We consider the problem of optimizing video delivery for a network supporting video clients streaming stored video. Specifically, we consider the problem of jointly optimizing network resource allocation and video quality adaptation. Our objective is to fairly maximize video clients' Quality of Experience (QoE) realizing tradeoffs among the mean quality, temporal variability in quality, and fairness, incorporating user preferences on rebuffering and cost of video delivery. We present a simple asymptotically optimal online algorithm, NOVA, to solve the problem. NOVA is asynchronous, and using minimal communication, distributes the tasks of resource allocation to network controller, and quality adaptation to respective video clients. Video quality adaptation in NOVA is also optimal for standalone video clients, and is well suited for use with DASH framework. Further, we extend NOVA for use with more general QoE models, networks shared with other traffic loads and networks using fixed/legacy resource allocation. arXiv:1307.7210v3 [cs.NI] 25 Sep 2013 4 models, [14] and [15] study the resource allocation component for video delivery accounting for user dynamics. A major weakness of the aforementioned papers is the limited nature of the associated QoE models (that are essentially just the mean quality) and their lack of flexibility in managing/incorporating user preferences related to rebuffering and cost.While [8] presents a novel algorithm for realizing mean-variability tradeoffs for video delivery (see [18] for genearalizations), the model involves a strong assumption of synchrony-the download of a segment of each video client starts at the beginning of a (network) slot and finishes by the end of the slot. This assumption on synchrony precludes any explicit control over rebuffering as it limits the ability of a video client to get ahead (by downloading more segments) during periods when channel is good and/or network is underloaded. Relaxed/different versions of this assumption can be found in the theoretical frameworks used in many previous papers (e.g., decision making in [16], [12], [13] is synchronous) as it facilitates an easier extension of tools from classical NUM framework. However, this assumption of synchrony is not ideal for DASH-based video clients in a wireless network that operate 'at their own pace'-downloading variable sized segments (with variable download times) one after the other. In this paper, we drop the assumption of synchrony which allows us to exploit opportunism across video clients' state of playback buffer (channels and features of video content like quality rate tradeoffs), and base our adaptation decision concerning a segment on network state information relevant to the download period of the segment. We also tackle the consequent novel technical challenges related to distributed asynchronous algorithms operating in a stochastic setting. Further, the rebuffering constraint in our asynchronous setting effectively induces a new type of constraint involving averages measured over two time...

Clustering is an effective method to increase the available parallelism in VLIW datapaths without incurring severe penalties associated with a large number of register file ports. Efficient utilization of a clustered datapath requires careful binding/assignment of operations to clusters. The article proposes a binding algorithm that effectively explores trade-offs between in-cluster operation serialization and delays associated with data transfers between clusters. Extensive experimental evidence is provided showing that the algorithm generates high quality solutions for representative kernels, with up to 33% improvement over a state-of-the-art binding algorithm.

No abstract

Large scale Content Delivery Networks (CDNs) are one of the key components of today's information infrastructure. This paper proposes and analyzes a simple stochastic model for a file-server system wherein servers can work together, as a pooled resource, to meet individual user requests. In such systems basic questions include: How and where to replicate files? What is the impact of dynamic service allocation across request types, and whether it can provide substantial gains over simpler load balancing policies? What are tradeoffs amongst performance, reliability and recovery costs, and energy? The paper provides both explicit and asymptotic approximations for large systems towards addressing these basic questions.

Network utility maximization (NUM) is a key conceptual framework to study reward allocation amongst a collection of users/entities in disciplines as diverse as economics, law and engineering. However when the available resources and/or users' utilities vary over time, reward allocations will tend to vary, which in turn may have a detrimental impact on the users' overall satisfaction or quality of experience. This paper introduces a generalization of the NUM framework which incorporates the detrimental impact of temporal variability in a user's allocated rewards. It explicitly incorporates tradeoffs amongst the mean and variability in users' reward allocations, as well as fairness across users. We propose a simple online algorithm to realize these tradeoffs, which, under stationary ergodic assumptions, is shown to be asymptotically optimal, i.e., achieves a long term performance equal to that of an offline algorithm with knowledge of the future variability in the system. This substantially extends work on NUM to an interesting class of relevant problems where users/entities are sensitive to temporal variability in their service or allocated rewards. Index Terms-Network utility maximization (NUM).