Changes between Version 19 and Version 20 of cypress/XeonPhi
- Timestamp:
- 10/21/15 11:16:51 (9 years ago)
Legend:
- Unmodified
- Added
- Removed
- Modified
-
cypress/XeonPhi
v19 v20 1 1 = Programming for the Xeon Phi Coprocessor on Cypress = 2 3 ==== Workshop Reminder ====4 To take advantage of the workshop QOS:5 {{{#!bash6 export MY_PARTITION=workshop7 export MY_QUEUE=workshop8 idev -c 4 --gres=mic:09 }}}10 11 To follow along, you can copy the source code and submission scripts12 {{{13 cp -r /lustre/project/workshop/PhiExamples/ .14 }}}15 2 16 3 The Xeon Phi coprocessor is an accelerator used to provide many cores to parallel applications. While it has fewer threads available than a typical GPU accelerator, the processors are much "smarter" and the programming paradigm is similar to what you experience coding for CPUs. … … 154 141 * Get onto a compute node using idev 155 142 {{{ 156 [tuhpc002@cypress1 Day2]$ export MY_PARTITION=workshop157 [tuhpc002@cypress1 Day2]$ export MY_QUEUE=workshop158 143 [tuhpc002@cypress1 Day2]$ idev -c 4 --gres=mic:0 159 Requesting 1 node(s) task(s) to workshop queue of workshoppartition144 Requesting 1 node(s) task(s) to normal queue of default partition 160 145 1 task(s)/node, 4 cpu(s)/task, 2 MIC device(s)/node 161 146 Time: 0 (hr) 60 (min). … … 271 256 Let's start out in interactive mode and then we will look at job submission: 272 257 {{{#!bash 273 export MY_PARTITION=workshop274 export MY_QUEUE=workshop275 258 idev -c 4 --gres=mic:0 276 259 }}} … … 464 447 {{{#!bash 465 448 #!/bin/bash 466 #SBATCH --qos=workshop # Quality of Service 467 #SBATCH --partition=workshop #Partition 449 #SBATCH --qos=normal # Quality of Service 468 450 #SBATCH --job-name=nativeTest # Job Name 469 451 #SBATCH --time=00:10:00 # WallTime … … 625 607 {{{#!bash 626 608 #!/bin/bash 627 #SBATCH --qos=workshop # Quality of Service 628 #SBATCH --partition=workshop #Partition 609 #SBATCH --qos=normal # Quality of Service 629 610 #SBATCH --job-name=offloadTest # Job Name 630 611 #SBATCH --time=00:10:00 # WallTime … … 654 635 * Minimal data transfer 655 636 656 657 == Future Training ==658 We've only scratched the surface on the potential of the Xeon Phi coprocessor. If you are interested in learning more, Colfax International will be giving two days of instruction on coding for the Xeon Phi at Tulane at the end of September. Interested parties can register at659 660 CDT 101: http://events.r20.constantcontact.com/register/event?oeidk=a07eayq4gvha16a1237&llr=kpiwi7pab661 662 CDT 102: http://events.r20.constantcontact.com/register/event?oeidk=a07eayqb5mwf5397895&llr=kpiwi7pab663 664 [[Image(ColfaxInvite.png, center, 80%)]]