Understanding GPU Power

Bridges, Robert A.; Imam, Neena; Mintz, Tiffany M.

doi:10.1145/2962131

Cited by 80 publications

(2 citation statements)

References 64 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For instance, GPU power modeling can benefit from HighRPM, as it shares similarities with CPU power modeling. However, GPU power modeling often requires additional considerations specific to the architecture [15]. Adapting HighRPM to GPU would involve adjusting the model to utilize GPU performance counters and collecting training data on the target platform.…”

Section: Hyperparametric Analysis We Now Analyze How Sensitivementioning

confidence: 99%

Proceedings of the 52nd International Conference on Parallel Processing

2023

View full text Add to dashboard Cite

In an era where power and energy are the first-class constraints of computing systems, accurate power information is crucial for energy efficiency optimization in high-performance computing (HPC) systems. Existing power monitoring techniques rely on either software-centric power models that suffer from poor accuracy or integrated hardware measurement schemes that have a low reading update frequency and coarse granularity. These result in a low spatiotemporal resolution for power monitoring. This paper introduces HighRPM, a new method for accurately measuring power consumption on HPC systems. HighRPM combines coarse-grained power sensor readings and software power modeling techniques to improve temporal and spatial resolutions. To provide high-frequent power readings in the temporal domain, HighRPM employs statistical modeling and machine learning techniques to predict the long-term power trend and the short-term fluctuations in power consumption. To improve spatial coverage, HighRPM takes lowtime resolution node-level power consumption and uses a neural network to distribute the power readings to lower-level computing components like CPUs and memory components. We evaluate HighRPM by applying it to both ARM-based and X86-based platforms. Experimental results show that HighRPM improves time resolution by 10 times, provides accurate readings for CPUs and memory, and reduces error by 7-24% compared to other power modeling methods. CCS CONCEPTS• General and reference → Power modeling.

show abstract

Section: Hyperparametric Analysis We Now Analyze How Sensitivementioning

confidence: 99%

Proceedings of the 52nd International Conference on Parallel Processing

2023

View full text Add to dashboard Cite

show abstract

“…A useful review of the DVFS technique is provided by Mittal and Vetter [24]. The review by Bridges et al [25] looked into the modelling of the power consumption by GPUs.…”

Section: Related Workmentioning

confidence: 99%

Efficiency Near the Edge: Increasing the Energy Efficiency of FFTs on GPUs for Real-Time Edge Computing

Adámek¹,

Novotný²,

Thiyagalingam³

et al. 2021

IEEE Access

View full text Add to dashboard Cite

The Square Kilometre Array (SKA) is an international initiative for developing the world's largest radio telescope with a total collecting area of over a million square meters. The scale of the operation, combined with the remote location of the telescope, requires the use of energy-efficient computational algorithms. This, along with the extreme data rates that will be produced by the SKA and the requirement for a real-time observing capability, necessitates in-situ data processing in an edge style computing solution. More generally, energy efficiency in the modern computing landscape is becoming of paramount concern. Whether it be the power budget that can limit some of the world's largest supercomputers, or the limited power available to the smallest Internet-of-Things devices. In this paper, we study the impact of hardware frequency scaling on the energy consumption and execution time of the Fast Fourier Transform (FFT) on NVIDIA GPUs using the cuFFT library. The FFT is used in many areas of science and it is one of the key algorithms used in radio astronomy data processing pipelines. Through the use of frequency scaling, we show that we can lower the power consumption of the NVIDIA A100 GPU when computing the FFT by up to 47% compared to the boost clock frequency, with less than a 10% increase in the execution time. Furthermore, using one common core clock frequency for all tested FFT lengths, we show on average a 43% reduction in power consumption compared to the boost core clock frequency with an increase in the execution time still below 10%. We demonstrate how these results can be used to lower the power consumption of existing data processing pipelines. These savings, when considered over years of operation, can yield significant financial savings, but can also lead to a significant reduction of greenhouse gas emissions.

show abstract

Simultaneous accessing of multiple SRAM subregions forming configurable and automatically generated memory fields

Simopoulos

Alexiou

Spyridopoulos

et al. 2021

Circuit Theory & Apps

View full text Add to dashboard Cite

SummaryA novel SRAM architecture that simultaneously accesses multiple and non‐overlapped memory subregions, located even inside the same memory page, is presented in this work. This new memory architecture is called multi‐field SRAM, and the multiple selection of its subregions—referred to as the memory fields—is succeeded using special designed intra‐encoders, placed as one for every memory field. The block diagram of the typical SRAM model is extended in order to support the addressing of the multi‐field SRAM using a modified crossbar addressing approach. The support of this new memory is based on the automated generation of intra‐encoders that are adjacent to the memory fields. A test circuit, consisting of 16 memory fields, is developed in order to prove the correct operation of this new memory. Each memory storage cell uses an 8T model in order to support the crossbar addressing scheme, which is also presented. Finally, the simulation results of the test circuit are shown, verifying the successful and simultaneous multi‐field access of the memory.

show abstract

Understanding GPU Power

Cited by 80 publications

References 64 publications

Proceedings of the 52nd International Conference on Parallel Processing

Proceedings of the 52nd International Conference on Parallel Processing

Efficiency Near the Edge: Increasing the Energy Efficiency of FFTs on GPUs for Real-Time Edge Computing

Simultaneous accessing of multiple SRAM subregions forming configurable and automatically generated memory fields

Contact Info

Product

Resources

About