Tarek's Hotpar 2010 article online

Article: Parallelism at HotPar 2010
By: David Kanter (dkanter.delete@this.realworldtech.com), July 27, 2010 8:29 am
Room: Moderated Discussions
Today is shaping up to be an excellent day for a number of reasons:

We have an excellent new contributor, Tarek Chammah. Tarek is a graduate student at the University of Waterloo who specializes in software approaches to parallelization, including run-times, languages, APIs, etc.

I recently had the opportunity to go to lunch with Tarek and we had an excellent time and I learned quite a lot about the trends on the software side of the equation. One of the points that Tarek emphatically made is that with the emergence of parallel processing, the software is becoming equally important as the hardware. Times are a changing, and it's not just about your good old compiler getting code into shape for the hardware; software is truly an essential part of the glue that binds together the system, and I hope to be able to discuss software more in the future at RWT.

Second, Tarek has provided us with an excellent article covering some of the highlights of the HotPar 2010 workshop. Hot Par was held in Berkeley this year, and included a fair number of papers - but almost all of them were software focused. This is a nice change of pace from our usual coverage of ISSCC, Hot Chips or IEDM:

http://www.realworldtech.com/page.cfm?ArticleID=RWT072610001641

Please join me in thanking Tarek for his contribution, and I look forward to some lively discussions.


David
 Next Post in Thread >
TopicPosted ByDate
Tarek's Hotpar 2010 article onlineDavid Kanter07/27/10 08:29 AM
  Tarek's Hotpar 2010 article onlineAnon07/27/10 02:57 PM
    Tarek's Hotpar 2010 article onlineDavid Kanter07/27/10 11:48 PM
      Tarek's Hotpar 2010 article onlineAnon07/28/10 03:44 PM
        Tarek's Hotpar 2010 article onlineanon07/28/10 05:10 PM
        CPU+GPUDavid Kanter07/29/10 12:38 PM
          CPU+GPUMark Christiansen07/30/10 07:36 AM
            CPU+GPUhobold07/30/10 07:54 AM
              CPU+GPUGabriele Svelto07/30/10 08:18 AM
                CPU+GPUhobold07/30/10 03:46 PM
              CPU+GPUAaron Spink08/02/10 06:32 PM
                CPU+GPUhobold08/03/10 04:33 AM
                  CPU+GPURicardo B08/03/10 05:19 AM
            CPU+GPU - the software problemVincent Diepeveen07/31/10 05:50 PM
              CPU+GPU - the software problemmpx08/02/10 12:19 PM
                CPU+GPU - the software problemRichard Cownie08/02/10 01:48 PM
                  CPU+GPU - the software problemGabriele Svelto08/03/10 12:10 AM
                    CPU+GPU - the software problemhobold08/03/10 04:41 AM
                      CPU+GPU - the software problemRichard Cownie08/03/10 06:17 AM
                        CPU+GPU - the software problemhobold08/04/10 07:45 AM
                          NVidia in a holeRichard Cownie08/04/10 08:23 AM
                            NVidia in a holeRohit08/04/10 08:33 AM
                              NVidia in a holeRichard Cownie08/04/10 08:50 AM
                                NVidia in a holeRohit08/05/10 07:31 AM
                                  NVidia in a holeGroo08/05/10 02:08 PM
                                NVidia in a holeGroo08/05/10 02:07 PM
                              NVidia in a holeMark Roulo08/04/10 09:26 AM
                                NVidia in a holeLaughabee08/18/10 08:28 AM
                            Can nVidia survive without a sub-$100 GPU market?Mark Roulo08/04/10 09:29 AM
                              Can nVidia survive without a sub-$100 GPU market?Richard Cownie08/04/10 09:42 AM
                                Can nVidia survive without a sub-$100 GPU market?Mark Roulo08/04/10 09:55 AM
                                  Can nVidia survive without a sub-$100 GPU market?Linus Torvalds08/04/10 11:35 AM
                                    "profit" is subjectiveRichard Cownie08/04/10 11:49 AM
                                      "profit" is subjectivea reader08/06/10 04:54 AM
                                        no price warRichard Cownie08/06/10 07:18 AM
                                  Can nVidia survive without a sub-$100 GPU market?Aaron Spink08/04/10 12:19 PM
                                  Can nVidia survive without a sub-$100 GPU market?Konrad Schwarz08/06/10 08:13 AM
                                CPU lineup != GPU lineupRohit08/05/10 07:43 AM
                                  CPU lineup != GPU lineupRichard Cownie08/05/10 08:12 AM
                                    CPU lineup != GPU lineupRohit08/05/10 08:37 AM
                                      CPU lineup != GPU lineupRichard Cownie08/05/10 08:56 AM
                                        CPU lineup != GPU lineupRohit08/05/10 11:19 AM
                                          CPU lineup != GPU lineuphobold08/06/10 04:08 AM
                                            CPU lineup != GPU lineupRohit08/06/10 07:24 AM
                                              CPU lineup != GPU lineuphobold08/06/10 07:41 AM
                              Can nVidia survive without a sub-$100 GPU market?David Hess08/04/10 11:31 AM
                              Can nVidia survive without a sub-$100 GPU market?Ricardo B08/04/10 01:16 PM
                              Semiconductor economicsDavid Kanter08/05/10 09:13 AM
                                Semiconductor economicsRichard Cownie08/05/10 09:51 AM
                                  NV buying ViA could shake up thingsRohit08/05/10 11:09 AM
                                    NV buying ViA could shake up thingsRichard Cownie08/05/10 11:25 AM
                                      Not likelyDavid Kanter08/05/10 02:39 PM
                                        Not likelyrandom08/05/10 02:49 PM
                                          Not likelyAaron Spink08/05/10 03:18 PM
                                        Not likelyGroo08/05/10 10:05 PM
                                        Not likelyRohit08/06/10 07:25 AM
                                    NV buying ViA could shake up thingsGroo08/05/10 10:02 PM
                                    NV buying ViA could shake up thingsehud08/06/10 01:53 AM
                                      NV buying ViA could shake up thingsRohit08/06/10 07:30 AM
                                        NV buying ViA could shake up thingsKevin G08/06/10 05:54 PM
                                          ViA buying NV could shake up thingsBrendan08/19/10 06:23 AM
                                        NV buying ViA could shake up thingsGroo08/06/10 08:24 PM
                            NVidia in a holempx08/04/10 12:01 PM
                              NVidia in a holeRichard Cownie08/04/10 12:31 PM
                                NVidia in a holercf08/04/10 03:40 PM
                              NVidia in a holeRohit08/05/10 07:51 AM
                              NVidia in a holeGroo08/05/10 09:50 PM
                              NVidia in a holeAaron Spink08/05/10 11:52 PM
                          Anyone else wish Intel would buy Nvidia ? (NT)Silent08/07/10 01:30 AM
                            Hell noa08/07/10 08:37 PM
                        CPU+GPU - the software problemno thanks08/04/10 07:54 PM
                          CPU+GPU - the software problemJouni Osmala08/05/10 01:41 AM
                            CPU+GPU - the software problemIntelUser200008/05/10 07:54 PM
                              CPU+GPU - the software problemJouni Osmala08/05/10 09:36 PM
                            CPU+GPU - the software problemLinus Torvalds08/06/10 08:08 AM
                              CPU+GPU - the software problemanonymous08/06/10 09:39 AM
                                CPU+GPU - the software problemLinus Torvalds08/06/10 10:56 AM
                                  non-existent Via+NV vs real OntarioRichard Cownie08/06/10 12:12 PM
                              CPU+GPU - the software problemkoby m.08/07/10 02:06 AM
                  CPU+GPU - the software problemhobold08/03/10 04:48 AM
                CPU+GPU - the software problemhobold08/03/10 04:45 AM
                  CPU+GPU - the software problemRichard Cownie08/03/10 06:22 AM
          CPU+GPUAnon07/30/10 09:11 PM
            CPU+GPUanon07/31/10 03:19 AM
            CPU+GPUAaron Spink08/02/10 06:45 PM
          100x speedups -- here we go againAM08/02/10 01:37 AM
            100x speedups -- here we go againanon08/02/10 03:32 AM
              100x speedups -- here we go againanon08/02/10 11:14 PM
            GPU raw H/W only has about a 10x advantage over a CPUMark Roulo08/02/10 09:41 AM
              GPU raw H/W only has about a 10x advantage over a CPUMichael S08/02/10 03:31 PM
                GPU raw H/W only has about a 10x advantage over a CPUanon08/02/10 09:36 PM
                  GPU raw H/W only has about a 10x advantage over a CPUMichael S08/03/10 12:41 AM
                    GPU raw H/W only has about a 10x advantage over a CPUMark Roulo08/03/10 08:22 AM
                      GPU raw H/W only has about a 10x advantage over a CPURohit08/04/10 07:08 AM
                      There is reasonable explanation...Jouni Osmala08/04/10 10:10 PM
                  GPU raw H/W only has about a 10x advantage over a CPUMark Roulo08/03/10 08:19 AM
              simple reality check for youAM08/03/10 01:45 AM
                simple reality check for younone08/03/10 07:27 AM
                  simple reality check for youAM08/03/10 11:50 PM
                    simple reality check for youanon08/04/10 12:06 AM
                      simple reality check for youAM08/05/10 03:51 AM
                        catch the same boat as everyone else next time (NT)anon08/05/10 04:44 AM
                Simple reality checkDavid Kanter08/03/10 08:24 AM
                  Simple reality checkGabriele Svelto08/03/10 10:51 PM
                  Simple reality checkAM08/03/10 11:57 PM
                    Simple reality checkanon08/04/10 12:12 AM
                      Simple reality checkAM08/05/10 03:59 AM
                        Simple reality checkanon08/05/10 04:47 AM
                          Study the papers, troll (NT)AM08/06/10 12:01 AM
                    Simple reality checkhobold08/04/10 05:05 AM
                    Simple reality checkDean Kent08/04/10 09:30 AM
                      Simple reality checkMark Roulo08/04/10 09:59 AM
                      Suggestion for David Kanter and Mark RouloAM08/05/10 03:57 AM
                        Suggestion for David Kanter and Mark Roulonone08/05/10 04:22 AM
                          Suggestion for David Kanter and Mark RouloAM08/06/10 12:04 AM
                        Suggestion for David Kanter and Mark Roulonone08/05/10 04:38 AM
                          Suggestion for David Kanter and Mark Rouloanon08/05/10 04:43 AM
                            Suggestion for David Kanter and Mark RouloAM08/06/10 12:11 AM
                              Suggestion for David Kanter and Mark Rouloanon08/06/10 07:52 AM
                          Suggestion for David Kanter and Mark RouloAM08/06/10 12:10 AM
                            Suggestion for David Kanter and Mark Roulonone08/06/10 01:35 AM
                              price doesn't matter only in theories (NT)AM08/09/10 02:13 AM
                            Not apple4appleMichael S08/06/10 05:13 AM
                              Not apple4appleAM08/09/10 02:09 AM
                                Not apple4appleMichael S08/09/10 04:35 AM
                                  Not apple4appleAM08/10/10 01:05 AM
                                    Not apple4applenone08/10/10 01:38 AM
                                      Back to topicAM08/11/10 12:03 AM
                                      Final note (and some remarks on Intel's paper)AM08/13/10 04:28 AM
                                        Final note (and some remarks on Intel's paper)none08/13/10 09:10 AM
                                          Final note (and some remarks on Intel's paper)AM08/16/10 01:14 AM
                                            Final note (and some remarks on Intel's paper)none08/16/10 06:29 AM
                                              One example of 1000X shown to be wrong.sea08/16/10 07:55 PM
                                                You're short of some factsAM08/17/10 01:13 AM
                                                  Maybe you need to read beyond the PR statementSteve Underwood08/17/10 02:39 AM
                                                    Maybe you need to read beyond the PR statementGroo08/17/10 08:18 PM
                                                    check the hw they usedAM08/18/10 03:33 AM
                                                Please realize what Monte Carlo meansVincent Diepeveen08/18/10 03:28 AM
                                              Final note (and some remarks on Intel's paper)AM08/17/10 01:18 AM
                                        Final note (and some remarks on Intel's paper)anon08/14/10 01:22 AM
                                          Final note (and some remarks on Intel's paper)AM08/16/10 01:15 AM
                                            Final note (and some remarks on Intel's paper)anon08/16/10 02:10 PM
                                              Final note (and some remarks on Intel's paper)AM08/17/10 01:15 AM
                                                Final note (and some remarks on Intel's paper)Michael S08/17/10 02:58 AM
                                                  Final note (and some remarks on Intel's paper)AM08/18/10 03:17 AM
                                                    Final note (and some remarks on Intel's paper)gallier208/18/10 04:52 AM
                                                    Final note (and some remarks on Intel's paper)Michael S08/18/10 05:33 AM
                                                      Final note (and some remarks on Intel's paper)Gabriele Svelto08/18/10 06:11 AM
                                                      Final note (and some remarks on Intel's paper)Steve Underwood08/18/10 04:03 PM
                                                        Intel might be moving from ISA to platformhobold08/19/10 03:58 AM
                                                          Intel might be moving from ISA to platformSteve Underwood08/22/10 07:00 PM
                                                            Intel might be moving from ISA to platformAnon08/22/10 10:43 PM
                                                              Intel might be moving from ISA to platformajensen08/23/10 12:37 AM
                                                                Intel might be moving from ISA to platformMichael S08/23/10 02:13 AM
                                                                  Intel might be moving from ISA to platformSteve Underwood08/23/10 02:35 AM
                                                                  Intel might be moving from ISA to platformhobold08/26/10 04:37 AM
                                                                Intel might be moving from ISA to platformAnon08/23/10 03:47 PM
                                                              Intel might be moving from ISA to platformSteve Underwood08/23/10 02:25 AM
                                                                Intel might be moving from ISA to platformhobold08/23/10 03:03 AM
                                                                  Intel might be moving from ISA to platformSteve Underwood08/23/10 04:26 AM
                                                                Intel might be moving from ISA to platformAnon08/23/10 03:55 PM
                                                                Intel might be moving from ISA to platformrwessel08/23/10 06:41 PM
                                                                  Intel might be moving from ISA to platformSteve Underwood08/23/10 07:30 PM
                                                                    Intel might be moving from ISA to platformrwessel08/23/10 09:50 PM
                                                                      Intel might be moving from ISA to platformSteve Underwood08/23/10 10:34 PM
                                                                        Intel might be moving from ISA to platformrwessel08/24/10 01:03 AM
                                                              Intel might be moving from ISA to platformIan Ollmann08/23/10 09:21 PM
                                                          Intel might be moving from ISA to platformajensen08/22/10 10:36 PM
                                                            Intel might be moving from ISA to platformIan Ollmann08/23/10 09:45 PM
                                                      Final note (and some remarks on Intel's paper)AM08/19/10 01:29 AM
                                                        Final note (and some remarks on Intel's paper)Richard08/19/10 04:51 AM
                                                          I stand corrected.Michael S08/19/10 05:30 AM
                simple reality check for youanonymous08/03/10 10:25 AM
                  simple reality check for youAM08/03/10 11:47 PM
                    simple reality check for youanonymous08/04/10 10:17 AM
                      simple reality check for youanon08/04/10 08:33 PM
                      SPECint/FPAM08/05/10 03:49 AM
                  simple reality check for youRohit08/13/10 08:31 PM
                    simple reality check for youhobold08/16/10 06:29 AM
              maybe texture caches ?Richard Cownie08/03/10 09:27 AM
                maybe texture caches ?none08/03/10 09:59 AM
                  maybe texture caches ?Richard Cownie08/03/10 10:15 AM
                    maybe texture caches ?Mark Roulo08/03/10 10:23 AM
                      maybe texture caches ?Richard Cownie08/03/10 10:33 AM
                        maybe texture caches ?Mark Roulo08/03/10 10:37 AM
                          the Hess paperRichard Cownie08/03/10 11:16 AM
                    maybe texture caches ?Aaron Spink08/03/10 01:23 PM
                      maybe texture caches ?Mark Roulo08/03/10 04:13 PM
                        maybe texture caches ?Michael S08/04/10 01:47 AM
                  maybe texture caches ?Michael S08/03/10 10:23 AM
                maybe texture caches ?Mark Roulo08/03/10 10:19 AM
                  maybe texture caches ?Richard Cownie08/03/10 10:42 AM
                Cache to cache comparisonDavid Kanter08/03/10 12:28 PM
                  Cache to cache comparisonRichard Cownie08/03/10 12:52 PM
                    Cache to cache comparisonDavid Kanter08/03/10 01:21 PM
                      Cache to cache comparisonRichard Cownie08/03/10 03:55 PM
                      Cache to cache comparisonGabriele Svelto08/03/10 11:23 PM
                        Nehalem vs Core2Richard Cownie08/04/10 04:24 PM
                          Nehalem vs Core2IntelUser200008/04/10 07:26 PM
                          Nehalem vs Core2Gabriele Svelto08/04/10 10:51 PM
                            Nehalem vs Core2Richard Cownie08/05/10 03:27 AM
                              Nehalem vs Core2Gabriele Svelto08/05/10 06:00 AM
                                Nehalem vs Core2Richard Cownie08/05/10 07:10 AM
                                  Nehalem vs Core2Gabriele Svelto08/05/10 07:41 AM
                                    Nehalem vs Core2Michael S08/05/10 10:02 AM
                              Nehalem vs Core2Michael S08/05/10 10:12 AM
                          Nehalem vs Core2Michael S08/05/10 12:36 AM
                            Nehalem vs Core2Richard Cownie08/05/10 08:33 AM
                              Nehalem vs Core2Michael S08/05/10 09:44 AM
                                Nehalem vs Core2Richard Cownie08/05/10 10:03 AM
                                  Nehalem vs Core2Michael S08/05/10 04:21 PM
                                    Nehalem vs Core2Richard Cownie08/05/10 07:15 PM
                                      I'm wrong, you're right - i's 2 dieRichard Cownie08/05/10 07:21 PM
                                    Nehalem vs Core2Richard Cownie08/06/10 08:12 AM
                                      Nehalem vs Core2Michael S08/06/10 08:33 AM
                                      Nehalem L1 cache latency a hedge for higher clock speeds?Mark Roulo08/06/10 08:36 AM
                                        Nehalem L1 cache latency a hedge for higher clock speeds?Kevin G08/07/10 10:14 AM
                                          Nehalem L1 cache latency a hedge for higher clock speeds?IntelUser200008/07/10 02:32 PM
                                        Nehalem L1 cache latency a hedge for higher clock speeds?someone08/07/10 02:35 PM
                          difference between C2D/C2Q chipsetsMichael S08/05/10 12:57 AM
                          Nehalem core goals ... my takeMark Roulo08/06/10 07:53 AM
                            Bulldozer single-threadRichard Cownie08/06/10 04:05 PM
                          Nehalem vs Core2Carlie Coats08/07/10 08:41 AM
                            scalingMichael S08/07/10 11:12 AM
                            why X5460?Michael S08/07/10 11:12 AM
                              why X5460?Carlie Coats08/08/10 04:34 AM
                                why X5460?Michael S08/08/10 04:42 AM
                                  Polling/PIO based COMM libraries?Michael S08/08/10 05:14 AM
                    Cache to cache comparisonAnts Aasma08/04/10 10:00 AM
                      Cache to cache comparisonRichard Cownie08/04/10 10:08 AM
                        Cache to cache comparisonAnts Aasma08/04/10 11:28 AM
                      Cache to cache comparisonMichael S08/04/10 11:33 AM
                        Cache to cache comparisonMark Roulo08/04/10 12:29 PM
                          Cache to cache comparisonAnts Aasma08/04/10 01:10 PM
                            Fermi and G200 instruction latencyMark Roulo08/04/10 02:10 PM
                              Fermi and G200 instruction latencyAnts Aasma08/04/10 03:01 PM
                        Cache to cache comparisonAnts Aasma08/04/10 01:00 PM
                        Register read comparisonDavid Kanter08/05/10 09:16 AM
                          Register read comparisonGabriele Svelto08/05/10 11:42 PM
                maybe texture caches ?Aaron Spink08/03/10 01:17 PM
              GPU raw H/W only has about a 10x advantage over a CPURichard08/09/10 04:58 AM
                GPU raw H/W only has about a 10x advantage over a CPUnone08/09/10 05:19 AM
                  GPU raw H/W only has about a 10x advantage over a CPUVincent Diepeveen08/24/10 07:18 AM
                    Forget it, they won't hear itanon08/24/10 04:37 PM
                    GPU raw H/W only has about a 10x advantage over a CPURichard08/25/10 01:03 PM
                This forum supports pre tagMichael S08/09/10 06:03 AM
    Tarek's Hotpar 2010 article onlineRich Vuduc07/28/10 10:24 AM
      Tarek's Hotpar 2010 article onlineAnon07/28/10 04:18 PM
  Low hanging fruitMoritz02/03/11 12:20 PM
    Low hanging fruitJukka Larja02/06/11 05:01 AM
Reply to this Topic
Name:
Email:
Topic:
Body: No Text
How do you spell blue?