Short-term Load Prediction and Energy-Aware Load Balancing for Data Centers Serving Online
@inproceedings{Stachecki2015ShorttermLP,
title={Short-term Load Prediction and Energy-Aware Load Balancing for Data Centers Serving Online},
author={Tyler Stachecki and Kanad Ghose},
year={2015}
}We introduce and evaluate an automated technique for dynamically provisioning server capacity in a data center that caters to on-line services. Shortterm load prediction is used to realize an effective energy-aware data center load balancing technique that achieves significant energy savings without compromising the delivered performance. In our heterogeneous datacenter consisting of 150 Linux servers, the load balancer is able to achieve a 30.8% reduction in overall server energy consumption…
2 Citations
Virtual Melting Temperature: Managing Server Load to Minimize Cooling Overhead with Phase Change Materials
- Computer Science2018 ACM/IEEE 45th Annual International Symposium on Computer Architecture (ISCA)
- 2018
VMT is proposed, a thermal aware job placement technique that adds an active, tunable component to enable greater control over datacenter thermal output and reduces peak cooling load by up to 12.8% to provide over two million dollars in cost savings when a smaller cooling system is installed.
An Experimental Analysis of Hot Aisle Containment Systems
- Physics2018 17th IEEE Intersociety Conference on Thermal and Thermomechanical Phenomena in Electronic Systems (ITherm)
- 2018
In recent years, various airflow containment systems have been deployed in data centers to improve the cooling efficiency by minimizing the mixing of hot and cold air streams. The goal of this study…
References
SHOWING 1-10 OF 17 REFERENCES
Energy-Aware Load Direction for Servers: A Feasibility Study
- Computer Science2012 20th Euromicro International Conference on Parallel, Distributed and Network-based Processing
- 2012
This work proposes an automated technique for allocating workload to servers to operate the fewest number of servers that are needed to cope with the instantaneous workload, leaving some headroom for workload surges.
AutoScale: Dynamic, Robust Capacity Management for Multi-Tier Data Centers
- Computer ScienceTOCS
- 2012
A dynamic capacity management policy, AutoScale, is introduced that greatly reduces the number of servers needed in data centers driven by unpredictable, time-varying load, while meeting response time SLAs and robustness.
Multi-mode energy management for multi-tier server clusters
- Computer Science2008 International Conference on Parallel Architectures and Compilation Techniques (PACT)
- 2008
Experimental results using realistic dynamic workloads based on the TPC-W benchmark show that exploiting multiple sleep states results in significant additional cluster-wide energy savings up to 23% with little or no performance degradation.
Energy-Efficient Real-Time Heterogeneous Server Clusters
- Computer Science12th IEEE Real-Time and Embedded Technology and Applications Symposium (RTAS'06)
- 2006
A cluster-wide QoS-aware technique that dynamically reconfigures the cluster to reduce energy consumption during periods of reduced load is presented and results show it is possible to save up to 45% of the total energy consumed by the servers.
Towards energy proportionality for large-scale latency-critical workloads
- Computer Science2014 ACM/IEEE 41st International Symposium on Computer Architecture (ISCA)
- 2014
PEGASUS is presented, a feedback-based controller that significantly improves the energy proportionality of WSC systems, as demonstrated by a real implementation in a Google search cluster.
SleepScale: Runtime joint speed scaling and sleep states management for power efficient data centers
- Computer Science2014 ACM/IEEE 41st International Symposium on Computer Architecture (ISCA)
- 2014
This paper presents SleepScale, a runtime power management tool designed to efficiently exploit existing power control mechanisms, and evaluates it using workload traces from data centers to achieve significant power savings relative to conventional power management strategies.
Power, pollution and the internet
- Power, pollution and the internet
- 2012
Web pages of Low Power Computing's server line at: http://lopoco
- Web pages of Low Power Computing's server line at: http://lopoco
- 2015
Estimating total power consumption by servers in the US and the world
- Estimating total power consumption by servers in the US and the world
- 2007
The sorry state of server utilization and the impending post-hypervisor era
- The sorry state of server utilization and the impending post-hypervisor era
- 2013