Energy expenses are becoming an increasingly important fraction of data center operating costs. At the same time, the energy expense per unit of computation can vary significantly between two different locations. In this paper, we characterize the variation due to fluctuating electricity prices and argue that existing distributed systems should be able to exploit this variation for significant economic gains. Electricity prices exhibit both temporal and geographic variation, due to regional demand differences, transmission inefficiencies, and generation diversity. Starting with historical electricity prices, for twenty nine locations in the US, and network traffic data collected on Akamai’s CDN, we use simulation to quantify the possible economic gains for a realistic workload. Our results imply that existing systems may be able to save millions of dollars a year in electricity costs, by being cognizant of locational computation cost differences. Categories andSubject Descriptors

The garage innovator creates new web applications which may rocket to popular success – or sink when the flash crowd that arrives melts the web server. In the web context, utility computing provides a path by which the innovator can, with minimal capital, prepare for overwhelming popularity. Many components required for web computing have recently become available as utilities. We analyze the design space of building a loadbalanced system in the context of garage innovation. We present six experiments that inform this analysis by highlighting limitations of each approach. We report our experience with three services we deployed in “garage” style, and with the flash crowds that each drew. 1

Energy consumption is a major and costly problem in data centers. A large fraction of this energy goes to powering idle machines that are not doing any useful work. We identify two causes of this inefficiency: low server utilization and a lack of power-proportionality. To address this problem we present a design for an power-proportional cluster consisting of a power-aware cluster manager and a set of heterogeneous machines. Our design makes use of currently available energy-efficient hardware, mechanisms for transitioning in and out of low-power sleep states, and dynamic provisioning and scheduling to continually adjust to workload and minimize power consumption. With our design we are able to reduce energy consumption while maintaining acceptable response times for a web service workload based on Wikipedia. With our dynamic provisioning algorithms we demonstrate via simulation a 63 % savings in power usage in a typically provisioned datacenter where all machines are left on and awake at all times. Our results show that we are able to achieve close to 90 % of the savings a theoretically optimal provisioning scheme would achieve. We have also built a prototype cluster which runs Wikipedia to demonstrate the use of our design in a real environment.

Abstract—Power consumption imposes a significant cost for data centers implementing cloud services, yet much of that power is used to maintain excess service capacity during periods of predictably low load. This paper investigates how much can be saved by dynamically ‘right-sizing ’ the data center by turning off servers during such periods, and how to achieve that saving via an online algorithm. We prove that the optimal offline algorithm for dynamic right-sizing has a simple structure when viewed in reverse time, and this structure is exploited to develop a new ‘lazy ’ online algorithm, which is proven to be 3-competitive. We validate the algorithm using traces from two real data center workloads and show that significant cost-savings are possible. I.

With the increasing popularity of Internet-based information retrieval and cloud computing, saving energy in Internet data centers (a.k.a. hosting centers, server farms) is of increasing importance. Current research approaches are based on dynamically adjusting the active server set in order to turn off a portion of the servers and save energy without compromising the quality of service; the workload is then distributed, conventionally equally (i.e. balanced), across the active servers. Although there is ample work that demonstrates energy savings through dynamic server provisioning, there is little work on thermal-aware server provisioning. This paper provides a formulation of the thermal aware active server set provisioning (TASP), in a nonlinear minimax binary integer programming form, and a series of heuristic approaches to solving them, namely MiniMax, bb-sLRH, CP-sLRH and sLRH. Furthermore, it introduces thermal-aware workload distribution (TAWD) among the active servers. The proposed heuristics are evaluated using a thermal model of the ASU HPCI data center, while the request traffic is based on real web traces of the 1998 FIFA World Cup as well as the SPECweb2009 suite. The TASP heuristics are found to outperform a power-aware–only server set selection scheme (CPSP), by up to 9.3 % for the simulated scenario. The order of achieved energy efficiency is: MiniMax (9.3 % savings), CP-sLRH (9.2%), bb-sLRH (8.6%), sLRH (5.8%), compared to CPSP.

We present Cypress, a novel framework to archive and query massive time series streams such as those generated by sensor networks, data centers, and scientific computing. Cypress applies multi-scale analysis to decompose time series and to obtain sparse representations in various domains (e.g. frequency domain and time domain). Relying on the sparsity, the time series streams can be archived with reduced storage space. We then show that many statistical queries such as trend, histogram and correlations can be answered directly from compressed data rather than from reconstructed raw data. Our evaluation with server utilization data collected from real data centers shows significant benefit of our framework. 1.

Significant opportunities for power optimization exist at application design stage and are not yet fully exploited by system and application designers. We describe the challenges developers face in optimizing software for energy efficiency by exploiting applicationlevel knowledge. To address these challenges, we propose the development of automated tools that profile the energy usage of various resource components used by an application and guide the design choices accordingly. We use a preliminary version of a tool we have developed to demonstrate how automated energy profiling helps a developer choose between alternative designs in the energyperformance trade-off space. 1.

Abstract—Networking devices today consume a non-trivial amount of energy and it has been shown that this energy consumption is largely independent of the load through the devices. With a strong need to curtail the rising operational costs of IT infrastructure, there is a tremendous opportunity for introducing energy awareness in the design and operation of enterprise and data center networks. We focus on these networks as they are under the control of a single administrative domain in which network-wide control can be consistently applied. In this paper, we describe and analyze three approaches to saving energy in single administrative domain networks, without significantly impacting the networks ’ ability to provide the expected levels of performance and availability. We also explore the tradeoffs between conserving energy and meeting performance and availability requirements. We conduct an extensive case study of our algorithms by simulating a real Web 2.0 workload in a real data center network topology using power characterizations that we obtain from real network hardware. Our results indicate that for our workload and data center scenario, 16 % power savings (with no performance penalty and small decrease in availability) can be obtained merely by appropriately adjusting the active network elements (links). Significant additional savings (up to 75%) can be obtained by incorporating network traffic management and server workload consolidation. I.

RACNet is a sensor network for monitoring a data center’s environmental conditions. The high spatial and temporal fidelity measurements that RACNet provides can be used to improve the data center’s safety and energy efficiency. RACNet overcomes the network’s large scale and density and the data center’s harsh RF environment to achieve data yields of 99 % or higher over a wide range of network sizes and sampling frequencies. It does so through a novel Wireless Reliable Acquisition Protocol (WRAP). WRAP decouples topology control from data collection and implements a token passing mechanism to provide network-wide arbitration. This congestion avoidance philosophy is conceptually different from existing congestion control algorithms that retroactively respond to congestion. Furthermore, WRAP adaptively distributes nodes among multiple frequency channels to balance load and lower data latency. Results from two testbeds and an ongoing production data center deployment indicate that RACNet outperforms previous data collection systems, especially as network load increases.

Abstract—Cloud is the state-of-the-art back-end infrastructure for most large-scale web services. This paper studies what effect energy proportionality has on the energy savings of cloud data center management, under various equipment compositions and power densities. Our findings show that although it is a common expectation that improved energy proportionality should diminish the benefits of power management’s server provisioning, this is not true in all cases. Results show that equipping server provisioning with thermal awareness can keep it as a useful technique when the data center exhibits power consumption heterogeneity and non-uniform heat recirculation phenomena. Index Terms—Energy proportionality; data centers; server provisioning; thermal awareness. I.