Title :
Fine-Grained Profiling for Data-Intensive Workflows
Author :
Dun, Nan ; Taura, Kenjiro ; Yonezawa, Akinori
Author_Institution :
Dept. of Comput. Sci., Univ. of Tokyo, Tokyo, Japan
Abstract :
Profiling is an effective dynamic analysis approach to investigate complex applications. ParaTrac is a user-level profiler using file system and process tracing techniques for data-intensive workflow applications. In two respects ParaTrac helps users refine the orchestration of workflows. First, the profiles of I/O characteristics enable users to quickly identify bottlenecks of underlying I/O subsystems. Second, ParaTrac can exploit fine-grained data-processes interactions in workflow execution to help users understand, characterize, and manage realistic data-intensive workflows. Experiments on thoroughly profiling Montage workflow demonstrate that ParaTrac is scalable to tracing events of thousands of processes and effective in guiding fine-grained workflow scheduling or workflow management systems improvements.
Keywords :
Application software; Cloud computing; Computer science; Data engineering; Databases; File systems; Grid computing; Statistical analysis; Statistical distributions; Workflow management software; profiling; workflow;
Conference_Titel :
Cluster, Cloud and Grid Computing (CCGrid), 2010 10th IEEE/ACM International Conference on
Conference_Location :
Melbourne, Australia
Print_ISBN :
978-1-4244-6987-1
DOI :
10.1109/CCGRID.2010.29