Cloud data centers commonly use virtualization technologies to provision compute capacity with a level of indirection between virtual machines and physical resources. In this paper we explore the use of that level of indirection as a means for autonomic data center configuration optimization and propose a sensor-actuator model to capture optimization-relevant relationships between data center events, monitored metrics (sensors data), and management actions (actuators). The model characterizes a wide spectrum of actions to help identify the suitability of different actions in specific situations, and outlines what (and how often) data needs to be monitored to capture, classify, and respond to events that affect the performance of data center operations.

Dynamic Voltage and Frequency Scaling (DVFS), CPU pinning, horizontal, and vertical scaling, are four techniques that have been proposed as actuators to control the performance and energy consumption on data center servers. This work investigates the utility of these four actuators, and quantifies the power-performance tradeoffs associated with them. Using replicas of the German Wikipedia running on our local testbed, we perform a set of experiments to quantify the influence of DVFS, vertical and horizontal scaling, and CPU pinning on end-to-end response time (average and tail), throughput, and power consumption with different workloads. Results of the experiments show that DVFS rarely reduces the power consumption of underloaded servers by more than 5%, but it can be used to limit the maximal power consumption of a saturated server by up to 20% (at a cost of performance degradation). CPU pinning reduces the power consumption of underloaded server (by up to 7%) at the cost of performance degradation, which can be limited by choosing an appropriate CPU pinning scheme. Horizontal and vertical scaling improves both the average and tail response time, but the improvement is not proportional to the amount of resources added. The load balancing strategy has a big impact on the tail response time of horizontally scaled applications.

Server power capping limits the power consumption of a server to not exceed a specific power budget. This allows data center operators to reduce the peak power consumption at the cost of performance degradation of hosted applications. Previous work on server power capping rarely considers Quality-of-Service (QoS) requirements of consolidated services when enforcing the power budget. In this paper, we introduce ALPACA, a framework to reduce QoS violations and overall application performance degradation for consolidated services. ALPACA reduces unnecessary high power consumption when there is no performance gain, and divides the power among the running services in a way that reduces the overall QoS degradation when the power is scarce. We evaluate ALPACA using four applications: MediaWiki, SysBench, Sock Shop, and CloudSuite’s Web Search benchmark. Our experiments show that ALPACA reduces the operational costs of QoS penalties and electricity by up to 40% compared to a non optimized system. 

Cloud computing infrastructures are powering most of the web hosting services that we use at all times. A recent failure in the Amazon cloud infrastructure made many of the website that we use on a hourly basis unavailable¹. This illustrates the importance of cloud applications being able to absorb peaks in workload, and at the same time to tune their power requirements to the power and energy capacity offered by the data center infrastructure. In this paper we combine an established technique for response time control — brownout — with power capping. We use cascaded control to take into account both the need for predictability in the response times (the inner loop), and the power cap (the outer loop). We execute tests on real machines to determine power usage and response times models and extend an existing simulator. We then evaluate the cascaded controller approach with a variety of workloads and both open- and closed-loop client models.

Constantly growing power consumption of data centers is a major concern from environmental and economical reasons. Current approaches to reduce negative consequences of high power consumption focus on limiting the peak power consumption. During high workload periods, power consumption of highly utilized servers is throttled to stay within the power budget. However, the peak power reduction affects performance of hosted applications and thus leads to Quality of Service violations. In this paper, we introduce Power Shepherd, a hierarchical system for application performance aware power shifting. Power Shepherd reduces the data center operational costs by redistributing the available power among applications hosted in the cluster. This is achieved by, assigning server power budgets by the cluster controller, enforcing these power budgets using Running Average Power Limit (RAPL), and prioritizing applications within each server by adjusting the CPU scheduling configuration. We implement a prototype of the proposed solution and evaluate it in a real testbed equipped with power meters and using representative cloud applications. Our experiments show that Power Shepherd has potential to manage a cluster consisting of thousands of servers and limit the increase of operational costs by a significant amount when the cluster power budget is limited and the system is overutilized. Finally, we identify some outstanding challenges regarding model sensitivity and the fact that this approach in its current from is not beneficial to be used in all situations, e.g., when the system is underutilized.

Power budgeting is a commonly employed solution to reduce the negative consequences of high power consumption of large scale data centers. While various power budgeting techniques and algorithms have been proposed at different levels of data center infrastructures to optimize the power allocation to servers and hosted applications, testing them has been challenging with no available simulation platform that enables such testing for different scenarios and configurations. To facilitate evaluation and comparison of such techniques and algorithms, we introduce a simulation model for Quality-of-Service aware power budgeting and its implementation in CloudSim. We validate the proposed simulation model against a deployment on a real testbed, showcase simulator capabilities, and evaluate its scalability.