:24 Mike McGreevy escrigué:
> Hi all,
> 
> A pre-release version of HDF5 1.8.5 is available for testing, and can be
> downloaded at the following link:
> 
> ftp://ftp.hdfgroup.uiuc.edu/pub/outgoing/hdf5/hdf5-1.8.5-pre2/
> 
> If you have a few moments to test this, we would greatly appreciate it.
> While we have tested it on the platforms that we have access to, there
> are many others that we are unable to test on ourselves. Please report
> any errors to [email protected], preferably within the next week or so.
> We are planning to create and distribute the official release by
> mid-June if no critical errors are reported beforehand.

Looks good here on Linux platform (openSUSE 64-bit).  My measured performance 
in critical places seems okay too.  By the way, I like the new performance 
excerpts during test time.  In particular, I'm curious about this one:

============================
 iopipe  Test Log
============================
[clip]
I/O request size is 1000000.0MB
Before getrusage() call
fill raw        0.00user 0.00system 0.01elapsed 8509.83MB/s
fill hdf5       0.00user 0.00system 0.01elapsed 6306.27MB/s
out raw         0.00user 0.03system 0.03elapsed 1425.14MB/s
out hdf5        0.00user 0.03system 0.03elapsed 1350.30MB/s
in raw          0.00user 0.02system 0.02elapsed 1778.29MB/s
in hdf5         0.00user 0.02system 0.02elapsed 2098.98MB/s
in hdf5 partial 0.01user 0.02system 0.03elapsed 1410.44MB/s
0.05user 0.28system 0:01.71elapsed 20%CPU (0avgtext+0avgdata 0maxresident)k
184inputs+3984outputs (2major+15920minor)pagefaults 0swaps

where HDF5 seems to behave well compared with pure POSIX, except for the fill 
test where POSIX is significantly faster.  Is this expected?

-- 
Francesc Alted

_______________________________________________
Hdf-forum is for HDF software users discussion.
[email protected]
http://mail.hdfgroup.org/mailman/listinfo/hdf-forum_hdfgroup.org

Reply via email to