Massively parallel applications often require periodic data checkpointing for program restart
and post-run data analysis. Although high performance computing systems provide massive …

This work is inspired by the observation that file systems are unable to cater for a group of
related clients accessing shared files. This hampers their ability to fine-tune the consistency …

Massively parallel applications often require periodic data checkpointing for program restart
and post-run data analysis. For many high performance applications uncoordinated I/O is …