Practical Data Center Power Proportionality
نویسندگان
چکیده
Energy accounts for a significant fraction of the operational costs of a data center and data center owners are increasingly interested in moving towards power proportional designs. However, we show that current power proportional solutions are inherently limited in the energy savings they can offer; the reason is that they focus on managing the data center IT equipment (disks, servers) and neglect the large power overheads resulting from non-IT infrastructure such as cooling, and power distribution equipment. We propose a solution, where the power management unit is as large as a rack, or even an entire containerized data center. We show that it is practical today to design data centers to power down idle racks or containers – and in fact, current online service trends strongly enable this model. Finally, we show that energy savings increase by an order of magnitude with larger power management units, while performance remains unaffected.
منابع مشابه
Adaptive Subcarrier Assignment and Power Distribution in Multiuser OFDM Systems with Proportional Data Rate Requirement
A low complexity dynamic subcarrier and power allocation methodology for downlink communication in an OFDM-based multiuser environment is developed. The problem of maximizing overall capacity with constraints on total power consumption, bit error rate and data rate proportionality among users requiring different QOS specifications is formulated. Assuming perfect knowledge of the instantaneo...
متن کاملPower Consumption Evaluation of Distributed Computing Network Considering Traffic Locality
When computing resources are consolidated in a few huge data centers, a massive amount of data is transferred to each data center over a wide area network (WAN). This results in increased power consumption in the WAN. A distributed computing network (DCN), such as a content delivery network, can reduce the traffic from/to the data center, thereby decreasing the power consumed in the WAN. In thi...
متن کاملTowards Energy-Proportional Computing Using Subsystem-Level Power Management
Massive data centers housing thousands of computing nodes have become commonplace in enterprise computing, and the power consumption of such data centers is growing at an unprecedented rate. Adding to the problem is the inability of the servers to exhibit energy proportionality, i.e., provide energy-efficient execution under all levels of utilization, which diminishes the overall energy efficie...
متن کاملExploratory assessment of dose proportionality: review of current approaches and proposal for a practical criterion.
This article reviews currently used approaches for establishing dose proportionality in Phase I dose escalation studies. A review of relevant literature between 2002 and 2006 found that the power model was the preferred choice for assessing dose proportionality in about one-third of the articles. This article promotes the use of the power model and a conceptually appealing extension, i.e. a cri...
متن کاملOn the Energy Proportionality of Scale-Out Workloads
Our increasing reliance on the cloud has led to the emergence of scale-out workloads. These scale-out workloads are latency-sensitive as they are user driven. In order to meet strict latency constraints, they require massive computing infrastructure, which consume significant amount of energy and contribute to operational costs. This cost is further aggravated by the lack of energy proportional...
متن کامل