Analysis and modeling of social influence in high performance computing workloads
KAUST DepartmentComputer, Electrical and Mathematical Sciences and Engineering (CEMSE) Division
Computer Science Program
Machine Intelligence & kNowledge Engineering Lab
MetadataShow full item record
AbstractSocial influence among users (e.g., collaboration on a project) creates bursty behavior in the underlying high performance computing (HPC) workloads. Using representative HPC and cluster workload logs, this paper identifies, analyzes, and quantifies the level of social influence across HPC users. We show the existence of a social graph that is characterized by a pattern of dominant users and followers. This pattern also follows a power-law distribution, which is consistent with those observed in mainstream social networks. Given its potential impact on HPC workloads prediction and scheduling, we propose a fast-converging, computationally-efficient online learning algorithm for identifying social groups. Extensive evaluation shows that our online algorithm can (1) quickly identify the social relationships by using a small portion of incoming jobs and (2) can efficiently track group evolution over time. © 2011 Springer-Verlag.
PublisherSpringer Science + Business Media
Conference/Event name17th International Conference on Parallel Processing, Euro-Par 2011