Intel® Stratix® 10 SoC: Highest Performance and Most Power-Efficient Processing
Intel® Stratix® 10 SoCs, manufactured on Intel’s 14 nm process technology, combine a quad-core ARM* Cortex*-A53 MPCore* hard processor system with the revolutionary Intel® HyperFlex™ FPGA Architecture to deliver breakthrough advantages in performance, power efficiency, density, and system integration. Intel Stratix 10 devices deliver 2X core performance gains over previous-generation, high-performance FPGAs with up to 70% lower power.†
The Intel® Stratix® 10 SoCs HPS architecture now includes a System Memory Management Unit, which enables hardware virtualization across the processor and FPGA domains. Intel® Stratix® 10 SoCs add a Cache Coherency Unit to provide one-way (I/O) cache coherency with the Cortex*-A53 MPCore* processor. Intel® Stratix® 10 SoCs also include up to 10 TFLOPS of hardened floating-point digital signal processing (DSP) blocks, embedded high-speed transceivers, hard memory controllers, and protocol intellectual property (IP) controllers - all in a single, highly integrated package.
By integrating the FPGA and the ARM* processor, Intel® Stratix® 10 SoCs provide an ideal solution for 5G wireless communication, software defined radios, secure computing for military applications, network function virtualization (NFV), and data center acceleration.
Intel® Stratix® 10 Device Demo Videos
28G Transceivers
In this video, we look at the unique transceiver architecture of Intel® Stratix® 10 FPGAs. See H-Tile transceivers that are connected via Intel's EMIB technology and operating at 28 Gbps backplane performance.
Intel® HyperFlex™ FPGA Architecture
Intel® HyperFlex™ FPGA Architecture in Intel® Stratix® 10 devices provides 2X the fMAX performance†. This video shows a side-by-side comparison of an original design and a Hyper-Optimized design.
PCIe* Gen3 DMA to DDR4 SDRAM
Intel® Stratix® 10 devices, which include PCI Express* (PCIe*) and memory controller hard intellectual property (IP) blocks, combined with Avalon® Memory Mapped and direct memory access (DMA) functions to create a high-performance reference design.
Achieve Performance Breakthroughs
- Quad-core 64-bit ARM* Cortex*-A53 HPS operating at up to 1.5 GHz
- ARM's next-generation 64-bit architecture (ARMv8)
- Ground-breaking Intel® HyperFlex™ FPGA Architecture delivering 2X the core performance gains†
- Up to 10 TFLOPS of single-precision floating-point DSP performance
Break Through the Bandwidth Barrier
- Up to 144 transceivers with data rates up to 30 Gbps deliver 4X serial transceiver bandwidth from previous generation FPGAs for high port count designs†
- 30 Gbps backplane capability for versatile data switching applications
- A path to 56 Gbps chip-to-chip/module capability for leading-edge interface standards
- Over 2.5 Tbps bandwidth for serial memory with support for Hybrid Memory Cube
- Over 2.3 Tbps bandwidth for parallel memory interfaces with support for DDR4 SDRAM at 2,666 Mbps
Lower Operating Expense
- Quad-core ARM C*ortex*-A53 processor optimized for performance per watt
- Leveraging Intel's leadership in process technology, Intel® Stratix® 10 devices offer the most power-efficient technologies
- Up to 70% lower power than prior-generation high-end SoCs†
- Up to 80 GFLOPS/Watt of single-precision floating point power efficiency†
Achieve High Levels of System Integration
- 64-bit quad-core ARM Cortex-A53 processor to enable hardware virtualization, system management and monitoring capabilities, acceleration pre-processing, and more
- Highest density FPGA fabric with 5.5M logic elements in a monolithic implementation
- Heterogeneous 3D SiP solutions including transceivers and other advanced components
Obtain comprehensive high-performance FPGA security capabilities
- Integrated Secure Device Manager (SDM) for flexibility to update configuration code
- Multi-factor authentication
- Physically Unclonable Function (PUF)
Get Faster Time to Market
- Start designing with Intel® Arria® 10 SoC and migrate to Intel® Stratix® 10 SoC
- Code compatability with previous generation SoCs
- Cortex-A53 processor supports 32-bit execution mode
- Complementary Intel® Enpirion® PowerSoCs offer complete and validated power solution for Intel® Stratix® 10 SoCs with higher performance, lower system power, higher reliability, smaller footprint, and faster time to market
Achieve high designer productivity with optimized FPGA and SoC design software
- Heterogeneous debug, profiling, and whole chip visualization with Intel® FPGA SoC FPGA Embedded Development Suite (EDS) featuring the ARM* Development Studio 5* (DS-5*) Intel® SoC FPGA Edition Toolkit
- New engine optimized for multi-million logic elements (LE) FPGA designs providing
- Up to 8X faster compile times†
- Significant reduction in design iterations
- Hyper-Aware design flow to optimize designs for the Intel® HyperFlex™ FPGA Architecture
- C-based design entry using the Intel® FPGA SDK for OpenCL™, offering a design environment that is easy to implement on SoC FPGAs
- Heterogeneous C-based modeling and hardware design with the Intel® FPGA SDK for OpenCL™
Intel® Stratix® 10 SoCs: Designed for Productivity
Design productivity is one of the driving philosophies of the Intel® Stratix® 10 SoC architecture. Intel® Stratix® 10 SoCs offer full software compatibility with previous generation SoCs, a broad ecosystem of ARM* software and tools, and the enhanced FPGA and digital signal processing (DSP) hardware design flow.
- Extensive ecosystem of ARM for software development
- Intel® SoC FPGA Embedded Development Suite (EDS) featuring the ARM* Development Studio 5* (DS-5*) Intel® SoC FPGA Edition Toolkit
- Board support packages for popular operating system including Linux*, Wind River’s VxWorks, Micrium’s uC/OS-II and uC/OS-III, and more
- Software compatibility between 28 nm Cyclone® V and Arria® V SoCs and 20 nm Intel® Arria® 10 SoCs
- Intel® Quartus® Prime software featuring:
- High-level automated design flow with Open Computing Language (OpenCL™) compiler
- Model-based DSP hardware design with the DSP Builder for Intel® FPGAs
- Intel® Stratix® 10 SoC Virtual Platform to enable early software development and verification
Documentation
- A New FPGA Architecture and Leading-Edge FinFET Process Technology Promise to Meet Next-Generation System Requirements White Paper (PDF)
- Leveraging the Intel® HyperFlex™ FPGA Architecture in Intel Stratix® 10 Devices to Achieve Maximum Power Reduction White Paper (PDF)
- Enabling Next-Generation Platforms Using Intel’s 3D System-in-Package Technology White Paper (PDF)
- Secure Device Manager for Intel® Stratix® 10 Devices Provides FPGA and SoC Security White Paper (PDF)
- Understanding How the New Intel® HyperFlex™ FPGA Architecture Enables NextGeneration High-Performance Systems White Paper (PDF)
- Using Intel® Quartus® Prime Software to Maximize Performance in the Intel HyperFlex™ FPGA Architecture White Paper (PDF)
- The Breakthrough Advantage for FPGAs with Tri-Gate Technology White Paper (PDF)
OpenCL and the OpenCL logo are trademarks of Apple Inc. used by permission by Khronos.
†Comparison based on Stratix V vs. Stratix 10 using Quartus Prime Pro 16.1 Early Beta. Stratix V Designs were optimized using 3 step optimization process of Hyper-Retiming, Hyper-Pipelining, and Hyper-Optimization in order to utilize Stratix 10 architecture enhancements of distributed registers in core fabric. Designs were analyzed using Quartus Prime Pro Fast Forward Compile performance exploration tool. For more details, refer to HyperFlex FPGA Architecture Overview White Paper: https://www.altera.com/content/dam/altera-www/global/en_US/pdfs/literature/wp/wp-01220-hyperflex-architecture-fpga-socs.pdf. Actual performance users will achieve varies based on level of design optimization applied. Tests measure performance of components on a particular test, in specific systems. Differences in hardware, software, or configuration will affect actual performance. Consult other sources of information to evaluate performance as you consider your purchase. For more complete information about performance and benchmark results, visit www.intel.com/benchmarks.