Parallel Compression Checkpointing for Socket-Level Heterogeneous Systems

Check pointing is an effective fault tolerant technique to improve the reliability of large scale parallel computing systems. However, check pointing causes a large number of computation nodes to store a huge amount of data into file system simultaneously. It does not only require a huge storage spa...

Ausführliche Beschreibung

Bibliographische Detailangaben

Hauptverfasser:	Yongpeng Liu, Hong Zhu, Yongyan Liu, Feng Wang, Baohua Fan
Format:	Tagungsbericht
Sprache:	eng
Schlagworte:	Checkpoint and restart Checkpointing Communication networks Computer architecture Data compression GPU Graphics processing unit Pipeline Pipeline processing SIMD parallelism Socket-level heterogeneous architecture
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!