Analysis of the Tradeoffs Between Energy and Run Time for Multilevel Checkpointing

Balaprakash, Prasanna; Bautista-Gomez, Leonardo; Bouguerra, Mohamed Slim; Wild, Stefan M.; Cappello, Franck; Hovland, Paul

doi:10.1007/978-3-319-17248-4_13

Cited by 6 publications

(15 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Hakkarinen and Chen [11] considered multi-level diskless checkpointing for tolerating simultaneous failures of multiple processors. Balaprakash et al [1] studied the trade-off between performance and energy for general multi-level checkpointing schemes.…”

Section: Related Workmentioning

confidence: 99%

“…Observation 2. There are two types of execution overheads for a pattern: (1). Error-free overhead, denoted as o ef , is the total cost of all the checkpoints placed in the pattern.…”

Section: Observationsmentioning

confidence: 99%

“…These protocols allow different levels of checkpoints to be set, each with a different checkpointing overhead and recovery ability. Typically, each level corresponds to a specific fault 1 type, and is associated to a storage device that is resilient to that type. For instance, a two-level system would deal with (i) transient memory errors (level 1) by storing key data in main memory; and (ii) node failures (level 2) by storing key data in stable storage (remote redundant disks).…”

Section: Introductionmentioning

confidence: 99%

“…The first option is to use independent checkpointing periods for each level. This option raises several difficulties, the most prominent one being overlapping 1. We use the terms fault, failure and error indifferently.…”

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Towards Optimal Multi-Level Checkpointing

Benoît

Cavelan

Fèvre

et al. 2017

IEEE Trans. Comput.

View full text Add to dashboard Cite

We provide a framework to analyze multi-level checkpointing protocols, by formally defining a k-level checkpointing pattern. We provide a first-order approximation to the optimal checkpointing period, and show that the corresponding overhead is in the order of k =1 √ 2λ C , where λ is the error rate at level , and C the checkpointing cost at level. This nicely extends the classical Young/Daly formula on single-level checkpointing. Furthermore, we are able to fully characterize the shape of the optimal pattern (number and positions of checkpoints), and we provide a dynamic programming algorithm to determine the optimal subset of levels to be used. Finally, we perform simulations to check the accuracy of the theoretical study and to confirm the optimality of the subset of levels returned by the dynamic programming algorithm. The results nicely corroborate the theoretical study, and demonstrate the usefulness of multi-level checkpointing with the optimal subset of levels.

show abstract

Section: Related Workmentioning

confidence: 99%

“…Observation 2. There are two types of execution overheads for a pattern: (1). Error-free overhead, denoted as o ef , is the total cost of all the checkpoints placed in the pattern.…”

Section: Observationsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

Towards Optimal Multi-Level Checkpointing

Benoît

Cavelan

Fèvre

et al. 2017

IEEE Trans. Comput.

View full text Add to dashboard Cite

show abstract

“…A similar technique has also been used to determine optimal checkpoint intervals for energy efficient fault tolerance [1].…”

Section: Related Workmentioning

confidence: 99%

Metrics for Energy-Aware Software Optimisation

Roberts

Wright

Fahmy

et al. 2017

Lecture Notes in Computer Science

View full text Add to dashboard Cite

A note on versions:The version presented here may differ from the published version or, version of record, if you wish to cite this item you are advised to consult the publisher's version. Please see the 'permanent WRAP url' above for details on accessing the published version and note that access may require a subscription.For more information, please contact the WRAP Team at: wrap@warwick.ac.uk Metrics for Energy-Aware Software OptimisationStephen I. Roberts, Steven A. Wright, Suhaib A. Fahmy, and Stephen A. Jarvis University of Warwick, U.K. S.I.Roberts@warwick.ac.uk Abstract. Energy consumption is rapidly becoming a limiting factor in scientific computing. As a result, hardware manufacturers increasingly prioritise energy efficiency in their processor designs. Performance engineers are also beginning to explore software optimisation and hardware/software co-design as a means to reduce energy consumption. Energy efficiency metrics developed by the hardware community are often re-purposed to guide these software optimisation efforts. In this paper we argue that established metrics, and in particular those in the Energy Delay Product (Et n ) family, are unsuitable for energy-aware software optimisation. A good metric should provide meaningful values for a single experiment, allow fair comparison between experiments, and drive optimisation in a sensible direction. We show that Et n metrics are unable to fulfil these basic requirements and present suitable alternatives for guiding energy-aware software optimisation. We finish with a practical demonstration of the utility of our proposed metrics.

show abstract

Performance and power modeling and prediction using MuMMI and 10 machine learning methods

Taylor

Lan

2022

Concurrency and Computation

View full text Add to dashboard Cite

Energy-efficient scientific applications require insight into how high performance computing system features impact the applications' power and performance. This insight can result from the development of performance and power models. In this article, we use the modeling and prediction tool MuMMI (Multiple Metrics Modeling Infrastructure) and 10 machine learning methods to model and predict performance and power consumption and compare their prediction error rates. We use an algorithm-based fault-tolerant linear algebra code and a multilevel checkpointing fault-tolerant heat distribution code to conduct our modeling and prediction study on the Cray XC40 Theta and IBM BG/Q Mira at Argonne National Laboratory and the Intel Haswell cluster Shepard at Sandia National Laboratories. Our experimental results show that the prediction error rates in performance and power using MuMMI are less than 10% for most cases. By utilizing the models for runtime, node power, CPU power, and memory power, we identify the most significant performance counters for potential application optimizations, and we predict theoretical outcomes of the optimizations. Based on two collected datasets, we analyze and compare the prediction accuracy in performance and power consumption using MuMMI and 10 machine learning methods.

show abstract

Analysis of the Tradeoffs Between Energy and Run Time for Multilevel Checkpointing

Cited by 6 publications

References 13 publications

Towards Optimal Multi-Level Checkpointing

Towards Optimal Multi-Level Checkpointing

Metrics for Energy-Aware Software Optimisation

Performance and power modeling and prediction using MuMMI and 10 machine learning methods

Contact Info

Product

Resources

About