MulticoreInfo.com has published 1800 posts in 2009 linking to useful resources of multicore related information. Among those, we believe the following top 10 posts stood out in our view and were viewed by many readers. We observe a trend of software efforts to catch up with utilizing multicore processors.
Multicore Research Papers - 2009: Parallel computing, computer architecture conferences publish numerous multicore related papers. We could collect a small set of these papers and posted on this page.
Parallel Programming Tutorial Series: There are a lot of online tutorials on parallel computing. We categorized some of these tutorials into Basic Parallel computing, MapReduce, Cell Processor Programming, OpenMP, pthreads, Intel TBB, MPI, CUDA, and OpenCL. Many readers found these very useful, especially CUDA and OpenCL ones in the wake of heterogeneous computing.
MATLAB Release 2009a Expands Multicore Capabilities: MATLAB released 2009a expanding multicore capabilities, both in the Parallel Computing Toolbox as well as the base MATLAB product.
Another related link to this Matlab release
Snow Leopard in Depth: Grand Central Dispatch: Apple released and open sourced Grand Central Dispatch [pdf] that helps developers more easily write software for multicore systems. This article introduces the technology.
Concurrent and Parallel Are Not The Same: Douglas Eadline writes this article on Linux magazine explaining the difference between concurrent and parallel, which are often used interchangeably.
Comparing Parallel Haskell Implementations for Multicore: Functional programming has been making some progress in becoming a programming paradigm for multicore processors. This paper by Dr. Jost Berthold investigates the differences and trade offs imposed by two parallel Haskell dialects running on multicore machines.
CULA Tools: GPU Accelerated LAPACK: Model of providing libraries to utilize GPUs is growing. There are many CUDA libraries including CUDPP (CUDA Data Parallel Primitives Library), SpMV (Sparse Matrix-Vector Multiplication in CUDA). A lot more libraries can be found at CUDA Zone.
Istanbul versus Nehalem, some notes: AnandTech’s Johan De Gelas explains the differences between AMD’s Istanbul and Intel’s Nehalem architectures.
OpenCL demo on a GPU: This demo of OpenCL on NVIDIA GPU was based on OpenCL API/driver interface.
Vectorization: Writing C/C++ code in VECTOR Format: Vectorization has been key optimization principle for more than a decade. Mukkaysh Srivastav writes an excellent article on writing vectorized C/C++ codes.
Other popular posts from MulticoreInfo.com
Intel may ship 8-core Nehalem EX in early 2010
Multicore Related Books
Multicore Research Papers - 2008
Multicore White Papers
Multicore Related Blogs
Call for Papers: Special Issue of JPDC on “Data Intensive Computing”
Multicore Review: Best Multicore Posts of 2008