Global cost diversity aware dispatch algorithm for heterogeneous data centers

Sankaranarayanan, Ananth Narayan; Sharangi, Somsubhra; Fedorova, Alexandra

doi:10.1145/1958746.1958787

Cited by 11 publications

(7 citation statements)

References 10 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…A two-level technique to reduce the electricity cost in multi location DC has been proposed [8]. The spatial variations in electricity prices were exploited at the first level and scheduling strategies were proposed at the second level to distribute the requests among heterogeneous servers available at the selected DC.…”

Section: Load Balancing By Spatial Pricingmentioning

confidence: 99%

Dynamic Cost-Aware Routing of Web Requests

Velusamy

Lent

2018

Future Internet

View full text Add to dashboard Cite

Work within next generation networks considers additional network convergence possibilities and the integration of new services to the web. This trend responds to the ongoing growth of end-user demand for services that can be delivered anytime, anywhere, on any web-capable device, and of traffic generated by new applications, e.g., the Internet of Things. To support the massive traffic generated by the enormous user base and number of devices with reliability and high quality, web services run from redundant servers. As new servers need to be regularly deployed at different geographical locations, energy costs have become a source of major concern for operators. We propose a cost aware method for routing web requests across replicated and distributed servers that can exploit the spatial and temporal variations of both electricity prices and the server network. The method relies on a learning automaton that makes per-request decisions, which can be computed much faster than regular global optimization methods. Using simulation and testbed measurements, we show the cost reductions that are achievable with minimal impact on performance compared to standard web routing algorithms.

show abstract

Section: Load Balancing By Spatial Pricingmentioning

confidence: 99%

Dynamic Cost-Aware Routing of Web Requests

Velusamy

Lent

2018

Future Internet

View full text Add to dashboard Cite

show abstract

“…Virtualization and server consolidation, e.g., in [24], [25], [26], [27], [28], can reduce the traffic dynamics by consolidating applications, by which the number of active servers can be reduced. Schemes in these studies either only work for the large-time-scale traffic variation, or can only be applied to a single data center operation Most recently, cross-IDC power management has received significant attention, e.g., in [29], [30], [31], [32], [33], [34], [35], [36], [37], [38], [39], [40]. One central idea for cross-IDC power management is to leverage the diversity of geolocation and the corresponding difference in electricity price.…”

Section: Related Workmentioning

confidence: 99%

Efficient Server Provisioning and Offloading Policies for Internet Data Centers with Dynamic Load-Demand

Liu

Fan

2015

IEEE Trans. Comput.

View full text Add to dashboard Cite

In data centers, traffic demand varies in both large and small time scales. A data center with dynamic traffic often needs to over-provision active servers to meet the peak demand, which incurs significant energy cost. In this paper, our goal is to reduce energy cost of a set of distributed Internet data centers (IDCs) while maintaining the quality of service of the dynamic traffic. In particular, we consider the outage probability as the QoS metric, where outage is defined as service demand exceeding the capacity. We require the outage probability at each IDC to be smaller than a predefined threshold. Our goal is thus to minimize total energy cost over all IDCs, subject to the outage probability constraint. We achieve the goal by dynamically adjusting server capacity and performing load shifting in different time scales. We propose three different load-shifting and joint capacity allocation schemes with different complexity and performance. Our schemes leverage both stochastic multiplexing gain and electricity-price diversity. Thus, improving over prior work, our schemes reduce energy consumption/cost even when all IDCs have the same electricity price. We use both simulated load traces and real traffic traces to evaluate the performance of the proposed schemes. Results show that our proposed schemes are efficient in reducing energy cost, and robust in QoS provisioning.

show abstract

“…The job migration requires bandwidth intensive migration of the application's state and data, consequently, they incorporate bandwidth cost in the electricity cost optimization problem. In [8] Sankaranarayanan et al exploit the heterogeneity of data centers for achieving additional energy efficiency and power savings via intelligent scheduling of requests to heterogeneous servers at each data center. In [7] the authors propose an optimization framework for throughput-intensive applications like web search engine.…”

Section: Related Workmentioning

confidence: 99%

Dynamic Control of Electricity Cost with Power Demand Smoothing and Peak Shaving for Distributed Internet Data Centers

Yao

et al. 2012

2012 IEEE 32nd International Conference on Distributed Computing Systems

View full text Add to dashboard Cite

Abstract-Internet based service providers, such as Amazon, Google, Yahoo etc, build their data centers (IDC) across multiple regions to provide reliable and low latency of services to clients. Ever-increasing service demand, complexity of services and growing client population cause enormous power consumptions by these IDCs incuring a major part of their running costs. Modern electric power grid provides a feasible way to dynamically and efficiently manage the electricity cost of distributed IDCs based on the Locational Marginal Pricing (LMP) policy. While recent works exploit LMP by electricity-price based geographic load distribution, the dynamic workload and high volatility of electricity prices induce highly volatile power demand and critical power peak problem. The benefit of cost minimization via geographic load distribution is counterbalanced with the high cost incurred by violating the peak power. In this paper, we study the dynamic control of electricity cost to provide low volatility in power demand and shaving of power peaks. To this end, a Model Predictive Control (MPC) electricity cost minimization problem is formulated based on a time-continuous differential model. The proposed solution minimizes electricity costs, provides low variation in power demand by penalizing the change in workload and alleviates the power peaks by tracking the available power budget. By providing extensive simulation results based on real-life electricity price traces we show the effectiveness of our approach.

show abstract

Global cost diversity aware dispatch algorithm for heterogeneous data centers

Cited by 11 publications

References 10 publications

Dynamic Cost-Aware Routing of Web Requests

Dynamic Cost-Aware Routing of Web Requests

Efficient Server Provisioning and Offloading Policies for Internet Data Centers with Dynamic Load-Demand

Dynamic Control of Electricity Cost with Power Demand Smoothing and Peak Shaving for Distributed Internet Data Centers

Contact Info

Product

Resources

About