Demand for data center compute resources is growing at an unprecedented rate, driven by the need to deliver, manipulate and analyze enormous amounts of data generated within the cloud. AMD draws on more than a decade of x86 innovation, harnesses the power of industry-leading graphics technology, and is pioneering the 64-bit ARM server space to more efficiently execute some of the most challenging server workloads. The AMD Server Zone exists to give developers the tools, documentation, and forum they need to optimize existing applications as well as create the next generation of workloads powering the datacenter. Welcome to a unique community that is transforming server processing and re-imagining the server. Read below for resources, documentation and articles for AMD server processors. Also, check out the Developer Resources and Blog sites. There you will find the full catalog of drivers, articles, instructional videos and more.
Use the right tool for the right job. These software development tools and resources have been optimized for AMD Server processors:
Runtimes & Compilers
Libraries & Tools
- AMD Core Math Library (ACML) Highly tuned implementations of BLAS and CFFT complex-complex Fast Fourier Transforms for x86 CPUs. APU version now in beta
- clMath OpenCL implementations of BLAS and FFT
- libM Optimized implementations of standard C99 math functions
- CodeXL Powerful GPU debugger, GPU and CPU profiler, and OpenCL analyzer (C, C++, OpenCL)
- The Emerging ARM 64 Server Software Ecosystem
- Compiler Options Quick Reference Guide for AMD Opteron™ X1100 series and AMD Opteron™ X2100 series APU
- Compiler Options Quick Reference Guide for AMD Opteron™ 6300 Series Processors (“Abu Dhabi”) & Opteron™ 4300 Series Processors (“Seoul”)
- “Bulldozer” and “Piledriver” Instructions Guide
- RedHat Fedora 21 and Xen 4.5 Ready for AMD Opteron A1100 Series Processors
- The ARM Server Ecosystem Continues to Gain Momentum
- ACML 6.1 Accelerates your Existing Code AMD Core Math Library 6.1 continues to improve our support for heterogeneous acceleration of your code.
- ArrayFire is Now Open Source ArrayFire has released their flagship accelerated computing software library as open source.
- PGI Accelerator Compilers Add Support for AMD APUs and GPUs PGI Beta release supports OpenACC directive-based accelerator programming for AMD APUs and discrete GPUs
- AMD releases APPML source code, creates clMath library AMD is opening the source to the APPML project and making it available on GitHub under the name clMath
- … More Articles
AMD server products now span three different architectures. Be part of the server transformation – visit below for product specs and documentation:
- AMD Opteron™ A1100-Series developer kit now available
- AMD and I/O Virtualization 2.0 on “Magny-Cours” Processors
- Magny-Cours” and Direct Connect Architecture 2.0 The increase in cores to 8 and 12 on each processor is made possible by our Direct Connect Architecture. This technical article outlines what enhancements were made and how they will benefit your code
- AMD and PathScale Join the OpenACC Standards Group 11/17/2014 By joining the OpenACC standards group, AMD and PathScale bring considerable expertise in high performance computing technology and programming capabilities, providing developers with more diversity and choice in computing architectures and software tools, and promising to extend the reach of OpenACC-supported solutions to new and existing users worldwide.
- AMD FirePro Awarded Top Spot on the Green500 List 11/20/2014 With 5.27 GFLOPS per watt, the GSI L-CSC cluster using AMD FirePro S9150 server GPUs delivers 20 percent better efficiency than the top ranked cluster on the June 2014 Green500 List.
- AMD Demonstrates First Network Function Virtualization Solution on 64-bit AMD and ARM Technology 10/1/2014 The AMD Embedded R-Series SoC (“Hierofalcon”), sampling to customers, transforms telecommunications infrastructure with software partners
- AMD Unleashes World’s Most Powerful Server GPU for HPC 8/6/2014 AMD FirePro™ S9150 server card delivers up to 2.53 TFLOPS of peak double-precision floating point performance
- AMD Announces the Availability of 64-bit ARM Opteron Developer Kits 7/30/2014
- AMD’s SeaMicro SM15000™ Server Sets Industry Benchmark Record for Hyperscale OpenStack Clouds 5/13/2014 Canonical Ubuntu OpenStack 14.04 (Icehouse) on SeaMicro SM15000 Server Scales to 168,000 Virtual Machines
- AMD Unveils Ambidextrous Computing Roadmap 5/5/2014 Announces 64-bit ARM Core Architecture License and Future “K12” ARM-based Core; Pin-Compatible x86 and ARM Processors; Demonstrates AMD Opteron A-Series 64-bit ARM Server CPU Codename “Seattle”
For more articles visit the AMD Newsroom