KALRAY announces world's most efficient manycore PCI Express board for intensive computing application at ISC 2014
High Density, Low power PCI Express board, 50W TDP for 1 TFLOPS
Orsay, Paris -- June 23, 2014 — KALRAY, the developer of the low-power, high-performance MPPA MANYCORE processor, today announced at ISC 2014 in Leipzig, the MPPA TURBOCARD2, the world’s most efficient 1024-core, 32GB DDR PCIe board. The TURBOCARD2 is based on KALRAY MPPA-256 manycore processor and provides a unique solution to boost servers and workstations used in Oil & Gas, Computational Finance, Cryptanalysis, Numerical simulation or Bio computing.
After the success of its MPPA DEVELOPER station launched in February 2013 and the embedded board EMB01 launched in December 2013, KALRAY now offers a high performance solution for intensive computing applications through its brand new multi-MPPA acceleration PCIe board. The TURBOCARD2 offers the performance of four MPPA-256 processors to reach up to 1 TFLOPS of computing power for a total power consumption of 50 watts. The TURBOCARD2 is a full length, full height PCIe board that can use either active or passive cooling.
Each of the 4 MPPA-256 processor provides 256 compute cores addressing 4 or 8 GB of DDR3 memory as well as a series of high speed interfaces such as two 8-lane PCIe Gen3, eight 10 Gb Ethernet and unique low latency inter-chip interfaces that scale up to 80 Gbps.
"We are delighted to answer the demand of our customers in the server market by addressing intensive computing and video processing applications with this TURBOCARD2. There is a big expectation from the market on this card. Not only does it bring a huge computing power but it does so for a fraction of the energy needed by all other acceleration solutions available on the market today" said Eric Baissus, CEO of KALRAY.
The TURBOCARD2 will be shipped to key customers in September 2014 and general availability will be in Q4 2014. KALRAY’s MPPA ACCESSCORE SDK provides a nonesuch trace and debug tool to quickly develop and optimize an application, speeding up the development time by a factor of 4, according to the long list of customers who already bought this SDK. KALRAY’s TURBOCARD family of accelerators will allow the deployment of 1 PFLOPS of double precision floating point compute in two industry standard racks which is state of the art in the server domain.
MPPA TURBOCARD2 Features
With the TURBOCARD family, KALRAY offers the most efficient PCIe acceleration solution for MIMD applications.
- 1024 cores on a single board
- Up to 32GB of DDR3x64 1600 MT/s with ECC
- Four KALRAY’s MPPA®-256 processor linked by high speed, low latency links (NoCX®)
- 16-lane PCIe Gen3
- Low latency/High speed board to board NoCX connection
- Trace and debug system
- Active or passive cooling system
- Form factor: Full Height ; Full Length ; Single Slot 111.15 mm (H) x 312 mm (L)
Typical applications include:
- Oil and Gas: e.g. RTM (i.e. Reverse Time Migration) and FWI (i.e. Full Waveform Inversion algorithms),
- Video: e.g. Real-Time 4K HEVC Encoding, Transcoding and Decoding,
- Security: e.g. Cryptography, Cryptanalysis and Homomorphic Encryption,
- Finance: e.g. MapReduce based Monte Carlo or Black-Scholes options pricing and HFT (i.e. High-Frequency Trading) algorithms
- Numerical Simulations: e.g. LBM (i.e. Lattice Boltzmann Methods) for CFD (i.e. Computational Fluid Dynamics),
- Storage: e.g. Seismic Data Compression,
- Biology: Genome DNA Sequencing & MD (i.e. Molecular Dynamics),
- Cloud based Services: e.g. Video Security and Augmented Reality, Big Scientific Instruments.
ABOUT KALRAY
KALRAY is a fabless semiconductor & software company developing and selling a new generation of manycore processors for high performance applications . Our products boost the development of innovative digital applications in the fields of image and signal processing, telecommunication, intensive computing , industrial automation , data management , or transport .
|
Related News
- PLDA and PerfectVIPs, team up to provide the industry's fastest and most reliable IP/VIP Solution for ASIC Based on the PCI Express 3.0 Specification
- TriCN Offers Most Area Efficient x1 PCI Express PHY
- Synopsys Demonstrates Industry's First Interoperability of PCI Express 6.0 IP with Intel's PCIe 6.0 Test Chip
- Verific's Rick Carlson Appointed Advisory Board Member for the College of Computing at Illinois Institute of Technology
- Synopsys Launches Industry's First Complete IP Solution for PCI Express 6.0
Breaking News
- Frontgrade Gaisler Unveils GR716B, a New Standard in Space-Grade Microcontrollers
- Blueshift Memory launches BlueFive processor, accelerating computation by up to 50 times and saving up to 65% energy
- Eliyan Ports Industry's Highest Performing PHY to Samsung Foundry SF4X Process Node, Achieving up to 40 Gbps Bandwidth at Unprecedented Power Levels with UCIe-Compliant Chiplet Interconnect Technology
- CXL Fabless Startup Panmnesia Secures Over $60M in Series A Funding, Aiming to Lead the CXL Switch Silicon Chip and CXL IP
- Cadence Unveils Arm-Based System Chiplet
Most Popular
- Cadence Unveils Arm-Based System Chiplet
- CXL Fabless Startup Panmnesia Secures Over $60M in Series A Funding, Aiming to Lead the CXL Switch Silicon Chip and CXL IP
- Esperanto Technologies and NEC Cooperate on Initiative to Advance Next Generation RISC-V Chips and Software Solutions for HPC
- Eliyan Ports Industry's Highest Performing PHY to Samsung Foundry SF4X Process Node, Achieving up to 40 Gbps Bandwidth at Unprecedented Power Levels with UCIe-Compliant Chiplet Interconnect Technology
- Arteris Selected by GigaDevice for Development in Next-Generation Automotive SoC With Enhanced FuSa Standards
E-mail This Article | Printer-Friendly Page |