Tag Archives: distributed systems

Remediating the Internet’s outstanding SPOFs

Mike Masnick writes a little article forecasting the engineers re-writing the single points of failure out of the internet. He entitles his article, Building A More Decentralized Internet: It’s Happening Faster Than People Realize. He cross references to two articles written by himself back in 2010, Operation Payback And Wikileaks Show The Battle Lines Are About Distributed & Open vs. Centralized & Closed and The Revolution Will Be Distributed: Wikileaks, Anonymous And How Little The Old Guard Realizes What’s Going On in which he, more accurately, recognises the current and future power of distributed and private networks. It should be remembered that these predictions all occurred before the Arab spring and the recent protests in Turkey and the state responses to the use of networks. One of the key initiatives proposed in my mind, is to develop a P2P name service resolver, while others propose a P2P file system. I wrote a wiki article, called “Ruggedising the Internet” which points at several further resources and projects. I might even join in.

Read more …

Cloud why now?

I wrote and published a paper, called Cloud why now? on the Knowledge Transfer Network site. How new is Cloud Computing? It is a clear evolution if two trends in IT architecture that have had an immense and limited success. The successful trend is distributed computing and the less successful one utility computing. What is driving this evolution and why now? This article has a quick look at the trends that have brought this to this point and looks at the fact that like most economic revolutions, it’s a confluence of both socio/economics and science. The article concludes by looking at the paradox of data, to see how it is both a driver of change, and an inhibitor.

Read more …

Data Centre Economies of Scale

At the Waters:Power09, last week, Bob Giffords argued there are three ‘gravitational’ forces leading to the mega data centres and cloud computing.

  • There’s too much data to move, it needs to stay where its created.
  • Intra system & total latency is still a problem, and hence systems are best co-located with the data.
  • He argues that energy management is a gravitational issue.
Read more …

Sun’s Software Engineering Summit 2005

Greg Papapdopolus (SUNW CTO), opened the conference and talked about some developments and forecasts on which Sun medium term plans are based, he only took a couple of minutes and will be speaking at length later in the week. One of his points is the growing support for “the Network is the Computer”. Greg argued that we’re reaching a world in which people think that if they can’t get google, the computer’s broken. You can read his own words here…..

Read more …

Utility Computing

I attended an all day seminar in Utility Computing. Couple of interesting presentations, the day was opened by Jim Baty and closed by Bill Vass. Frankly, its been a patchy day and I’m still not sure how firms requiring competitive advantage from their IT can leverage utility offerings because the suppliers will need a degree of homogeneity. It was definitely interesting and heartening to hear that several customers, who have the expertise to build the complex grids on which today’s utility offerings are to be based, are still coming to Sun;

Read more …

Real Options & Flexible Planning

Kieron Bradley, one of my colleagues at Sun, during a piece of client consulting recently had reason to use financial option theory & language to justify why CPU’s in Sun’s large systems are more expensive to buy than those in the smaller ones. He and the customer had examined all the TCOO factors they thought were relevant and the fact remained that if one wanted to take a utility view of CPU supply, it was cheaper to buy and run smaller systems rather than larger ones. (This particular analysis did not perform a variable utilisation analysis. It was assumed, (or defined as policy) that all CPUs would run at a given % utilisation. Contradicting this assumption, it is a fact that large (and flexibly partitioned) systems are easier to keep busy.)

Read more …