Titlebook: High Performance Computing - HiPC 2008; 15th International C Ponnuswamy Sadayappan,Manish Parashar,Viktor K. Pr Conference proceedings 2008

Dawdle

頌揚(yáng)國(guó)家

Yu Chen,Wenlong Li,Junmin Lin,Aamer Jaleel,Zhizhong TangThe algorithms are implemented in the open-source verification tool TAPAAL and we document the efficiency of our approach on a large benchmark of Petri net models and queries from the Model Checking Contest 2017.

閃光你我

Improving Performance of Digest Caches in Network Processorshes in network processors by exploiting the disparity between the number of flows and the number of packets transferred. Our experimental results demonstrate that SP performs better than the widely used LRU cache replacement policy in size constrained caches. Further, we characterize the misses experienced by flow identifiers in digest caches.

SLING

Optimization of BLAS on the Cell Processorn better performance. Using these techniques for double precision, we obtain up to 70–80% of peak performance for different memory bandwidth bound level 1 and 2 routines and up to 80–90% for computation bound level 3 routines.

intuition

The Design and Architecture of MAQAOAdvisor: A Live Tuning Guide. It is based on static analysis and dynamic profile of assembly code. It feeds information back to help end-user detect and understand performance problems. It proposes optimization recommendations to guide a user to perform the best transformations to get the best performance.

evasive

A Load Balancing Framework for Clustered Storage Systems server implementation. The evaluation scenario considers consolidating two real world systems, with hundreds of users each: a six-node clustered storage system supporting engineering workloads and a legacy system supporting three email severs.

切割

xanthelasma

Emg827

An Open Computing Resource Management Framework for Real-Time Computingate the suitability and versatility of the framework. We compute a metric that relates the computing resource management success to its complexity and conclude that adjusting the cost function’s parameter is more efficient than augmenting the ..-mapping’s window size.

CLOWN