AMD Performance Libraries and Math Libraries work great within Visual Studio and they give you quick access to multi-threaded functions and sophisticated math routines.
» Framewave Project
» AMD Core Math Library (ACML)

Continuing our software and driver leadership, AMD has just released the ATI Catalyst™ Windows 7 Preview driver to coincide with the pre-beta release of Microsoft's Windows 7 OS.
» ATI Catalyst™ Windows 7 Preview Driver

Downloads from MSDN:
» Overview of Visual Studio 2008
» Try Visual Studio 2008

Visual Studio 2008 includes SSE4a intrinsic support specifically for “Barcelona” and “Shanghai” processors:

» __lzcnt16, __lzcnt, __lzcnt64
» __popcnt16, __popcnt, __popcnt64
» __mm_stream_ss , __stream_sd
» __ rdtscp
» _mm_extract_si64, _mm_extracti_si64
» _mm_insert_si64, _mm_inserti_si64

“Barcelona” and “Shanghai” processors (CPUID Family 10h), including Third Generation AMD Opteron™ processors, also now support the use of these intrinsic functions available in Visual Studio 2008:
» _InterlockedCompareExchange128
» _mm_monitor & _mm_mwait

PDC 2008 AMD In-Booth Theater Presentations
» Multi-core processors are here, but how do you resolve bottlenecks in native code?
» AMD Developer Central, Tools & Resources for Software Developers
» Quad-core AMD Opteron™ features and roadmaps
» Virtualization, An overview of the hardware features that give rise to world-class virtualization and cloud computing

Microsoft® Visual Studio® 2008 and AMD Opteron™ processors provide an exciting combination for Microsoft Windows® and .NET application developers
» The Next-Generation Development and Deployment Platform

Using AMD’s Performance Library on Windows®
» How To Get Smashing Performance

.NET Resources
» .NET Framework 3.5 Service Pack 1

Microsoft and AMD continue to collaborate with each new product release. Read this article to learn more about what AMD and Microsoft® did in Windows® 7 to improve power, graphics and virtualization.
» Ten Things Developers Should Know About Windows 7

Download the Mandelbrot code set and get hands-on training in 64-bit performance optimization and parallel programming techniques in a Windows environment in this two part series by Michael Wall, Principal Member of Technical Staff, AMD Inc.
» Performance Optimization of Windows Applications on AMD Processors, Part II
» Performance Optimization of Windows Applications on AMD Processors, Part I

Get the latest compiler flag recommendations and see how to get the most out of your code on “Barcelona” and “Shanghai” processors with Visual Studio 2008.
» Develop Blazing Fast Code with Microsoft Visual Studio® 2008 and AMD Tools

Microsoft® Windows® Compute Cluster Server 2003 is opening up new options in the high-performance computing world. This article explores the architecture and network topology of CCS, the core networking protocols that it relies upon, the use of key Microsoft technologies and the application development process.
» Power Processing with the Microsoft Windows Compute Cluster Server 2003

AMD brought NUMA back to the x86 architecture. There are some distinct advantages that can be had given that you know what to do. This article explores Windows NUMA APIs and shows you how large server applications can get the best performance.
» The Windows NUMA API—What It Is and Why You Care

Microsoft Vista supports multicore processing in more ways than one. This article looks at the future of processor cores and what that means for parallel programming for Vista based operating systems.
» Exploiting Multi-Core Processors in Windows Vista

NUMA allows memory bandwidth to scale up naturally to feed multiple processors. Get the scoop on how to optimize your code for this memory architecture found in AMD processors.
» NUMA optimization in Windows Applications