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.

Population crunch in India: is it urban or still rural?

The article attempts to present analysis based on the provisional results of the Census 2011. While there is no doubt that the human social organization of the country is undergoing a transition, the nature of growth however is subject to the lens through which this is viewed. Noting the dichotomy of urban and rural definitions, we question the rationality of the ‘urban’ definition and its relevance.

Computational investigations on covalent dimerization/oligomerization of polyacenes: Is it relevant to soot formation?

We have postulated a novel pathway that could assist in the nucleation of soot particles through covalent dimerization and oligomerizations of a variety of PAHs. DFT calculations were performed with the objective of obtaining the relative thermal stabilities and formation probabilities of oligomeric species that exploit the facile dimerization that is known to occur in linear oligoacenes. We propose that the presence of small stretches of linear oligoacence (tetracene or longer) in extended PAH, either embedded or tethered, would be adequate for enabling the formation of such dimeric and oligomeric adducts; these could then serve as nuclei for the growth of soot particles. Our studies also reveal the importance of p-stacking interactions between extended aromatic frameworks in governing the relative stabilities of the oligomeric species that are formed. (c) 2012 Wiley Periodicals, Inc.