Optimal multicommodity flow through the complete graph with random edge capacities

We consider a multicommodity flow problem on a complete graph whose edges have random, independent, and identically distributed capacities. We show that, as the number of nodes tends to infinity, the maximumutility, given by the average of a concave function of each commodity How, has an almost-sure limit. Furthermore, the asymptotically optimal flow uses only direct and two-hop paths, and can be obtained in a distributed manner.

Bandwidth management in a hybrid wireless network: For superstore applications

The integration of different wireless networks, such as GSM and WiFi, as a two-tier hybrid wireless network is more popular and economical. Efficient bandwidth management, call admission control strategies and mobility management are important issues in supporting multiple types of services with different bandwidth requirements in hybrid networks. In particular, bandwidth is a critical commodity because of the type of transactions supported by these hybrid networks, which may have varying bandwidth and time requirements. In this paper, we consider such a problem in a hybrid wireless network installed in a superstore environment and design a bandwidth management algorithm based on the priority level, classification of the incoming transactions. Our scheme uses a downlink transaction scheduling algorithm, which decides how to schedule the outgoing transactions based on their priority level with efficient use of available bandwidth. The transaction scheduling algorithm is used to maximize the number of transaction-executions. The proposed scheme is simulated in a superstore environment with multi Rooms. The performance results describe that the proposed scheme can considerably improve the bandwidth utilization by reducing transaction blocking and accommodating more essential transactions at the peak time of the business.

Believe it or Not! Multicore CPUs can Match GPUs for FLOP-intensive Applications

In this work, we evaluate performance of a real-world image processing application that uses a cross-correlation algorithm to compare a given image with a reference one. The algorithm processes individual images represented as 2-dimensional matrices of single-precision floating-point values using O(n4) operations involving dot-products and additions. We implement this algorithm on a nVidia GTX 285 GPU using CUDA, and also parallelize it for the Intel Xeon (Nehalem) and IBM Power7 processors, using both manual and automatic techniques. Pthreads and OpenMP with SSE and VSX vector intrinsics are used for the manually parallelized version, while a state-of-the-art optimization framework based on the polyhedral model is used for automatic compiler parallelization and optimization. The performance of this algorithm on the nVidia GPU suffers from: (1) a smaller shared memory, (2) unaligned device memory access patterns, (3) expensive atomic operations, and (4) weaker single-thread performance. On commodity multi-core processors, the application dataset is small enough to fit in caches, and when parallelized using a combination of task and short-vector data parallelism (via SSE/VSX) or through fully automatic optimization from the compiler, the application matches or beats the performance of the GPU version. The primary reasons for better multi-core performance include larger and faster caches, higher clock frequency, higher on-chip memory bandwidth, and better compiler optimization and support for parallelization. The best performing versions on the Power7, Nehalem, and GTX 285 run in 1.02s, 1.82s, and 1.75s, respectively. These results conclusively demonstrate that, under certain conditions, it is possible for a FLOP-intensive structured application running on a multi-core processor to match or even beat the performance of an equivalent GPU version.

Efficient Algorithms for Combinatorial Auctions with Volume Discounts Arising in Web Service Composition

Web services are now a key ingredient of software services offered by software enterprises. Many standardized web services are now available as commodity offerings from web service providers. An important problem for a web service requester is the web service composition problem which involves selecting the right mix of web service offerings to execute an end-to-end business process. Web service offerings are now available in bundled form as composite web services and more recently, volume discounts are also on offer, based on the number of executions of web services requested. In this paper, we develop efficient algorithms for the web service composition problem in the presence of composite web service offerings and volume discounts. We model this problem as a combinatorial auction with volume discounts. We first develop efficient polynomial time algorithms when the end-to-end service involves a linear workflow of web services. Next we develop efficient polynomial time algorithms when the end-to-end service involves a tree workflow of web services.

Resource usage monitoring for KVM based virtual machines

Realization of cloud computing has been possible due to availability of virtualization technologies on commodity platforms. Measuring resource usage on the virtualized servers is difficult because of the fact that the performance counters used for resource accounting are not virtualized. Hence, many of the prevalent virtualization technologies like Xen, VMware, KVM etc., use host specific CPU usage monitoring, which is coarse grained. In this paper, we present a performance monitoring tool for KVM based virtualized machines, which measures the CPU overhead incurred by the hypervisor on behalf of the virtual machine along-with the CPU usage of virtual machine itself. This fine-grained resource usage information, provided by the above tool, can be used for diverse situations like resource provisioning to support performance associated QoS requirements, identification of bottlenecks during VM placements, resource profiling of applications in cloud environments, etc. We demonstrate a use case of this tool by measuring the performance of web-servers hosted on a KVM based virtualized server.

Eco-friendly wood polymer composites for sustainable design applications

Environmental inputs can improve the level of innovation by interconnecting them with traditional inputs regarding the properties of materials and processes as a strategic eco-design procedure. Advanced engineered polymer composites are needed to meet the diverse needs of users for high-performance automotive, construction and commodity products that simultaneously maximize the sustainability of forest resources. In the current work, wood polymer composites (WPC) are studied to promote long-term resource sustainability and to decrease environmental impacts relative to those of existing products. A series of polypropylene wood–fiber composite materials having 20, 30, 40 and 50 wt. % of wood–fibers were prepared using twin-screw extruder and injection molding machine. Tensile and flexural properties of the composites were determined. Polypropylene (PP) as a matrix used in this study is a thermoplastic material, which is recyclable. Suitability of the prepared composites as a sustainable product is discussed.