Comments on: Multicore’s Not-So-Secret Problem http://gigaom.com/2008/06/19/multicores-not-so-secret-problem/ Business, Internet, Technology & Strategy Wed, 15 Jul 2009 07:13:40 +0000 http://wordpress.com/ hourly 1 By: Microsoft and Proprietary Chips : Beyond Search http://gigaom.com/2008/06/19/multicores-not-so-secret-problem/#comment-937088 Microsoft and Proprietary Chips : Beyond Search Fri, 10 Apr 2009 15:10:37 +0000 http://gigaom.com/?p=13832#comment-937088 [...] her write up was: The issue of getting software performance to scale linearly with the addition of more cores has become a vexing problem. Plus, as data center operators look for better application performance without expending as many [...] [...] her write up was: The issue of getting software performance to scale linearly with the addition of more cores has become a vexing problem. Plus, as data center operators look for better application performance without expending as many [...]

]]> By: Is Microsoft Turning Away From Commodity Servers? http://gigaom.com/2008/06/19/multicores-not-so-secret-problem/#comment-936986 Is Microsoft Turning Away From Commodity Servers? Thu, 09 Apr 2009 19:59:40 +0000 http://gigaom.com/?p=13832#comment-936986 [...] issue of getting software performance to scale linearly with the addition of more cores has become a vexing problem. Plus, as data center operators look for better application performance without expending as many [...] [...] issue of getting software performance to scale linearly with the addition of more cores has become a vexing problem. Plus, as data center operators look for better application performance without expending as many [...]

]]> By: The Hunt for a Universal Compiler Gets $16M http://gigaom.com/2008/06/19/multicores-not-so-secret-problem/#comment-936581 The Hunt for a Universal Compiler Gets $16M Tue, 07 Apr 2009 23:01:41 +0000 http://gigaom.com/?p=13832#comment-936581 [...] Research Projects Agency to develop a universal compiler that will run on heterogeneous hardware and multicore platforms, which are found in everything from supercomputers to embedded systems, such as those used in [...] [...] Research Projects Agency to develop a universal compiler that will run on heterogeneous hardware and multicore platforms, which are found in everything from supercomputers to embedded systems, such as those used in [...]

]]> By: Big Computer Brains Need Big Memory Bandwidth http://gigaom.com/2008/06/19/multicores-not-so-secret-problem/#comment-930285 Big Computer Brains Need Big Memory Bandwidth Sat, 07 Mar 2009 01:01:11 +0000 http://gigaom.com/?p=13832#comment-930285 [...] Daniel Reed, Microsoft’s scalable and multicore computing strategist, calls this a hidden problem that’s just as big as the challenges of developing code that optimizes multicore chips. [...] [...] Daniel Reed, Microsoft’s scalable and multicore computing strategist, calls this a hidden problem that’s just as big as the challenges of developing code that optimizes multicore chips. [...]

]]> By: Looking At Solutions From The Designer and Economist Points Of View at MakerLab Blog http://gigaom.com/2008/06/19/multicores-not-so-secret-problem/#comment-915033 Looking At Solutions From The Designer and Economist Points Of View at MakerLab Blog Fri, 28 Nov 2008 08:54:12 +0000 http://gigaom.com/?p=13832#comment-915033 [...] This quote from Stacey Higginbotham of GigaOM summed it up: As anyone who’s ever hosted a demolition party well knows, you can only throw so many workers at a problem before people start to linger at the edges, swill your alcohol and generally stop helping. [...] [...] This quote from Stacey Higginbotham of GigaOM summed it up: As anyone who’s ever hosted a demolition party well knows, you can only throw so many workers at a problem before people start to linger at the edges, swill your alcohol and generally stop helping. [...]

]]> By: Multicore’s Not-So-Secret Problem http://gigaom.com/2008/06/19/multicores-not-so-secret-problem/#comment-905483 Multicore’s Not-So-Secret Problem Fri, 17 Oct 2008 11:47:15 +0000 http://gigaom.com/?p=13832#comment-905483 [...] Full Story [...] [...] Full Story [...]

]]> By: IBM’s Cell Chip Gets Another Home: TV - GigaOM http://gigaom.com/2008/06/19/multicores-not-so-secret-problem/#comment-902491 IBM’s Cell Chip Gets Another Home: TV - GigaOM Wed, 01 Oct 2008 17:48:53 +0000 http://gigaom.com/?p=13832#comment-902491 [...] pushing the Cell processor for data center applications, too, analysts scoffed, saying there were few programs out there to take advantage of the multicore architecture. Yet today, multicore chips are in everything from servers and desktops to routers, and firms from [...] [...] pushing the Cell processor for data center applications, too, analysts scoffed, saying there were few programs out there to take advantage of the multicore architecture. Yet today, multicore chips are in everything from servers and desktops to routers, and firms from [...]

]]> By: Happy 50th Birthday, Dear Microchip - GigaOM http://gigaom.com/2008/06/19/multicores-not-so-secret-problem/#comment-899038 Happy 50th Birthday, Dear Microchip - GigaOM Fri, 12 Sep 2008 12:00:36 +0000 http://gigaom.com/?p=13832#comment-899038 [...] will need to change. Already chip companies are turning from increasing the clock speed of chips to increasing the number of cores on a chip. They’re also trying to replace wires with light or store more data in semiconductors that [...] [...] will need to change. Already chip companies are turning from increasing the clock speed of chips to increasing the number of cores on a chip. They’re also trying to replace wires with light or store more data in semiconductors that [...]

]]> By: John Keane http://gigaom.com/2008/06/19/multicores-not-so-secret-problem/#comment-894916 John Keane Wed, 20 Aug 2008 07:36:18 +0000 http://gigaom.com/?p=13832#comment-894916 It is a tough one. Just remember that computing is still in infancy. Shame on Microsoft and Intel. I do hope those universities are not offended by the loose change offered. It is a tough one. Just remember that computing is still in infancy. Shame on Microsoft and Intel. I do hope those universities are not offended by the loose change offered.

]]> By: Five Multicore Startups to Watch - GigaOM http://gigaom.com/2008/06/19/multicores-not-so-secret-problem/#comment-888861 Five Multicore Startups to Watch - GigaOM Thu, 17 Jul 2008 00:00:16 +0000 http://gigaom.com/?p=13832#comment-888861 [...] software developers and engineers mindset needs to adapt. For to really take advantage of multiple cores, a programmer needs to look at ways to make her code parallel, splitting jobs into different parts rather than [...] [...] software developers and engineers mindset needs to adapt. For to really take advantage of multiple cores, a programmer needs to look at ways to make her code parallel, splitting jobs into different parts rather than [...]

]]> By: DreamWorks and Big Oil Put Multicore to Work - GigaOM http://gigaom.com/2008/06/19/multicores-not-so-secret-problem/#comment-887992 DreamWorks and Big Oil Put Multicore to Work - GigaOM Fri, 11 Jul 2008 09:00:54 +0000 http://gigaom.com/?p=13832#comment-887992 [...] it comes to multicore chips, big bucks are on the line as the chip firm or software company that figures out how to write code to take advantage of them stands to make boatloads of [...] [...] it comes to multicore chips, big bucks are on the line as the chip firm or software company that figures out how to write code to take advantage of them stands to make boatloads of [...]

]]> By: Hunter http://gigaom.com/2008/06/19/multicores-not-so-secret-problem/#comment-886044 Hunter Fri, 27 Jun 2008 14:53:55 +0000 http://gigaom.com/?p=13832#comment-886044 Dave, DataRush is similar to PeakStream and RapidMind only its a Java framework for developing <a href="http://www.pervasivedatarush.com/node/66" rel="nofollow">dataflow</a> applications. Dave, DataRush is similar to PeakStream and RapidMind only its a Java framework for developing dataflow applications.

]]> By: Mats http://gigaom.com/2008/06/19/multicores-not-so-secret-problem/#comment-885238 Mats Tue, 24 Jun 2008 09:04:08 +0000 http://gigaom.com/?p=13832#comment-885238 Hi, another angle of the problem is described very well in Michael Feldman's article in HPC Wire. http://www.hpcwire.com/features/19362014.html Hi, another angle of the problem is described very well in Michael Feldman’s article in HPC Wire. http://www.hpcwire.com/features/19362014.html

]]> By: Marc’s Voice » Blog Archive » Blogging on June 22nd, 2008 http://gigaom.com/2008/06/19/multicores-not-so-secret-problem/#comment-885029 Marc’s Voice » Blog Archive » Blogging on June 22nd, 2008 Mon, 23 Jun 2008 11:02:50 +0000 http://gigaom.com/?p=13832#comment-885029 [...] Could multicore be the NEXT new black? [...] [...] Could multicore be the NEXT new black? [...]

]]> By: whoopie http://gigaom.com/2008/06/19/multicores-not-so-secret-problem/#comment-884653 whoopie Fri, 20 Jun 2008 17:18:37 +0000 http://gigaom.com/?p=13832#comment-884653 the future is in functional programming, and shared-nothing concurrency. google is already there - look at the tutorials for map/reduce. they are in haskell. imperative programming is dead, most imperative tools will end up performaning worse on massively multicore systems utilizing relatively cheap processors the future is in functional programming, and shared-nothing concurrency. google is already there – look at the tutorials for map/reduce. they are in haskell.

imperative programming is dead, most imperative tools will end up performaning worse on massively multicore systems utilizing relatively cheap processors

]]> By: ronald http://gigaom.com/2008/06/19/multicores-not-so-secret-problem/#comment-884534 ronald Thu, 19 Jun 2008 21:29:52 +0000 http://gigaom.com/?p=13832#comment-884534 @ twyrick Sorry the brain doesn't do fuzzy logic. And the pattern logic is build upon _1_ simple algorithm which generates different models in different regions and layers of the brain. This is what makes it so resilient to errors, where Boolean logic is really prone to errors. This also enables what you call fuzzy, but my guess it's just an evolutionarily approach to handle new or defect brain regions/layers. I don't think this will replace hard core Boolean systems for math modeling, or the GUI for example. I see Boolean more as an Co-Processor for math, while decisions are made on the other site. Best of both worlds. If you know what Intelligence is let me know. @ twyrick
Sorry the brain doesn’t do fuzzy logic.
And the pattern logic is build upon _1_ simple algorithm which generates different models in different regions and layers of the brain. This is what makes it so resilient to errors, where Boolean logic is really prone to errors. This also enables what you call fuzzy, but my guess it’s just an evolutionarily approach to handle new or defect brain regions/layers.
I don’t think this will replace hard core Boolean systems for math modeling, or the GUI for example. I see Boolean more as an Co-Processor for math, while decisions are made on the other site. Best of both worlds.
If you know what Intelligence is let me know.

]]>