1. Cray XD1 Reconfigurable Computing for Application Acceleration

2. 2 Cray XD1 System Architecture Compute 12 AMD Opteron 32/64 bit, x86 processors High Performance Linux RapidArray Interconnect 12 communications processors 1 Tb/s switch fabric Active Management Dedicated processor Application Acceleration 6 co-processors Processors directly connected via integrated switch fabric 50 GFLOPS - 2+ TFLOPS 12 – 1024+ processors Entry/Mid range system optimized for sustained performance With reconfigurable computing capability

3. 3 Naval Research Lab 24 chassis / 2.5 TeraFlops System 288 AMD Dual-core Processors 144 Xilinx Virtex-II Pro Italy’s national supercomputer center 24 Affiliated universities & the National Research Council 634 GFLOP Cray XD1 system (144 Processor) with Application Acceleration FPGAs CINECA Air Force Maui Optical and Supercomputing Site 1.3 TFLOP Cray XD1 system (288 Processor) Part of $23M contract awarded to Cray from DoD HPC Modernization Program. Located at Maui Space Surveillance Site

4. 4 The Barriers to Reconfigurable Computing 0. Choosing the Right Application! Core functionality, increasing complexity, increasing demand on resources 1. Starving the FPGA Bandwidth and latency to the FPGA limited by PCI bus 2. FPGA, Processor Interaction Job scheduling, Linux integration, memory mapping 3. Programming Tools Programming hardware requires special tools, special expertise.

5. 5 The Right Application…. Industries  Life Sciences  Electronic Design Automation  Manufacturing  Energy & Natural Resources  Media  Government Example Applications  Seismic Processing  Astrophysics / Adaptive Optics  Graphics Acceleration  Quantum Physics  Bioinformatics  Cheminformatics  Vehicular Traffic Simulations  Financial Modeling  And many more Example Functions  Searching & Sorting  Signal & Image Processing  Encryption/Decryption  Error Correction  Coding/Decoding  Packet Processing  Random-number Generation  Bit Manipulation  And many more 1. Performance! 2. Performance! 3. Performance! 4. Lower Power Density 5. Deployments in the Field, Mobile Platforms, … Why FPGAs in a Cray product?

6. 6 Celoxica C->FPGA Compiler Handel-C compiler allows FPGA development in C (with extensions) It does not eliminate the need to have hardware awareness! Enables single programming model for Logic and PowerPCs Example: Mersennes Twister RNG 3 days to port onto FPGA (performance << Opteron!!) 2-3 weeks to optimize End result roughly comparable to VHDL development Beta Test underway with OSC index = 0; while (index < length){ if(table[index] = key) found=index; else index = index+1; }

7. 7 Mitrion The application: Thin Plate Splines -- image analysis of protein gels Image morphing based on natural logarithm computations Essential for comparing protein content Speedup per FPGA: 10-30x. Reduces analysis runtime from days to hours. 180 lines of Mitrion-c code generates 150,000 lines of VHDL code Speedup per FPGA: 10-30x Pure software programming – easy to learn for an HPC programmer No hardware design considerations

8. 8 Higher Level Abstractions: Mobius (www.codetronix.com) Pascal-like CSP based language (Types,records, arrays, fp arithmetic) Synchronization and communication by handshaking over channels Generate HW, SW or HW/SW code General purpose & dataflow algorithms Pipelined DES: ~ 120 lines of Mobius ~ 2 GB/s thruput ~ 2200 slices DES

9. 9 Easing FPGA Adoption … 1. Traditional Programming Model VHDL, Verilog 2. Off-The-Shelf Libraries Cray and third party acceleration libraries Prepackaged, turnkey applications 3. High-Level Compilers C, Graphical, Matlab Working to Create an Open Source FPGA Community