Altera's Stratix IV FPGAs Deliver Unprecedented Acceleration for World’s Most Powerful Reconfigurable Supercomputer
High-End FPGAs Double the System Memory and Acceleration While Reducing Power Consumption in CHREC's Novo-G Supercomputer
San Jose, Calif., November 11, 2011—Altera Corporation (NASDAQ: ALTR) today announced that the Center for High-Performance Reconfigurable Computing (CHREC) has selected the company's Stratix® IV FPGAs for its latest update to the Novo-G supercomputer. Altera's Stratix IV devices enabled CHREC to increase the system memory and acceleration of the world's most powerful supercomputer by two times. Additionally, CHREC was able to realize a greater performance-to-cost ratio with modest increases in size, power and cooling—a feat rarely achieved in conventional supercomputer upgrades.
"CHREC chose Altera Stratix IV FPGAs to provide our international team of academic researchers with a faster and more efficient platform for testing applications and tools that showcase the advantages of reconfigurable computing on a large scale," said Dr. Alan George, professor of ECE at the University of Florida and CHREC director. "Altera's devices uniquely provided the performance, scalability, power and cooling advantages we needed, as well as the ability to combine the versatility, flexibility and efficiency of a general-purpose processor."
CHREC is the world's leading research body focused on reconfigurable computing. Comprised of more than 30 organizations including Altera, the body is made up of representatives from the academic, industry and government sectors with a common interest in reconfigurable computing for a broad range of applications. The Novo-G supercomputer typically accelerates new algorithms by 10X or more in bioinformatics, signal and image processing and other applications where large amounts of data need to be managed and analyzed.
Stratix IV FPGAs were deployed on PROCStar-IV boards from GiDEL in the Novo-G to enable a scaling up (increased performance per node) in the supercomputer configuration and to double the overall system memory. This configuration also scales out (to many nodes) for use in large data centers where the inherent advantages of Altera FPGAs like parallelism, pipelining, bit manipulation, data streaming and lower power take place.
"In the information age, the need for algorithm acceleration for processing data is growing exponentially. The Stratix IV FPGAs in CHREC's Novo-G system are allowing academic research teams to double the FPGA logic and attached memory within the same data center space," said David Gamba, director of the computer applications business unit at Altera. "Altera is proud to be sponsoring the innovative research being done by CHREC and the Novo-G members."
About Altera
Altera programmable solutions enable system and semiconductor companies to rapidly and cost-effectively innovate, differentiate and win in their markets. Find out more about Altera's FPGA, CPLD and ASIC devices at www.altera.com.
Related Semiconductor IP
- Specialized Video Processing NPU IP for SR, NR, Demosaic, AI ISP, Object Detection, Semantic Segmentation
- Ultra-Low-Power Temperature/Voltage Monitor
- Multi-channel Ultra Ethernet TSS Transform Engine
- Configurable CPU tailored precisely to your needs
- Ultra high-performance low-power ADC
Related News
- Altera Starts Shipping Its 40-nm Stratix IV FPGA Family
- Altera Announces Stratix IV GT and Arria II GX FPGAs: Expands Industry's Broadest Integrated Transceiver Portfolio
- Altera's 40-nm Stratix IV GX FPGAs Achieve PCI-SIG Compliance for the PCI Express 2.0 Architecture
- Altera Delivers Stratix IV GX Transceiver Signal Integrity Development Kit
Latest News
- 10xEngineers and Andes Enable High-Performance AI Compilation for RISC-V AX46MPV Cores
- GUC Announces Tape-out of UCIe 64G IP on TSMC N3P Technology
- Caspia Launches New RTL Security Analyzer Enabling Agentic Silicon Security Verification
- Europe’s stealth leading-edge process technology
- Combined CapEx of Top Eight CSPs to Exceed $710 Billion in 2026; Google Leads ASIC Deployment with TPUs