Virtual Data in CMS Production
A. Arbree, P. Avery, D. Bourilkov, R. Cavanaugh, G. Graham, S. Katageri, J. Rodriguez, J. Voeckler, M. Wilde
Abstract
Initial applications of the GriPhyN Chimera Virtual Data System have been performed within the context of CMS Production of Monte Carlo Simulated Data. The GriPhyN Chimera system consists of four primary components: 1) a Virtual Data Language, which is used to describe virtual data products, 2) a Virtual Data Catalog, which is used to store virtual data entries, 3) an Abstract Planner, which resolves all dependencies of a particular virtual data product and forms a location and existence independent plan, 4) a Concrete Planner, which maps an abstract, logical plan onto concrete, physical grid resources accounting for staging in/out files and publishing results to a replica location service. A CMS Workflow Planner, MCRunJob, is used to generate virtual data products using the Virtual Data Language. Subsequently, a prototype workflow manager, known as WorkRunner, is used to schedule the instantiation of virtual data products across a grid.
Create a lesson
Related papers
Scheduling in Cloud,Fog, and Edge Computing: A Comparative Analysis and Decision Framework for Technique Selection
Rana Ghazali, Douglas G. Down
GPU-resident Execution for Timing-predictable Sensor Processing Systems
Soham Sinha, Rahul Choudhury, Shekhar Dwivedi et al.
CompileMask: Bridging the Gap Between Expressiveness and Efficiency in Attention Masking
Hongzheng Li, Jiacheng Wan, Zhaolong Xing et al.
ShoreKeeper: Accelerating Iterative Data Analytics in the Cloud via Distributed Recovery-Cost-Aware Caching
Jongseop Lee, Minsu Sun, Xinyue Ma et al.
Stacked Intelligent Metasurface-Based MultiModal Feature Fusion for Collaborative Inference in Edge Networks
Yushi Wang, Zhengxin Yu, Guhan Zheng et al.
Taking Tokenization off the Host for Agentic LLM Serving
Mohammad Siavashi, Pascal Schwinte, Ali Banaei et al.