Accurate Fine-Grained Processor Power Proxies

Huang, Wei; Lefurgy, Charles; Kuk, William; Buyuktosunoglu, Alper; Floyd, Michael S.; Rajamani, Karthick; Allen-Ware, Malcolm; Brock, Bishop

doi:10.1109/micro.2012.29

Cited by 44 publications

(35 citation statements)

References 13 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Ordinarily, the processor temperature would need to be included in the input parameters of a power or energy model, since leakage power increases with the temperature [9]. This would pose a problem, since we found no well-documented inband temperature sensor that is accessible to us.…”

Section: Methodsmentioning

confidence: 99%

See 1 more Smart Citation

Modeling CPU Energy Consumption of HPC Applications on the IBM POWER7

Gschwandtner

Knobloch

Mohr

et al. 2014

2014 22nd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing

View full text Add to dashboard Cite

Energy consumption optimization of HPC applications inherently requires measurements for reference and comparison. However, most of today's systems lack the necessary hardware support for power or energy measurements. Furthermore, in-band data availability is preferred for specific optimization techniques such as auto-tuning. For this reason, we present in-band energy consumption models for the IBM POWER7 processor based on hardware counters. We demonstrate that linear regression is a suitable means for modeling energy consumption, and we rely on already available, highlevel benchmarks for training instead of self-written or handtuned micro-kernels. We compare modeling efforts for different instruction mixes caused by two compilers (GCC and IBM XL) as well as various multi-threading usage scenarios, and validate across our training benchmarks and two real-world applications. Results show mean errors of approximately 1% and overall max errors of 5.3% for GCC.

show abstract

Section: Methodsmentioning

confidence: 99%

“…In [9] Huang et al present a per-core power proxy for the POWER7. They achieve a comparable mean error of 1.8% for their workloads, however their model relies on activity sensors that are not available in production (neither in-nor out-of-band) and they train with over 700 micro-benchmarks.…”

Section: Related Workmentioning

confidence: 99%

Modeling CPU Energy Consumption of HPC Applications on the IBM POWER7

Gschwandtner

Knobloch

Mohr

et al. 2014

2014 22nd Euromicro International Conference on Parallel, Distributed, and Network-Based Processing

View full text Add to dashboard Cite

show abstract

“…We conservatively model voltage/frequency transition latencies of 4 µs, longer than the 0.5 µs achievable with FIVR [5]. Power model: We evaluate Rubik's power savings using an accurate power model that we construct with a methodology similar to prior work [9,18,30,54]. We fit the model to a Supermicro 5018D-MTF server with a 4-core Haswell (Xeon E3-1240v3 [20]) and 32GB of unbuffered DDR3-1600 ECC DRAM.…”

Section: Experimental Methodologymentioning

confidence: 99%

Rubik

Kasture

Bartolini

Beckmann

et al. 2015

Proceedings of the 48th International Symposium on Microarchitecture

109

View full text Add to dashboard Cite

Latency-critical workloads (e.g., web search), common in datacenters, require stable tail (e.g., 95th percentile) latencies of a few milliseconds. Servers running these workloads are kept lightly loaded to meet these stringent latency targets. This low utilization wastes billions of dollars in energy and equipment annually.Applying dynamic power management to latency-critical workloads is challenging. The fundamental issue is coping with their inherent short-term variability: requests arrive at unpredictable times and have variable lengths. Without knowledge of the future, prior techniques either adapt slowly and conservatively or rely on application-specific heuristics to maintain tail latency.We propose Rubik, a fine-grain DVFS scheme for latency-critical workloads. Rubik copes with variability through a novel, general, and efficient statistical performance model. This model allows Rubik to adjust frequencies at sub-millisecond granularity to save power while meeting the target tail latency. Rubik saves up to 66% of core power, widely outperforms prior techniques, and requires no application-specific tuning.Beyond saving core power, Rubik robustly adapts to sudden changes in load and system performance. We use this capability to design RubikColoc, a colocation scheme that uses Rubik to allow batch and latencycritical work to share hardware resources more aggressively than prior techniques. RubikColoc reduces datacenter power by up to 31% while using 41% fewer servers than a datacenter that segregates latency-critical and batch work, and achieves 100% core utilization.

show abstract

“…Activity proxies constitute a well-known technique to estimate or track the power consumption profile of processor chips [12][13][14][15][16]. In the z13 design, they consist of a set of 16 activity counters.…”

Section: Activity Proxiesmentioning

confidence: 99%

Robust power management in the IBM z13

et al. 2015

View full text Add to dashboard Cite

The power management strategy adopted for the IBM z13i processor chip (referred to as the CP or Central Processor chip) is guided by three basic principles: (a) controlling the peak power consumption by setting a realistic limit on the so-called thermal design power or thermal design point (TDP) driven by customer workloads and maximum-power stress microbenchmarks; (b) reduction of the voltage margin by using a novel dynamic guard-banding technique; and (c) the creation of a rich new set of fine-grained, time-synchronized sensors that track performance, power, temperature, and power management behavior for a running machine. A prime requirement of the power management architecture is that the efficient control mechanisms be designed in such a manner that the high standards of IBM z Systemsi application performance and reliability be maintained without any compromise. In this paper, we describe the key features constituting the z13 CP robust power management architecture and design that meet the stipulated objectives.

show abstract

Accurate Fine-Grained Processor Power Proxies

Abstract: Abstract

Cited by 44 publications

References 13 publications

Modeling CPU Energy Consumption of HPC Applications on the IBM POWER7

Modeling CPU Energy Consumption of HPC Applications on the IBM POWER7

Rubik

Robust power management in the IBM z13

Contact Info

Product

Resources

About