Now, the company has announced that the Stratix 10 SoC devices , manufactured on Intel’s 14 nm Tri-Gate process, will incorporate a quad-core 64-bit ARM Cortex-A53 processor system, sitting alongside the device’s floating-point digital signal processing (DSP) blocks and FPGA fabric. Altera envisages applications in areas such as data centre computing acceleration, radar systems and communications infrastructure.
The Cortex-A53 processor is, says Altera, the first 64-bit processor to be used on an SoC FPGA, and when built in the 14 nm Tri-Gate process, “will achieve more than six times more data throughput compared to today’s highest performing SoC FPGAs.” The Cortex-A53 also delivers features such as virtualisation support, 256 TB memory addressing and error correction code (ECC) on L1 and L2 caches. The core can run in 32-bit mode, which will run Cortex-A9 operating systems and code unmodified, allowing an upgrade path from Altera’s 28 nm and 20 nm SoC FPGAs. Stratix 10 SoCs will have a programmable-logic performance level of more than 1GHz; twice the core performance of current high-end 28 nm FPGAs, the company says.
In the tools context, Altera places considerable emphasis on the OpenCL facilities in the tool chain. Embedded developers use Altera’s SoC Embedded Design Suite (EDS) featuring the ARM Development Studio 5 (DS-5) Altera Edition toolkit, “the industry’s only FPGA-adaptive debug tool”, as well as using Altera’s software development kit (SDK) for OpenCL to create heterogeneous implementations using the OpenCL high-level design language.
Heterogeneous computing is very much the name of the game according to Danny Biran, senior vice president, corporate strategy and marketing at Altera, who stresses the capability of combining different levels of performance with different computing elements, as required. “With Stratix 10 SoCs, designers will have a versatile and powerful heterogeneous compute platform enabling them to innovate and get to market faster... converged silicon continues to be the best solution for complex, high-performance applications.” Radar, he notes, is becoming more and