Abstract—Detecting the occurrence and location of performance anomalies (e.g., high jitter or loss events) is critical to ensuring the effective operation of network infrastructu...
Paul Barford, Nick G. Duffield, Amos Ron, Joel Som...
Anomaly detection is a first and important step needed to respond to unexpected problems and to assure high performance and security in IP networks. We introduce a framework and ...
Yin Zhang, Zihui Ge, Albert G. Greenberg, Matthew ...
For a grid middleware to perform resource allocation, prediction models are needed, which can determine how long an application will take for completion on a particular platform o...
— Since the advent of distributed computer systems an active field of research has been the investigation of scheduling strategies for parallel applications. The common approach...
GPU-based heterogeneous clusters continue to draw attention from vendors and HPC users due to their high energy efficiency and much improved single-node computational performance...