An approach to heterogeneous process state capture/recovery to achieve minimum performance overhead during normal execution

Bungale, Prashanth P.; Sridhar, Swaroop; Krishnamurthy, V.

doi:10.1109/ipdps.2003.1213213

Cited by 7 publications

(8 citation statements)

References 9 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The second method works based on the dynamic changing code. In this method, after receiving a migration request, the execution code is changed in such a way that the migration point is available in the code [17]. A method to manage the migration process is presented in [36] that utilizes checkpoints to keep the migration timing and program consistency during the migration process.…”

Section: Task Migrationmentioning

confidence: 99%

Using task migration to improve non-contiguous processor allocation in NoC-based CMPs

Modarressi

Asadinia

Sarbazi-Azad

2013

Journal of Systems Architecture

View full text Add to dashboard Cite

Section: Task Migrationmentioning

confidence: 99%

Using task migration to improve non-contiguous processor allocation in NoC-based CMPs

Modarressi

Asadinia

Sarbazi-Azad

2013

Journal of Systems Architecture

View full text Add to dashboard Cite

“…Large number of polling points results in a large run-time cost; whereas insufficient or misplaced points would increase PM's response time. The second is code-based PM [4,19,27], which modifies the program code at run-time to insert special code for migration initiation.…”

Section: Related Workmentioning

confidence: 99%

“…The overall migration execution time cost is in the order of seconds. Similarly, Bungale et al [4] and Nollet et al, [19] introduced poll-free mechanisms to perform run-time PM. No empirical data is given for these works as yet.…”

Section: Related Workmentioning

confidence: 99%

See 1 more Smart Citation

Fine-grained hardware/software methodology for process migration in MPSoCs

Ambrose

Parameswaran

2012

Proceedings of the International Conference on Computer-Aided Design

View full text Add to dashboard Cite

Process migration (PM) is a method used in Multi-Processor System on Chips (MPSoCs) to improve reliability, reduce thermal hotspots and balance loads. However, existing PM approaches are limited by coarse granularity (i.e. can only switch at application or operating systems boundaries), and thus respond slowly. Such slow response does not allow for fine control over temperature, nor does it allow frequent migration which is necessary in certain systems. In this paper, we propose Thor, an approach which is a fine-grained reliable PM scheme, for Embedded MPSoCs, to overcome the limitations of existing PM approaches. Our approach leverages custom instructions to integrate a base processor architecture, with PM functionality. We have proposed three schemes, Thor-BM (migration at basic block boundaries), Thor-BM/CR (migration at basic block boundaries with checkpoint and recovery), and ThorIM/CR (migration at instruction level with checkpoint and recovery). Our experiments show that the execution time overhead is less than 2%, while the additional area cost and power consumption costs are approximately 50% (excluding main memories, which if taken into account would substantially decrease this overhead). The average migration time cost is 289 cycles.

show abstract

“…The Tui System [2] constructed to provide a heterogeneous migration tool that use on four common architectures in the UNIX environment. In this system, the capture and recovery of the state of a process is done in a highly machine dependent manner, that requires complete knowledge of the fixed conventions.…”

Section: Related Workmentioning

confidence: 99%