Parallella, Supercomputing, and Ruby (Kenya)

Presented at RubyConf Kenya 2016. Updated to include information on FPGAs, the Porcupine expansion board, and the Raspberry Pi camera.

Transcript

1. Supercomputing, and Ruby RubyConf Kenya Africa Nazarene University - Nairobi, Kenya May 5-7, 2016 RayHightower.com
2. RJ-45 Power μUSB μHDMI μSD
3. RISC ARM + FPGA 18 cores: 2 ARM + 16 RISC
4. RayHightower.com
8. Why?
9. What can we do now that we could not do before?
 How do we show the difference visually?
10. What can we do in parallel that we cannot do serially?
 How do we show the difference visually? Parallel: So what?
11. Moore’s Law: 2x every 18 months
12. Moore’s Law: 2x every 18 months 1993 http://www.washingtonpost.com/blogs/innovations/wp/ 2015/04/14/10-images-that-explain-the-incredible-power-of- moores-law/ 2013
13. Parallelism
14. If one ox could not do the job they did not try to grow a bigger ox, but used two oxen. When we need greater computer power, the answer is not to get a bigger computer, but to build systems of computers and operate them in parallel. -Grace Hopper
15. Concurrency is not Parallelism. https://www.youtube.com/watch?v=cN_DpYBzKso&list=PLOnWKC1gI_OPU8SDIBnCLHsgzNLSbnPJQ&index=3 -Rob Pike, Go
16. Concurrency At least two threads are making progress. Parallelism At least two threads are executing simultaneously. Oracle Multithreaded Programming Guide http://docs.oracle.com/cd/E19455-01/806-5257/6je9h032b/index.html vs.
18. Watts & Dollars
20. 17.8 million watts $17.8 million per year
21. 5 watts for Parallella?
22. http://rayhightower.com/blog/2014/09/09/solar-powered-parallella/
23. http://rayhightower.com/blog/2014/09/09/solar-powered-parallella/ 5 volts 1 amp 5 watts Solar!
24. Reduced Instruction Set Computer RISC
25. Advanced (or Acorn) RISC Machine ARM
27. Find all primes up to 16,000,000. Serial on Parallella.
28. 1 #include <stdio.h> 2 #include <stdlib.h> 3 #include <math.h> 4 #include <time.h> 5 6 #define DEFAULT_MAX_TESTS 16000000 7 8 inline int isprime(unsigned long number) 9 { 10 unsigned long i; 11 unsigned long s = sqrt(number); 12 for(i=3;i<=s;i+=2) 13 { 14 if(number % i == 0) 15 return 0; 16 } 17 return 1; 18 } /* Copyright (c) Adapteva, contributed by M. Thompson with modifications by T. Malthouse. */
29. 8 inline int isprime(unsigned long number) 9 { 10 unsigned long i; 11 unsigned long s = sqrt(number); 12 for(i=3;i<=s;i+=2) 13 { 14 if(number % i == 0) 15 return 0; 16 } 17 return 1; 18 } /* Copyright (c) Adapteva, contributed by M. Thompson with modifications by T. Malthouse. */
31. 237.1 sec
32. Find all primes up to 16,000,000. Serial on Mac OS X.
33. Same serial code, written in C. Build it on OS X.
35. 14.4 sec
36. Find all primes up to 16,000,000. Parallel on Parallella.
37. 27 #include <e-hal.h> 28 29 // Default max number of prime tests per core 30 // Used if a limit it not provided in argv[1] 31 #define DEFAULT_MAX_TESTS 500000 32 33 int main(int argc, char *argv[]) 34 { 35 unsigned row, col, coreid, i, j; 36 e_platform_t platform; 37 e_epiphany_t dev; 38 /* Copyright (c) Adapteva, contributed by M. Thompson with modifications by T. Malthouse.*/
39. 18.6 sec
40. Summary: Finding Primes 0" 50" 100" 150" 200" 250" Parallel Parallella Serial Mac Serial Parallella 18.6 sec 14.4 sec 237.1 sec ($2,000.00 Apple MacBook Pro) ($150.00 Parallella)
41. Embarrassingly parallel problem.
42. Mandelbrot Set
44. Why?
45. https://www.e-education.psu.edu/worldofweather/node/2029 Grid spacing influences accuracy. Weather Prediction
46. break? wear out? work? Finite Element Analysis Will it {
47. Finite Element Analysis http://www.ce.ncsu.edu/news/article/21550/making-bridges-more-robust-to-earthquakes/
48. Finite Element Analysis http://www.ce.ncsu.edu/news/article/21550/making-bridges-more-robust-to-earthquakes/
49. Free Body Diagram F F applied friction F gravity F normal
50. Where can Ruby (or Python, Node.js, Erlang, etc.) fit?
51. Ruby
 +
 Sinatra
 +
 WebSockets
52. Original example by The Hybrid Group and Engine Yard. Modified by WisdomGroup.
53. Field Programmable Gate Array FPGA
55. AB 00 01 10 11 Q 1 1 1 0
56. WebPACK Edition http://www.xilinx.com/products/design-tools/vivado.html
60. •Finance: High-freq trading •Routers •Health care: MRIs, CAT scanners •Cell phone towers •Anything with lots of connections •HDMI interface on Parallella. FPGA Uses…
61. Embedded Systems
64. https://www.parallella.org/2015/06/01/the-open-camera-project-1000-bounty-for-open-firmwaredrivers-for-raspberry-pi-camera-module/
65. Parallella vs GPU
66. 384 cores MacBook Pro Video
67. Specialized vs. General
68. Kickstarter: Jan 2016 Pine 64 4 cores 64-bit $15.00
69. Pine 64 4 cores 64-bit 4K video Starts at $15.00 N ew !
71. Pine64 vs Parallella 4K video 4 cores 2GB RAM 2.5 watts Full-size ports no FPGA $15 - $29 HDMI 18 cores (2 + 16) 1GB RAM 5 watts Micro-ports FPGA $99 - $150
72. Pine64 vs Parallella 4K video 4 cores 2GB RAM 2.5 watts Full-size ports Larger board $15 - $29 HDMI 18 cores (2 + 16) 1GB RAM 5 watts Micro-ports Smaller board $99 - $150 Winner? Depends on design goals.
74. Thanks! RayHightower.com