Successfully reported this slideshow.
We use your LinkedIn profile and activity data to personalize ads and to show you more relevant ads. You can change your ad preferences anytime.

numPYNQ @ NGCLE@e-Novia 15.11.2017

90 views

Published on

numPYNQ @ NGCLE@e-Novia 15.11.2017

Published in: Engineering
  • Be the first to comment

  • Be the first to like this

numPYNQ @ NGCLE@e-Novia 15.11.2017

  1. 1. numPYNQ Luca Stornaiuolo luca.stornaiuolo@polimi.it Riccardo Pressiani riccardo.pressiani@mail.polimi.it Filippo Carloni filippo.carloni@mail.polimi.it 15/11/2017 e-Novia Sara Notargiacomo, Giuseppe Natale, Luca Cerina, Marco D. Santambrogio
  2. 2. numPYNQ Context Definition 2 Huge amount of data that need to be processed to get aggregated information Signal Processing Machine LearningData MiningImage Analysis
  3. 3. numPYNQ Context Definition 3 GPUs are the most popular choice to accelerate data analysis Field Programmable Gate Array (FPGA) are gaining momentum
  4. 4. numPYNQ Context Definition 4 GPUs are the most popular choice to accelerate data analysis Field Programmable Gate Array (FPGA) are gaining momentum Performance Energy Efficiency Flexibility Easy to Use GPU FPGA
  5. 5. numPYNQ 5 numPYNQ numPYNQ is an FPGA-based hardware library that offers data scientists an accelerated version of the core functions of NumPy to be used transparently from their applications. Performance Energy Efficiency Flexibility Easy to Use GPU FPGA
  6. 6. numPYNQ 6 Overlay Technology Crosscorrelation Matrix Dot Product
  7. 7. numPYNQ 7 Experimental Results
  8. 8. numPYNQ 8 Experimental Results
  9. 9. numPYNQ 9 Runtime Adaptivity Runtime Input Analysis Target Implementation
  10. 10. numPYNQ 10 Market The Big Data market is estimated to top $61 billion in 2020, a 26% compound annual growth rate for 2011-2020. Source: Wikibon research as covered by KDNuggets.com
  11. 11. numPYNQ 11 Competition Analisys numPYNQ TensorFlow PyCUDA
  12. 12. numPYNQ 12 Go to Market Data Analysis-based Companies Partnership deal to complete the development with the company support Revenue sharing Direct channel marketing through sales force, email and online marketing numPYNQ as a Service
  13. 13. numPYNQ 13 Freemium Model Basic Plan Premium Plan Advanced Plan Basic functions Advanced Functions — Dedicated Training — Dedicated Support — Custom Functions — — Cost Free $ $$$
  14. 14. numPYNQ 14 Team Luca Stornaiuolo PhD student in Computer Science and Engineering at Politecnico di Milano Role in numPYNQ project: Hardware Design and Optimization Manager Riccardo Pressiani MSc student in in Computer Science and Engineering at Politecnico di Milano and University of Illinois at Chicago Role in numPYNQ project: Hardware Implementation and Social Media Manager Filippo Carloni BSc student in Computer Science and Engineering at Politecnico di Milano Role in numPYNQ project: PYNQ Overlay and Interfaces Manager Sara Notargiacomo, Giuseppe Natale, Luca Cerina, Marco D. Santambrogio
  15. 15. numPYNQ https://twitter.com/numPYNQ_NECST https://www.facebook.com/numPYNQ https://www.slideshare.net/numPYNQ_at_NECST Thanks for the attention!

×