Memory energy management for an enterprise decision support system

Authors:
Karthik Kumar;Kshitij Doshi;Martin Dimitrov;Yung-Hsiang Lu
Affiliations:
Purdue University, West Lafayette, IN, USA;Intel Corporation, Chandler, AZ, USA;Intel Corporation, Chandler, AZ, USA;Purdue University, West Lafayette, IN, USA
Venue:
Proceedings of the 17th IEEE/ACM international symposium on Low-power electronics and design
Year:
2011

Citing 7
Cited 1

Power aware page allocation

ASPLOS IX Proceedings of the ninth international conference on Architectural support for programming languages and operating systems
Comparing Program Phase Detection Techniques

Proceedings of the 36th annual IEEE/ACM International Symposium on Microarchitecture
Design and implementation of power-aware virtual memory

ATEC '03 Proceedings of the annual conference on USENIX Annual Technical Conference
PowerNap: eliminating server idle power

Proceedings of the 14th international conference on Architectural support for programming languages and operating systems
Memory MISER: Improving Main Memory Energy Efficiency in Servers

IEEE Transactions on Computers
Scalable high performance main memory system using phase-change memory technology

Proceedings of the 36th annual international symposium on Computer architecture
MemScale: active low-power modes for main memory

Proceedings of the sixteenth international conference on Architectural support for programming languages and operating systems

RAMZzz: rank-aware dram power management with dynamic migrations and demotions

SC '12 Proceedings of the International Conference on High Performance Computing, Networking, Storage and Analysis

Quantified Score

Hi-index	0.00

Visualization

Abstract

Energy efficiency is an important factor in designing and configuring enterprise servers. In these servers, memory may consume 40% of the total system power. Different memory configurations (sizes, numbers of ranks, speeds, etc.) can have significant impacts on the performance and energy consumption of enterprise workloads. Many of these workloads, such as decision support systems (DSS), require large amounts of memory. This paper investigates the potential to save energy by making the memory configuration adaptive to workload behavior. We present a case study on how memory configurations affect energy consumption and performance for running DSS. We measure the energy consumption and performance of a commercial enterprise server, and develop a model to describe the conditions when energy can be saved with acceptable performance degradation. Using this model, we identify opportunities to save energy in future enterprise servers.