Zero ASIC
| Industry | Semiconductor industry |
|---|---|
| Founded | March 2008 |
| Founder | Andreas Olofsson |
| Headquarters | , US |
Key people | Andreas Olofsson, CEO |
| Products | Central processing units |
| Owner | Privately funded |
| Website | adapteva |
Adapteva is a fabless semiconductor company focusing on low power many core microprocessor design. The company was the second company to announce a design with 1,000 specialized processing cores on a single integrated circuit.[1][2]
Adapteva was founded in 2008 with the goal of bringing a ten times advancement in floating point performance per watt for the mobile device market. Products are based on its Epiphany multi-core MIMD architecture and its Parallella Kickstarter project promoting "a supercomputer for everyone" in September 2012. The company name is a combination of "adapt" and the Hebrew word "Teva" meaning nature.
History
Adapteva was founded in March 2008, by Andreas Olofsson. The company was founded with the goal of bringing a 10× advancement in floating point processing energy efficiency for the mobile device market. In May 2009, Olofsson had a prototype of a new type of massively parallel multi-core computer architecture. The initial prototype was implemented in 65 nm and had 16 independent microprocessor cores. The initial prototypes enabled Adapteva to secure US$1.5 million in series-A funding from BittWare, a company from Concord, New Hampshire, in October 2009.[3]
Adapteva's first commercial chip product started sampling to customers in early May 2011 and they soon thereafter announced the capability to put up to 4,096 cores on a single chip.
The Epiphany III, was announced in October 2011 using 28 nm and 65 nm manufacturing processes.
Products
Adapteva's main product family is the Epiphany scalable multi-core MIMD architecture. The Epiphany architecture could accommodate chips with up to 4,096 RISC out-of-order microprocessors, all sharing a single 32-bit flat memory space. Each RISC processor in the Epiphany architecture is superscalar with 64× 32-bit unified register file (integer or single precision) microprocessor operating up to 1 GHz and capable of 2 GFLOPS (single precision). Epiphany's RISC processors use a custom instruction set architecture (ISA) optimised for single precision floating point,[4] but are programmable in high level ANSI C using a standard GNU-GCC tool chain. Each RISC processor (in current implementations; not fixed in the architecture) has 32 KB of local memory. Code (possibly duplicated in each core) and stack space should be in that local memory; in addition (most) temporary data should fit there for full speed. Data can also be used from other processor cores local memory at a speed penalty, or off-chip RAM with much larger speed penalty.
The memory architecture is does not employ explicit hierarchy of hardware caches, similar to the Sony/Toshiba/IBM cell processor, but with the additional benefit of off-chip and inter-core loads and stores being supported ( which simplifies porting software to the architecture). It is a hardware implementation of partitioned global address space.[citation needed]
This eliminated the need for complex cache coherency hardware, which places a practical limit on the number of cores in a traditional multicore system. The design allows the programmer to leverage greater foreknowledge of independent data access patterns to avoid the runtime cost of figuring this out. All processor nodes are connected through a Network On Chip, allowing efficient message passing.[5]
Scalability
The architecture is designed to scale almost indefinitely, with 4 e-links allowing multiple chips to be combined in a grid topology, allowing for systems with thousands of cores.
Multi-core coprocessors
.jpg)
On August 19, 2012, Adapteva posted some specifications and information about Epiphany multi-core coprocessors.[6]
| Technical info for | E16G301[7] | E64G401[8] |
|---|---|---|
| Cores | 16 | 64 |
| Core MHz | 1000 | 800 |
| Core GFLOPS/s | 2 | 1.6 |
| "Sum GHz" | 16 | 51.2 |
| Sum GFLOPS | 32 | 102 |
| mm² | 8.96 | 8.2 |
| nm | 65 | 28 |
| W def. | 0.9 | 1.4 |
| W max. | 2 | 2 |
In September 2012, a 16-core version, the Epiphany-III (E16G301), was produced using 65 nm[9] (11.5 mm2, 500 MHz chip[10]) and engineering samples of 64-core Epiphany-IV (E64G401) were produced using 28 nm GlobalFoundries process (800 MHz).[11]
The primary markets for the Epiphany multi-core architecture include:
- Smartphone applications such as real-time facial recognition, speech recognition, translation, and augmented reality.
- Next generation supercomputers requiring drastically better energy efficiency to allow systems to scale to exaflop computing levels.
- Floating point acceleration in embedded systems based on field-programmable gate array architectures.
Parallella project
.jpg)
In September 2012, Adapteva started project Parallella on Kickstarter, which was marketed as "A Supercomputer for everyone." Architecture reference manuals for the platform were published as part of the campaign to attract attention to the project.[12] The US$750,000 funding goal was reached in a month, with a minimum contribution of US$99 entitling backers to obtain one device; although the initial deadline was set for May 2013, the first single-board computers with 16-core Epiphany chip were finally shipped in December 2013.[13]
Size of board is planned to be 86 mm × 53 mm (3.4 in × 2.1 in).[14][15][16]
The Kickstarter campaign raised US$898,921.[17][18] Raising US$3 million goal was unsuccessful, so no 64-core version of Parallella will be mass-produced.[19] Kickstarter users having donated more than US$750 will get "parallella-64" variant with 64-core coprocessor (made from initial prototype manufacturing with 50 chips yield per wafer).[20]
| Parallella-16 Micro Server | Parallella-16 Desktop Computer | Parallella-16 Embedded Platform | |
|---|---|---|---|
| Use Case | Ethernet connected headless server | A personal computer | Leading edge embedded systems |
| Processor | Dual-core 32-bit ARM Cortex-A9 with NEON at 1 GHz (part of Zynq Z7010 chip by Xilinx) | Dual-core 32-bit ARM Cortex-A9 with NEON at 1 GHz (part of Zynq Z7020 chip by Xilinx) | |
| Coprocessor | 16-core Epiphany III multi-core accelerator (E16) | ||
| Memory | 1 GB DDR3L RAM | ||
| Ethernet | 10/100/1000 | ||
| USB | — | 2× USB 2.0 (USB 2.0 HS and USB OTG) | |
| Display | — | HDMI | |
| Storage | 16 GB microSD | ||
| Expansion | — | 2 eLinks + 24 GPIO | 2 eLinks + 24 GPIO |
| FPGA | 28K Programmable Logic Cells 80 Programmable DSP Slices |
80K Programmable Logic Cells 220 Programmable DSP Slices | |
| Weight | 36 g (1.3 oz) | 38 g (1.3 oz) | |
| Size | 3.5 mm × 2.1 mm × 0.625 mm (0.1378 in × 0.0827 in × 0.0246 in) | ||
| SKU | P1600-DK-xx | P1601-DK-xx | P1602-DK-xx |
| HTS Code | 8471.41.0150 | ||
| Power | USB powered (2.5 W) or 5 V DC (~5 W) | ||
Epiphany V
As of 2016, the firm has taped out a 1024 core 64bit variant of their Epiphany architecture.[21] This features: larger local stores (64kb), 64bit addressing, double precision floating point arithmetic or SIMD single precision, and 64bit integer instructions, implemented in the 16nm process node. This design is backwards compatible with earlier products. There are instruction set enhancements aimed at deep-learning and cryptography applications. It is a DARPA funded project.
Performance
 |
Joel Hruska from Extremetech has the following opinion about the Parallella project: "Adapteva is drastically overselling what the Epiphany IV can actually deliver. 16–64 tiny cores with small amounts of memory, no local caches, and a relatively low clock speed can still be useful in certain workloads, but contributors aren't buying a supercomputer—they're buying the real-world equivalent of a self-sealing stem bolt."[22]
The criticism that the Epiphany chips cannot provide anywhere near the performance of modern supercomputers is nevertheless correct: actually, Epiphany chips with 16-cores or 64-cores and c. 25 or 100 GFLOPs in single precision, respectively, do not even match the floating-point performance of modern desktop PC processors (Core i7-4770K (Haswell), 4× cores @ 3.5 GHz AVX2: 177 GFLOPS,[23] double precision) – a fact that is acknowledged by Adapteva.
However, the latest Parallella boards with E16 Epiphany chips[24] can be compared to many ancient supercomputers in terms of raw performance (just as an example, the Cray 1 – the first supercomputer per se – had a peak performance of 80 MFLOPS at 1976, and its successor the Cray 2 had a peak performance of 1.9 GFLOPS at 1985), and can certainly be used for parallel code development; The architectural similarities to supercomputers (message passing and NUMA) make it a potentially useful development system, compared to traditional SMP machines.
The point being that for a power envelope of 5 W and in terms of GFLOPS/mm2 of chip die space, the current E16 Epiphany chips provide vastly more performance than anything else available to date, with an architecture designed to scale, and applicable to more than just embarrassingly parallel GPU tasks.[citation needed] (e.g. it would be capable of running the actor model with many concurrent, fully independent states). It is also suitable for DSP-like tasks where data could be fed directly on chip (from an FPGA or other ASIC) without having to create buffers in temporary memory as for a GPU), making it ideal for robotics & other intelligent sensor applications.
The architecture also allows parallella boards to be combined into a cluster with a fast inter-chip 'eMesh' interconnect, extending the logical grid of cores (creating almost unlimited scaling potential).
The 16-core Parallella has roughly 5.0 GFLOPs/W, and the 64-core Epiphany-IV made with 28 nm estimated as 50 GFLOPs/W (single precision),[25] and 32-board system based on them has 15 GFLOPS/W.[26] For comparison, top GPUs from AMD and Nvidia reached 10 GFLOPs/W for single precision in 2009–2011 timeframe.[27]
See also
- SW26010, a Chinese design featuring a similar architecture, used in their Sunway TaihuLight supercomputer
- Partitioned global address space, a similar memory model found in some HPC applications
- CELL, a multicore processor with scratchpad memory and inter-core DMA
- IBM TrueNorth, a design by IBM with a radically different neuromorphic processor, while having a similar network on chip, extendable with a grid of chips
- Kalray, a company producing a similar network-on-chip RISC processor, aimed at workloads such as convolutional neural networks for vision, again able to extend its grid between multiple chips
- Vision Processing Unit, a class of processor with significant overlapping features
- GRVI Phalanx, an FPGA RISC-V softcore with some similar features
- Asynchronous array of simple processors
- NUMA
References
- ^ Clark, Don (May 3, 2011). "Startup Has Big Plans for Tiny Chip Technology". Wall Street Journal. Retrieved May 3, 2011.
- ^ "IBM says Kilocore technology will outrun today's mobile processors". Tom's Hardware. 2006.
- ^ "From RTL to GDSII in Just Six Weeks". From RTL to GDSII in Just Six Weeks. EETimes. 2010. Retrieved October 26, 2010.
- ^ "Epiphany Architecture Reference Manual".
- ^ "Startup Launches Manycore Floating Point Acceleration Technology". Startup Launches Manycore Floating Point Acceleration Technology. HPCWire. 2011. Retrieved May 3, 2011.
- ^ "Epiphany Multicore IP. Example Configurations". August 19, 2012.
- ^ Epiphany-III 16-core 65nm Microprocessor (E16G301) // admin (August 19, 2012)
- ^ Epiphany-IV 64-core 28nm Microprocessor (E64G401) // admin (August 19, 2012)
- ^ Silicon devices // Adapteva
- ^ Linley Gwennap, Adapteva: More Flops, Less Watts. Epiphany Offers Floating-Point Accelerator for Mobile Processors. // Microprocessor Report, June 2011
- ^ Michael Feldman, Adapteva Unveils 64-Core Chip // HPCWire
- ^ Andreas Olofsson, Epiphany Documentation Release
- ^ Update #46: First Parallella User Created Video
- ^ Rick Merritt, Adapteva Kickstarts Hundred-Dollar Supercomputer // EETimes, September 27, 2012
- ^ "Parallella - Supercomputing for Everyone(slidecast)" // Adapteva Founder & CEO Andreas Olofsson (September 28, 2012)
- ^ Parallella: A Supercomputer For Everyone by Adapteva, Project page at Kickstarter
- ^ Parallella: A Supercomputer For Everyone // Kickstarter project, by Adapteva
- ^ Hiawatha Bray, Adapteva creates efficient, cheap microchip with help from Kickstarter. ‘Crowdfunding’ puts a tiny, fast computer closer to production // The Boston Globe, December 2, 2012
- ^ Andrew Back, Introducing the $99 Linux Supercomputer, Linux.com, January 24, 2013: "pledges of $99 or more being rewarded with at least one board with a 16-core device. ... The 16-core Epiphany chip delivers 26 GFLOPS of performance and with the entire Parallella computer consuming only 5 watts"
- ^ 64-core version of the Parallella board now offered! // Adapteva blog at Kickstarter, October 25, 2012: "The Epiphany-IV (64+2) core Parallella board will be offered for pledges above $750. ... the fact that we only get 50 dies per wafer for these initial prototype runs. We can't disclose wafer pricing and yields at 28nm,"
- ^ "epiphany v announcement".
- ^ Joel Hruska (September 28, 2012). "Adapteva turns to Kickstarter to fund massively parallel processor". Extremetech.
- ^ Dr. Donald Kinghorn (August 26, 2013). "Haswell Floating Point Performance". Puget Systems Blog.
- ^ Andreas Olofsson (July 14, 2014). "New Parallella Product Offerings". Parallella Blog. Retrieved September 3, 2014.
- ^ Feldman, Michael (August 22, 2012). "Adapteva Unveils 64-Core Chip". HPCWire. Retrieved September 3, 2014.
- ^ "Adapteva Reveals A-1 Supercomputing Platform at ISC14". HPCWire, press-release of Adapteva. June 23, 2014. Retrieved September 3, 2014.
- ^ "CPU, GPU and MIC Hardware Characteristics over Time. Raw Compute Performance - Comparison of GFLOP/sec per Watt for single precision arithmetics. Higher is better". Karl Rupp. June 24, 2013. Retrieved September 3, 2014.
Further reading
- Linley Gwennap, Adapteva: More Flops, Less Watts. Epiphany Offers Floating-Point Accelerator for Mobile Processors. // Microprocessor Report, June 2011