Cargando…
Parallelizing TTree::Draw functionality with PROOF
In ROOT, the software context of this project, multi-threading is not currently an easy option, because ROOT is not by construction thread-aware and thread-safeness can only be achieved with heavy locking. Therefore, for a ROOT task, multi-processing is currently the most eective way to achieve cunc...
Autor principal: | |
---|---|
Lenguaje: | eng |
Publicado: |
2014
|
Materias: | |
Acceso en línea: | http://cds.cern.ch/record/1756818 |
Sumario: | In ROOT, the software context of this project, multi-threading is not currently an easy option, because ROOT is not by construction thread-aware and thread-safeness can only be achieved with heavy locking. Therefore, for a ROOT task, multi-processing is currently the most eective way to achieve cuncurrency. Multi-processing in ROOT is done via PROOF. PROOF is used to enable interactive analysis of large sets of ROOT les in parallel on clusters of computers or many-core machines. More generally PROOF can parallelize tasks that can be formulated as a set of independent sub-tasks. The PROOF technology is rather ecient to exploit all the CPU's provided by many-core processors. A dedicated version of PROOF, PROOF-Lite, provides an out-of-the-box solution to take full advantage of the additional cores available in today desktops or laptops. One of the items on the PROOF plan of work is to improve the inte- gration of PROOF-Lite for local processing of ROOT trees. In this project we investigate the case of the Draw functionality and its implementation via PROOF. |
---|