Abstract-Server consolidation plays a key role to mitigate the continuous power increase of datacenters. The recent advent of scale-out applications (e.g., web search, MapReduce, etc.) necessitate the revisit of existing server consolidation solutions due to distinctively different characteristics compared to traditional high-performance computing (HPC), i.e., user interactive, latency critical, and operations on large data sets split across a number of servers. This paper presents a power saving solution for datacenters that especially targets the distinctive characteristics of the scale-out applications. More specifically, we take into account correlation information of core utilization among virtual machines (VMs) in server consolidation to lower actual peak server utilization. Then, we utilize this reduction to achieve further power savings by aggressively-yet-safely lowering the server operating voltage and frequency level. We have validated the effectiveness of the proposed solution using 1) multiple clusters of real-life scale-out application workloads based web search and 2) utilization traces obtained from real datacenter setups. According to our experiments, the proposed solution provides up to 13.7% power savings with up to 15.6% improvement of Quality-of-Service (QoS) compared to existing correlation-aware VM allocation schemes for datacenters. I. INTRODUCTIONThe soaring demand for computing has produced as collateral undesirable effect a surge in power consumption of servers and datacenters [1]. Server consolidation [3], which minimizes the number of active servers by packing workloads, or virtual machines (VMs) in a virtualized environment, into the minimal number of active servers, is one of the widely used techniques to reduce the power consumption of datacenters.However, in order to satisfy the performance demand of applications running on servers, server consolidation is usually conducted assuming the worst-case (or peak) utilization [2], [3]. Thus, in order to achieve power consumption without any significant Quality-of-Service (QoS) degradation, many works have presented aggressive consolidation schemes, which pack VMs based on off-peak (e.g., 90th/95th/99th percentile) of server utilization [4], [5]. Recently, correlation of resources utilization patterns among VMs are also exploited, such that, un-correlated VMs are co-located into a server to enable overprovision of VMs under negligible QoS degradation [6]- [9]. Nonetheless, these existing solutions are mostly designed for high-performance computing (HPC) applications and do not

Abstract

No abstract

Abstract-With the rising challenges in heat removal in integrated circuits (ICs), the development of thermal-aware computing architectures and run-time management systems has become indispensable to the continuation of IC design scaling. These thermal-aware design technologies of the future strongly depend on the availability of efficient and accurate means for thermal modeling and analysis. These thermal models must have not only the sufficient accuracy to capture the complex mechanisms that regulate thermal diffusion in ICs, but also a level of abstraction that allows for their fast execution for design space exploration. In this paper, we propose an innovative thermal modeling approach for full-chips that can handle the scalability problem of transient heat flow simulation in large 2-D/3-D multiprocessor ICs. This is achieved by parallelizing the computation-intensive task of transient temperature tracking using neural networks and exploiting the computational power of massively parallel graphics processing units. Our results show up to 35× run-time speedup compared to state-of-the-art IC thermal simulation tools while keeping the error lower than 1°C. Speedups scale with the size of the 3-D multiprocessor ICs and our proposed method serves as a valuable design space exploration tool.

Abstract-Free cooling, i.e., directly using outside cold air and/or water to cool down datacenters, can provide significant power savings of datacenters. However, due to the limited cooling capability, which is tightly coupled with climate conditions, free cooling is currently used only in limited locations (e.g., North Europe) and periods of the year. Moreover, the applicability of free cooling is further restricted along with the conservative assumption on workload characteristics and the virtual machine (VM) consolidation technique as they require to provision higher cooling capability. This paper presents a dynamic power management scheme, which extends the applicability of free cooling by judiciously consolidating VMs exploiting time-varying workload characteristics of datacenter as well as climate conditions, in order to minimize the power consumption of the entire datacenter while satisfying service-level agreement (SLA) requirements. Additionally, we propose the use of a receding horizon control scheme in order to prevent frequent cooling mode transitions. Experimental results show that the proposed solution provides up to 25.7% power savings compared to conventional free cooling decision schemes, which uses free cooling only when the outside temperature is lower than predefined threshold temperature.