comp.lang.idl-pvwave archive
Messages from Usenet group comp.lang.idl-pvwave, compiled by Paulo Penteado

Home » Public Forums » archive » CDF and HDF5 problems
Show: Today's Messages :: Show Polls :: Message Navigator
E-mail to friend 
Switch to threaded view of this topic Create a new topic Submit Reply
CDF and HDF5 problems [message #88639] Thu, 22 May 2014 05:49 Go to next message
Matthew Argall is currently offline  Matthew Argall
Messages: 286
Registered: October 2011
Senior Member
I read both CDF and HDF5 files and am running into similar problems for each. There are two routines that slow things way down, CDF_OPEN and H5A_GET_NUM_ATTRS. My program gives

IDL> profiler, /report
Module Type Count Only(s) Avg.(s) Time(s) Avg.(s)
CDF_OPEN (S) 8 21.990163 2.748770 21.990163 2.748770
H5A_GET_NUM_ATTRS
(S) 184 16.309904 0.088641 16.309904 0.088641

However, if I try the same operations from the IDL prompt, this time is reduced to a fraction of a second. Does anyone know why this would be?
Re: CDF and HDF5 problems [message #89058 is a reply to message #88639] Sun, 20 July 2014 14:27 Go to previous message
Matthew Argall is currently offline  Matthew Argall
Messages: 286
Registered: October 2011
Senior Member
Some progress on this...

There is a new CDF_SET_VALIDATE procedure that turns file validation on and off. In IDL 8.0 file validation became the default when CDF files are opened. Turning validation off, as the documentation states, significantly decreases the amount of time needed to open a file.

As for H5A_* functions, below is a link to a program that illustrates the compounding slowness of H5A_OPEN_IDX and H5A_OPEN_NAME. I am so far unable to reproduce the same problems with H5A_GET_NUM_ATTRS and H5A_GET_NAME; however, I have encountered the same problem with those as well.

https://dl.dropboxusercontent.com/u/42944960/test_h5a_v2.pro

Here is a link to a data file

http://www.rbsp-ect.lanl.gov/data_pub/rbspa/MagEphem/def/201 2/rbspa_def_MagEphem_TS04D_20120831_v1.0.0.h5

At the bottom of the program, there is a main-level script. Simply change the location of the data file and

IDL> .run test_h5a_v2
  Switch to threaded view of this topic Create a new topic Submit Reply
Previous Topic: mpfitfun
Next Topic: Multiresolution analysis based on maximum overlap discrete wavelet transform and the Haar wavelet

-=] Back to Top [=-
[ Syndicate this forum (XML) ] [ RSS ] [ PDF ]

Current Time: Wed Oct 08 15:07:35 PDT 2025

Total time taken to generate the page: 0.00370 seconds