|
Check access to TIFR (Jan 5)
|
|
|
Logistics of an Open Data analysis (Jan 10)
|
Computation for a CMS analysis is typically run in several steps: dataset skimming / flattening, observable creation, visualization, and statistical analysis.
Early steps such as dataset skimming and observable creation often particularly benefit from distributed computing.
The CMS Open Data Workshops provide tutorials for performing these analysis steps on either HTCondor or Google Cloud platforms.
|
|
HTCondor submission
|
The condor job control file can specify a docker container.
Each job’s executable file can specify code to access from Github to perform an analysis task.
References are included here to condor submission and monitoring guides.
|
|
Run your analysis
|
The hadd command allows you to merge ROOT files that have the same internal structure
Files can be extracted from TIFR to your local machine using scp
You can then analyze the POET ROOT files using other techniques from this workshop
|