Energy Management for Commercial Servers

By Charles R. Lefurgy, K. Rajamani, F. Rawson, Wes Felter, M. Kistler, T. Keller. Computer, 2003

495 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Hardware and Architecture, Information Systems

View Paper: https://www.semanticscholar.org/paper/539b79c98106ece36a646e3fa50b876f396afe62

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Hardware and Architecture

In Information Systems

Citations in the corpus

Data Center Energy Consumption Modeling: A Survey (2016, background): “it was observed that fan power dominated the small configuration of IBM p670 server power envelope (51%) while in large configuration it represented a considerable portion (28%) of the server power envelope [270].”
Memory power management via dynamic voltage/frequency scaling (2011, unknown): “[2] C.”
Scalable high performance main memory system using phase-change memory technology (2009, unknown): “Furthermore, main memory consisting entirely of DRAM is already hitting the power and cost limits [15].”
Enhancing lifetime and security of PCM-based Main Memory with Start-Gap Wear Leveling (2009, background): “Unfortunately, main memory consisting entirely of DRAM is already hitting power and cost limits [9].”
Multi-Objective Virtual Machine Placement in Virtualized Data Center Environments (2010, background): “To reduce these costs, the use of virtualization to consolidate workloads and tune off unloaded servers has been proposed to achieve greater energy savings [3][7][18][21][24].”
RAPL: Memory power estimation and capping (2010, background): “Several studies have shown the importance of limiting memory power in enterprise server systems [15,16].”
Deep Supervised Learning Using Local Errors (2017, background): “Moreover, energy needed to drive oﬀ-chip traﬃc from/to external memory as well as memory read/write energy often contribute signiﬁcantly to the overall energy consumption [51, 52].”
TOP-PIM: throughput-oriented programmable processing in memory (2014, background): “As a result, the memory system is often a performance bottleneck and accounts for a significant, and increasing, fraction of system level energy consumption [30, 43].”
Power and energy management for server systems (2004, background): “Researchers have suggested a few other potential avenues in [15].”
Phase-Change Memory Optimization for Green Cloud with Genetic Algorithm (2015, background): “However, addressing the speed issue, dynamic random access memory (DRAM) has considered the crucial technique for main memory that has almost reached its scalability upper limit [1].”
PowerNap: eliminating server idle power (2009, methodology): “Figure 2 reproduces typical server power breakdowns for the IBM p670 [14], Sun UltraSparc T2000 [12], and a generic server specified by Google [6], respectively.”
No "power" struggles: coordinated multi-level power management for the data center (2008, unknown): “Categories and Subject Descriptors C.0 [Computer Systems Organization], D.4 [Operating Systems] General Terms Algorithms, Design, Management, Measurement, Performance.”
Memory scaling: A systems architecture perspective (2013, background): “In particular, on the systems/architecture front, energy and power consumption have become key design limiters as the memory system continues to be responsible for a significant fraction of overall system energy/power [43].”
Server-Level Power Control (2007, background): “As modern enterprise data centers continue to increase computing capabilities to meet their growing business requirements, high-density servers become more and more desirable due to space considerations and better system management features.”
Just In Time Dynamic Voltage Scaling: Exploiting Inter-Node Slack to Save Energy in MPI Programs (2005, background): “Such work shows that power and energy management are critical for commercial workloads, especially web servers [6, 29 ].”
Processing Data Where It Makes Sense: Enabling In-Memory Computation (2019, background): “[133] showed, in 2003 that, in large commercial servers designed by IBM, the off-chip memory hierarchy (including, at that time, DRAM, interconnects, memory controller, and off-chip caches) consumed between 40% and 50% of the total system energy.”
pMapper: Power and Migration Cost Aware Application Placement in Virtualized Systems (2008, background): “Energy management in server clusters has been a popular area of research since the beginning of this decade [12, 1].”