http://www.linux-mag.com/id/6744/ Open Source, Open Standards Sat, 04 Jul 2009 23:06:29 +0000 http://wordpress.org/?v=2.0.11 http://www.linux-mag.com/id/6744/#comment-1435 Tue, 02 Sep 2008 02:51:41 +0000 http://www.linux-mag.com/id/6744/#comment-1435 [...] Again, John West reads more than I, and notes at InsideHPC.com blog, an article from Doug Eadline on Linux Magazine, all about really big clusters. [...] […] Again, John West reads more than I, and notes at InsideHPC.com blog, an article from Doug Eadline on Linux Magazine, all about really big clusters. […]

]]> http://www.linux-mag.com/id/6744/#comment-1434 Tue, 02 Sep 2008 00:58:56 +0000 http://www.linux-mag.com/id/6744/#comment-1434 [...] Doug Eadline has an interesting article in Linux Magazine from last week about what nature — in particular, really large colonies of organisms — can teach us about managing extremely large computing clusters As we approach the really big threshold, we may need to look at how nature has solved the scale problem. The first thing that hits me is redundancy and fault tolerance. Just like the ant colony, a large cluster and the software running on it will need to adapt to constant node and network failure. As clusters get bigger, nodes and networks are going to fail at increasing rates. Programs and networks may need to manage themselves without any reliance on a single point of control or at least minimizing those points as much as possible. That is, programs may need to find resources on their own and not wait for a central authority to allocate them. There is no one telling ant number 183,234 to go pick up that crumb over there. The colony adapts dynamically to its needs. [...] […] Doug Eadline has an interesting article in Linux Magazine from last week about what nature — in particular, really large colonies of organisms — can teach us about managing extremely large computing clusters As we approach the really big threshold, we may need to look at how nature has solved the scale problem. The first thing that hits me is redundancy and fault tolerance. Just like the ant colony, a large cluster and the software running on it will need to adapt to constant node and network failure. As clusters get bigger, nodes and networks are going to fail at increasing rates. Programs and networks may need to manage themselves without any reliance on a single point of control or at least minimizing those points as much as possible. That is, programs may need to find resources on their own and not wait for a central authority to allocate them. There is no one telling ant number 183,234 to go pick up that crumb over there. The colony adapts dynamically to its needs. […]

]]> http://www.linux-mag.com/id/6744/#comment-1418 Thu, 28 Aug 2008 14:46:59 +0000 http://www.linux-mag.com/id/6744/#comment-1418 Note this is not merely an academic exercise. The Franklin machine at NERSC has about 20,000 cores and is being upgraded to about 40,000 cores (each with 4 GBytes of RAM and high-speed low-latency network, including access to a 350 TByte disk farm). This is not the largest parallel machine in the world. Note this is not merely an academic exercise. The Franklin machine at NERSC has about 20,000 cores and is being upgraded to about 40,000 cores (each with 4 GBytes of RAM and high-speed low-latency network, including access to a 350 TByte disk farm). This is not the largest parallel machine in the world.

]]> http://www.linux-mag.com/id/6744/#comment-1416 Thu, 28 Aug 2008 12:58:16 +0000 http://www.linux-mag.com/id/6744/#comment-1416 this is an interesting article <a>Interesting article</a> this is an interesting article
Interesting article

]]>