Abstract
Thanks to its RDataFrame interface, ROOT now supports the execution of the same physics analysis code both on a single machine and on a cluster of distributed resources. In the latter scenario, it is common to read the input ROOT datasets over the network from remote storage systems, which often increases the time it takes for physicists to obtain their results. Storing the remote files much closer to where the computations will run can bring latency and execution time down. Such a solution can be improved further by caching only the actual portion of the dataset that will be processed on each machine in the cluster, reusing it in subsequent executions on the same input data. This paper shows the benefits of applying different means of caching input data in a distributed ROOT RDataFrame analysis. Two such mechanisms will be applied to this kind of workflow with different configurations, namely caching on the same nodes that process data or caching on a separate server.
Highlights
The high amount of data collected by the LHC experiments has made distributed computing a staple in High Energy Physics (HEP) data processing workflows for a long time, with the WLCG [1] being the prime example of efforts in that direction
Each scenario in turn presents three tests: the baseline test with caching disabled, one test with XRootD cache enabled on a server separate from the computing nodes and one test with TFilePrefetch cache enabled on the local filesystem of the computing nodes
The XRootD framework is quite well established in the community and its proxy plugin system may be used to cache remote files closer to the computing nodes
Summary
The high amount of data collected by the LHC experiments has made distributed computing a staple in High Energy Physics (HEP) data processing workflows for a long time, with the WLCG [1] being the prime example of efforts in that direction. It will be crucial to make the most out of current and future architectures In this regard, distributed computing will need to be revisited with new approaches, algorithms and frameworks. Letting the user interactively explore their dataset even as it grows larger and larger will be a requirement in many physics analysis groups Services such as SWAN [4] try to solve that need, providing a modern interactive interface for analysis through Jupyter notebooks and the possibility to run on distributed cluster resources on demand
Sign-in/Register to view highlights & summary Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
