GPU parallelism for real applications can achieve enormous performance gain. CPU-GPU Communication is one of the major bottlenecks that limit this performance gain. Among several libraries developed so far to optimize this communication, DyManD (Dynamically Managed Data) provides better communication optimization strategies and achieves better performance on a single GPU. SmithWaterman is a wel...

Energy consumption is an increasingly important concern in data centers. Today, nearly half of the energy in data centers is consumed by the cooling infrastructure. Existing policies on thermally-aware workload allocation do not consider applications that include many tasks (or threads) running on a large set of nodes with significant communication among the tasks. Such jobs, however, constitut...

Whenever multiple stakeholders try to optimize a common objective function in a distributed way, an adroit coordination mechanism is necessary. This contribution presents a formal model of distributed combinatorial optimization problems. Subsequently, a heuristic is introduced, that uses self-organizing mechanisms to optimize a common global objective as well as individual local objectives in a...

The multiplication of a vector by a sparse matrix is an important kernel in scientific computing. We study how to optimize the performance of this operation in parallel by reducing communication. We review existing approaches and present a new two-dimensional partitioning method for symmetric matrices, called corner partitioning. Our method is simple and can be implemented using existing softwa...

1 Natura l Emergence of Optimized Configurations Every day, enormous efforts are devoted to organizing the supply and demand of limited resources, so as to optimize their utility. Examples include the supply of foods and services to consumers, the scheduling of a t ransportat ion fleet, or the flow of information in communication networks within society or within a parallel computer. By contras...

We investigate services giving users an adequate insight on his or her energy consumption habits in order to optimize it in the long run. The explored energy awareness services are addressed to inhabitants of smart homes, equipped with smart meters, advanced communication facilities, sensors and actuators. To analyze the potential of such services, a game at a social network Facebook has been d...

We optimize the design of a short-range communication system using nondirected line-of-sight IR radiation. We propose a receiver structure comprising a spherical thin-film optical filter and a truncated spherical lens that can significantly outperform an optimized planar-filter system. We can make the passband of the spherical filter arbitrarily narrow without constraining the field of view by ...

This paper presents Quo Vadis, an evolving framework for intelligent tra c management in very large communication networks. Quo Vadis is designed to exploit topological properties of large networks as well as their spatio-temporal dynamics to optimize multiple performance criteria through cooperation among nodes in the network. It employs a distributed representation of network state informatio...

The Internet of Things (IoT) is the next big wave in computing characterized by large scale open ended heterogeneous network of things, with varying sensing, actuating, computing and communication capabilities. Compared to the traditional field of autonomic computing, the IoT is characterized by an open ended and highly dynamic ecosystem with variable workload and resource availability. These c...