N1GE6 Checkpointing and Berkeley Lab Checkpoint/Restart
Peng Liang, Lip Kian Ng · 2004
APSTC-TB-2004-005 Abstract: N1GE6, formerly known as Sun Grid Engine, is widely used in HPTC environment for efficient utilization of compute resources. As applications in such environment are generally compute intensive, fault tolerance is required to minimize the impact of hardware failure. N1GE6 has several fault tolerance features and in this report, the focus will be on the checkpointing support and the integration of Berkeley Lab Checkpoint/Restart will be used as an example.