Future servers will incorporate many active low-power modes for different system components, such as cores and memory. Though these modes provide flexibility for power management via Dynamic Voltage and Frequency Scaling (DVFS), they must be operated in a coordinated manner. Such coordinated control creates a combinatorial space of possible power mode configurations. Given the rapid growth of the number of cores, it is becoming increasingly challenging to quickly select the configuration that maximizes the performance under a given power budget. Prior power capping techniques do not scale well to large numbers of cores, and none of those works has considered memory DVFS. In this paper, we present FastCap, our optimization approach for system-wide power capping, using both CPU and memory DVFS. Based on a queuing model, FastCap formulates power capping as a non-linear optimization problem where we seek to maximize the system performance under a power budget, while promoting fairness across applications. Our FastCap algorithm solves the optimization online and efficiently (low complexity on the number of cores), using a small set of performance counters as input. To evaluate FastCap, we simulate it for a many-core server running different types of workloads. Our results show that FastCap caps power draw accurately, while producing better application performance and fairness than many existing CPU power capping methods (even after they are extended to use of memory DVFS as well).

Energy consumption is the main obstacle to the green cloud computing, particularly with the global climate warming and data center scale expanding. The green computing is gaining more and more attention because energy consumption increased rapidly. We propose a cloud computing management framework in order to ensure energy consumption of cloud computing to be a minimum. The framework uses virtual date center instead of virtual machine service mode of traditional service providers, and partition the virtual data center used in the management framework is based on the network communications of virtual machines. Then the virtual data center partition is placed into corresponding green data center in order to maximize revenue of providers and minimize the carbon emissions.

Power consumption in data centers has been growing significantly in recent years. To reduce power, servers are being equipped with increasingly sophisticated power management mechanisms. Different mechanisms offer dramatically different trade-offs between power savings and performance penalties. Considering the complexity, variety, and temporally-varying nature of the applications hosted in a typical data center, intelligently determining which power management policy to use and when is a complicated task. In this paper we analyze a system model featuring both performance scaling and low-power states. We reveal the interplay between performance scaling and low-power states via intensive simulation and analytic verification. Based on the observations, we present SleepScale, a runtime power management tool designed to efficiently exploit existing power control mechanisms. At run time, SleepScale characterizes power consumption and quality-of-service (QoS) for each low-power state and frequency setting, and selects the best policy for a given QoS constraint. We evaluate SleepScale using workload traces from data centers and achieve significant power savings relative to conventional power management strategies.

We study how coding in distributed storage reduces expected download time, in addition to providing reliability against disk failures. The expected download time is reduced because when a content file is encoded with redundancy and distributed across multiple disks, reading only a subset of the disks is sufficient for content reconstruction. For the same total storage used, coding exploits the diversity in storage better than simple replication, and hence gives faster download. We use a novel fork-join queueing framework to model multiple users requesting the content simultaneously, and derive bounds on the expected download time. Our system model and results are a novel generalization of the fork-join system that is studied in queueing theory literature. Our results demonstrate the fundamental trade-off between the expected download time and the amount of storage space. This trade-off can be used for design of the amount of redundancy required to meet the delay constraints on content delivery.

In this paper, we aim to develop scheduling policies to maximize the stability region of a wireless network under the assumption that mutual information accumulation is implemented at the physical layer. This enhanced physical layer capability enables the system to accumulate information even when the link between two nodes is not good and a packet cannot be decoded within a slot. The result is an expansion of the stability region of the system. The accumulation process does not satisfy the i.i.d assumption that underlies many previous analysis in this area. Therefore it also brings new challenges to the problem. We propose two dynamic scheduling algorithms to overcome this difficulty. One performs scheduling every T slot, which inevitably increases average delay in the system, but approaches the boundary of the stability region. The second constructs a virtual system with the same stability region. Through controlling the virtual queues in the constructed system, we avoid the non-i.i.d difficulty and attain the stability region. We derive performance bounds under both algorithms and compare them through simulation results.

We design and analyze a method to extract secret keys from the randomness inherent to wireless channels. We study a channel model for a multipath wireless channel and exploit the channel diversity in generating secret key bits. We compare the key extraction methods based both on entire channel state information (CSI) and on single channel parameter such as the received signal strength indicators (RSSI). Due to the reduction in the degree-of-freedom when going from CSI to RSSI, the rate of key extraction based on CSI is far higher than that based on RSSI. This suggests that exploiting channel diversity and making CSI information available to higher layers would greatly benefit the secret key generation. We propose a key generation system based on low-density parity-check (LDPC) codes and describe the design and performance of two systems: one based on binary LDPC codes and the other (useful at higher signal-to-noise ratios) based on four-ary LDPC codes.

This study deals with signal transmission and reception over a fast-fading wireless environment. Conventional methods usually consider the rapid change of channel to be harmful, which may decrease the performance of existing systems. There are also methods proposed to make use of the cause of degradation in existing systems – Doppler spread – to provide additional diversity, at the cost of folds increased complexity. In this study, the authors focus on exploiting the array gains provided by the fast-fading wireless channel multipath array gain (MAG) and Doppler array gain (DAG). The conceptions of MAG and DAG are straightforward extensions of the conception of array gain in multi-antenna systems. For methods which exploit multipath and Doppler diversities, complex signal processing is required at the receiver side, leading to a great increase in complexity. The MAG and DAG can be easily obtained using simple transmit waveform design. Therefore, no extra computational load is required. Both analytical and simulation results show that, with the same computational complexity, the proposed methods can significantly outperform the conventional ones.