Electrical Grid and Supercomputing Centers: An Investigative Analysis of Emerging Opportunities and Challenges

Bates, Natalie; Ghatikar, Girish; Abdulla, Ghaleb; Koenig, Gregory A.; Bhalachandra, Sridutt; Sheikhalishahi, Mehdi; Patki, Tapasya; Rountree, Barry; Poole, Stephen W

doi:10.1007/s00287-014-0850-0

Cited by 31 publications

(9 citation statements)

References 31 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently, awareness of these potentials has drawn attention to the capabilities of data centers to participate in DR programs. A survey conducted by the Lawrence Livermore National Laboratory in 2015 shows that about 50% of the participating data centers have interest in smart pricing demand side programs, such as load shedding to avoid peak demand (Bates et al, 2015). However, data centers are reluctant to participate in fast demand response programs such as providing frequency regulation (FR) in ancillary service market, for multiple reasons.…”

Section: Introductionmentioning

confidence: 99%

“…However, data centers are reluctant to participate in fast demand response programs such as providing frequency regulation (FR) in ancillary service market, for multiple reasons. One reported concern is that data centers are still learning the process of providing FR and that providing grid services on such a fast timescale can be "outside of their visibility or control" (Bates et al, 2015). This concern is well-founded considering that these programs provide novel and relatively unexplored territory from the point of view of traditional data center control and operations.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Multi-market optimization of a data center without storage systems

Fu¹,

Baker²

2020

Linköping Electronic Conference Proceedings

View full text Add to dashboard Cite

Data centers have numerous opportunities to participate in demand response programs considering their large capacities, flexible working environments and work loads, redundant design and operation, etc. Frequency regulation, as one service provided in demand response programs, can also benefits the data centers. This paper aims to develop a real-time multi-market optimization framework for a data center without storage systems to maximize their benefits from participating in both the energy market and the regulation market. Then a case study is conducted to numerically investigate the optimal bids at each hour by considering the energy cost, demand costs, and regulation revenues using a virtual data center located in PJM. Simulation results show that the proposed multi-market optimization framework can help data centers maintain minimum costs by getting maximum regulation revenues while satisfying energy and demand goals.

show abstract

Section: Introductionmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

Multi-market optimization of a data center without storage systems

Fu¹,

Baker²

2020

Linköping Electronic Conference Proceedings

View full text Add to dashboard Cite

show abstract

“…Other concerns related to these large computing facilities include, for instance, the variations in the power demand of these installations. In this regard, these peta-scale facilities present peak requirements of over 30 MW with fluctuations that can range up to a few megawatts over short periods of time [5]. One more example of the issues associated with these large facilities is that they have a limited cooling capability when trying to keep a given temperature for the server room by dissipating the heat generated by those servers [6].…”

Section: Introductionmentioning

confidence: 99%

Analyzing the performance/power tradeoff of the rCUDA middleware for future exascale systems

Reaño

Prades

Silla

2019

Journal of Parallel and Distributed Computing

View full text Add to dashboard Cite

The computing power of supercomputers and data centers has noticeably grown during the last decades at the cost of an ever increasing energy demand. The need for energy (and power) of these facilities has finally limited the evolution of high performance computing, making that many researchers are concerned not only about performance but also about energy efficiency. However, despite the many concerns about energy consumption, the search for computing power continues. In this regard, the research on exascale systems, able to deliver 10 18 floating point operations per second, has reached a widely consensus that these systems should operate within a maximum power budget of 20 megawatts. Many efficiency improvements are necessary for achieving this goal. One of these improvements is the usage of ARM low-power processors, as the Mont-Blanc project proposes. In this paper we analyze the combined use of ARM processors with the rCUDA remote GPU virtualization middleware as a way to improve efficiency even more. Results show that it is possible to speed up applications by almost 8x while reducing energy consumption up to 35% when rCUDA is used to access high-end GPUs. These improvements are achieved while maintaining a feasible average power consumption level for future exascale systems.

show abstract

“…Limiting this during a time period leads to controlling the energy consumption (as power.dt = energy). The survey provided in [2] gives a thorough analysis of related work on power management strategies along with details upon the relationship between supercomputing centers and electricity service providers in the US. Among the different studied techniques, powercapping in conjuction with job scheduling and shutdown mechanisms appeared as the most promising.…”

Section: Introductionmentioning

confidence: 99%

Towards Energy Budget Control in HPC

Dutot

Georgiou²,

Glesser

et al. 2017

2017 17th IEEE/ACM International Symposium on Cluster, Cloud and Grid Computing (CCGRID)

View full text Add to dashboard Cite

Energy consumption has become one of the most critical issues in the evolution of High Performance Computing systems (HPC). Controlling the energy consumption of HPC platforms is not only a way to control the cost but also a step forward on the road towards exaflops. Powercapping is a widely studied technique that guarantees that the platform will not exceed a certain power threshold instantaneously but it gives no flexibility to adapt job scheduling to a longer term energy budget control. We propose a job scheduling mechanism that extends the backfilling algorithm to become energy-aware. Simultaneously, we adapt resource management with a node shutdown technique to minimize energy consumption whenever needed. This combination enables an efficient energy consumption budget control on a cluster during a period of time. The technique is experimented, validated and compared with various alternatives through extensive simulations. Experimentation results show high system utilization and limited bounded slowdown along with interesting outcomes in energy efficiency while respecting an energy budget during a particular time period.

show abstract

Electrical Grid and Supercomputing Centers: An Investigative Analysis of Emerging Opportunities and Challenges

Cited by 31 publications

References 31 publications

Multi-market optimization of a data center without storage systems

Multi-market optimization of a data center without storage systems

Analyzing the performance/power tradeoff of the rCUDA middleware for future exascale systems

Towards Energy Budget Control in HPC

Contact Info

Product

Resources

About