Software Tools

Software Tools

Stata and Maintaining Version Stability

Stata and Maintaining Version Stability

Every two years we update Stata in our research computing environment (both desktops and on the HBSGrid). While this is an excellent opportunity to stay abreast of the new features, it does present a problem: how does one guarantee that you will obtain the same results with the new version for a project you've been working on for a number of years?

Current Stata: Version command

Stata has the excellent command version that maintains backwards compatibility for the execution engine:

version # sets the command interpreter and other features such as random-number generators (RNGs) to version number #. version # is used to allow old programs to run correctly under more recent versions of Stata and to ensure that new programs run correctly under future versions of Stata.

This guarantees that your code will still execute correctly, but it does not necessarily guarantee that the fundamental, underlying math libraries haven't changed, which may result in calculation results which are not consistent with previous runs. Please see https://www.stata.com/manuals/pversion.pdf for more information.

Running old Stata versions

For this reason, we keep at least one to two versions of Stata installed on the HBSGrid to guarantee that changes in versions (and the underlying math and software libraries) do not cause unexpected problems.

At this time, executing Stata via drop-down Application menus through NoMachine will run the latest version of Stata that is installed. Likewise, using the command line wrappers stata-mp4-5g and their permutations ([x]stata-[se|mp4|mp8][-5g|-10g-20g]) will also run the latest Stata version.

To run older versions of Stata, one must use custom LSF commands to submit a job and point to the location of the older Stata binary. On the HBSGrid, in general, all binaries are located within respective directories at /usr/local/app. Thus:

Stata v15 SE is at /usr/local/app/stata15-se/, and

Stata v15 MP4 is at /usr/local/app/stata15-mp4/.

And so on. So to run a Stata v15 SE interactive with 5 GB RAM, one would use the interactive form of the bsub command with the path to the Stata v15 SE GUI binary. So the full command would be

bsub -q long_int -Is -n 1 -app stata-se-5g /usr/local/app/stata15-se/xstata-se

To run a Stata v15 MP4 batch file with 30 GB RAM, one would use the non-GUI binary and use the appropriate Stata command line flags to run the *.do file in batch. So the full command would be

bsub -q long -n 4 -R "rusage[mem=30000]" -M 30000 -hl /usr/local/app/stata15-mp4/stata-mp -d do mydofile.do

If you plan to use ODBC with Stata, you must also include the following command before your job submission bsub:

export LD_LIBRARY_PATH=/usr/lib64:$LD_LIBRARY_PATH

Please contact RCS if you have any questions.