Energy efficient network data transport through adaptive compression using the DEEP platforms

Singh, Digvijay; Kaiser, William J.

doi:10.1109/wimob.2012.6379084

Cited by 2 publications

(7 citation statements)

References 14 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…However, none of them have tried to attribute the measured energy consumption of the disk drive to an application's individual processes. The DEEP platform is a direct energy measurement platform [2,10], which is the foundation for all the power measurements performed in this manuscript. Time-stamped direct energy measurements provided by the platform present important advantages, such as superior run-time accuracy, compared to model-based methods [15].…”

Section: Related Workmentioning

confidence: 99%

“…2, is based on a computing system with an N330 Intel Atom CPU, SDRAM memory, secondary storage, and network interfaces, which are monitored for power consumption by a high speed data sampling system [2,10]. By employing various event synchronization methods, the hardware power dissipation data received from the sampling system is synchronized with the operating system clock and kernel software events.…”

Section: A Deep Platformmentioning

confidence: 99%

“…Usage of kprobes, which enable user-defined runtime handler execution at break-points in software code, enable fast instrumentation of the software systems. Events from multiple CPU cores can also be synchronized using the system [2]. DEEP provides a power log at a data sampling rate of 10 kHz.…”

Section: A Deep Platformmentioning

confidence: 99%

“…In this case, the write operations to the disk didn't start until the program's execution was close to completion and continued for a significant time interval after the causative program had finished issuing I/O requests. Processing) platform [2]. Being an easy-to-use and nonintrusive direct energy measurement platform, DEEP avoids the limitations suffered by simplistic power meters, indirect energy models and simulation tools, such as inaccuracy and inability to capture the runtime dynamism in the power consumption of modern hardware and software systems [15].…”

Section: Introductionmentioning

confidence: 99%

“…Being an easy-to-use and nonintrusive direct energy measurement platform, DEEP avoids the limitations suffered by simplistic power meters, indirect energy models and simulation tools, such as inaccuracy and inability to capture the runtime dynamism in the power consumption of modern hardware and software systems [15]. DEEP also enables the synchronization of instantaneous power measurements with the operating system clock [2]. This paper presents a system to estimate the hard disk drive energy consumption for I/O write operations and attribute it to the causative processes.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Accurate and Low-Overhead Process-Level Energy Estimation for Modern Hard Disk Drives

Yan

Lonappan

Vajid

et al. 2013

2013 IEEE International Conference on Green Computing and Communications and IEEE Internet of Things and IEEE Cyber, Physical A

Self Cite

View full text Add to dashboard Cite

Measuring the energy consumed by modern secondary storage devices and attributing it to the causative application's processes can guide optimization of application energy efficiency. Prior work has focused on hard disk drive or file system energy management using custom-built measurement systems and models, but suffers from limitations that prohibit the attribution of the hardware energy consumption to the causative processes. This paper presents a novel process-level disk drive energy estimation system that monitors the operating system kernel to predict the storage power consumption created by individual processes of a software application. Evaluation reveals that the system is accurate with prediction errors around 7% even for workloads involving simultaneous file system operations from multiple processes. The system creates virtually no power consumption overhead in the CPU and about 4% overhead in the RAM module. Results using I/O benchmarks demonstrate that a large portion of the energy usage (up to 100% in some cases) in modern disk drives can occur after the causative processes have already completed execution. Hence, results obtained using power meters or direct energy measurement systems based solely on a benchmark's execution lifetime can drastically underestimate the benchmark application's storage energy demands.

show abstract

Section: Related Workmentioning

confidence: 99%

Section: A Deep Platformmentioning

confidence: 99%

Section: A Deep Platformmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

Accurate and Low-Overhead Process-Level Energy Estimation for Modern Hard Disk Drives

Yan

Lonappan

Vajid

et al. 2013

2013 IEEE International Conference on Green Computing and Communications and IEEE Internet of Things and IEEE Cyber, Physical A

Self Cite

View full text Add to dashboard Cite

show abstract

Energy efficient task scheduling on a multi-core platform using real-time energy measurements

Singh

Kaiser

2014

Proceedings of the 2014 International Symposium on Low Power Electronics and Design

Self Cite

View full text Add to dashboard Cite

This paper presents a large advance in energy-efficient operating system multiprocessor task scheduling with experimentally proven benefits for standard Linux multi-core computing platforms. This Energy Aware Scheduler (EAS) introduces micro-Operations executed Per Joule (OPJ) as a metric representing run-time task energy efficiency. A novel platform architecture permits eventresolved real-time energy measurements. EAS uses OPJ values for scheduling tasks to reduce resource contention. Compared to the Linux task scheduler (Completely Fair Scheduler), EAS improves energy efficiency by over 30% and execution time by over 24%.

show abstract

Energy efficient network data transport through adaptive compression using the DEEP platforms

Cited by 2 publications

References 14 publications

Accurate and Low-Overhead Process-Level Energy Estimation for Modern Hard Disk Drives

Accurate and Low-Overhead Process-Level Energy Estimation for Modern Hard Disk Drives

Energy efficient task scheduling on a multi-core platform using real-time energy measurements

Contact Info

Product

Resources

About