What is Adapteva Possibly the World146s Smallest Semiconductor Company 64core 28nm 100 GFLOPS Coprocessor 2W Now Also a System Company133 1 in Processor Energy Efficiency at 50 GFLOPSWatt T ID: 112320
Download Pdf The PPT/PDF document "Inventing the Future of ComputingKicksta..." is the property of its rightful owner. Permission is granted to download and print the materials on this web site for personal, non-commercial use only, and to display it on your personal computer provided you do not modify the materials and that you retain all copyright notices contained in the materials. By downloading content from our website, you accept the terms of this agreement.
Inventing the Future of ComputingKickstarting the Transition to Parallel Computing With Open HardwareAndreas Olofssonandreas@adapteva.com Linux Collaboration Summit 2013April 15th17th, 2013 San Francisco, What is Adapteva? Possibly the World’s Smallest Semiconductor Company 64core 28nm 100 GFLOPS Coprocessor @2W Now Also a System Company… #1 in Processor Energy Efficiency at 50 GFLOPS/Watt The Computing Energy Crisis: IT’S REAL!! 1,00010,000100,0001,000,00010,000,000100,000,000 System ProcessingNeeds Legacy ProcessingEfficiency “The Efficiency Gap”Von NeumannSaturation The Ideal World InfRAM CPUInfInfB/sec The Real World: Performance is Saturating! . Add Cores. SIMD1. ScaleFrequency Now What? . MultiIssue Power Consumption Thermal Density Memory Bottlenecks Latency Wall Wiring Frequency Wall Yield Issues Time to Market Software Complexity Amdahl’s Law10 Trends that Will Shape the Future of Computing Nature is Massively Parallel Efficient Robust Heterogeneous Parallel The Heterogeneous Computing Vision SYSTEMCHIP BIGCPU FPGA BIGCPU BIGCPU BIGCPU 100’s of small RISC CPUs GPU Analog The Current State of Parallel Programming How To Make Every Programmer a Parallel Programmer? Source: Github ��10 &#x/MCI; 7 ;&#x/MCI; 7 ;Industry Challenges Before Us Rebuild the computer ecosystemRewrite billions of lines of codeeducate millions of programmersRewrite the education curriculum 11 No Computing Parallel Computing “Von Neumann Age”Serial ComputingNo Electronic Computing Parallel ComputingAny Reason to Think the Future of Computing is NOT Parallel? What is Parallella?��12 &#x/MCI; 10;&#x 000;&#x/MCI; 10;&#x 000;“A $99 credit card sized parallel computing platform” Parallella Principles PARALLEL:Heterogeneousscalableparallel hardware OPEN:Opensource O/S (LinuxOpenchip documentationOpensource drivers and SDKOpensource hardware (board schematics, docs, layout)Openstandards (OpenCL, OpenMP, MPI, ...)ACESSIBLE:$99 starting point Easy to use novice progammers Parallella Kickstarter Stats ”$99 Linux supercomputer”5,000 backers6,300 boards ”presold” in 4 weeks67 countries, all 50 US states75% of backers are developers,000 more signups since Jan 1stCustomer Application Classes:SDR (GNU Radio)Ray tracing/renderingImage processingRoboticsGaming CryptographyMedia ServerDistributed ComputingSignal processing The Parallella Architecture Dual CoreARM A9 AXI BUS MIO SHARED DRAM “O/S” DRAM USB OTG USB 2.0 UART Ethernet CARD I2C DAC/ADC IF HDMIController MASTER SLAVE “GlueLogic” DaughterCard MASTER ZynqFPGA Zynq“Hard” OffChip MEMCTRL “Sandbox” EPIPHANY For the first time in public… PARALLELLAPARALLELLAZynq Dual Core ARM A9core Epiphany Accelerator1GB RAMGbE, USB, HDMI, uSD 6 GB/s expansion connectors$99 (long term goal)Same as PARALLELLA16, withcore Epiphany Accelerator First Parallella16 PowerUp was at 1:30pm on April 11 Parallella Bringup Day 1 Parallella Bringup Day 2 Parallella Bringup Day 3 Epiphany Coprocessor Read/Write Success Parallella What’s Next? Ship 6,300 Boards ASAP Start the Parallella Academic Program Start working on Parallella1024 Build a sustainable supply model