I’ve been interviewed for IBM’s developerWorks podcast, this time answering questions about my Designing a Scalable Grid series.
Hear the podcast: IBM developerWorks : Blogs : developerWorks podcasts blog
Read about Designing a Scalable Grid part 1
Read about Designing a Scalable Grid part 2
The second part of the two part series on designing a scalable grid system is now available at developerWorks. This tutorial looks at the hardware of the individual nodes and then at operating system and application considerations.
For example, I take a close look at the optimization of CPU and RAM in your grid node selections, and how to identify and resolve issues with RAM limitations. You can also use CPU selections to help drive and optimize your grid.
For the operating system, I look at the different benefits and advantages of operating system choice, including how some operating systems provide a richer security solution out of the box, while others may be better suited for the large memory or storage requirements associated with a grid manage or distribution node.
In terms of the application, I cover the use of a component, rather than a single application model, and how the methods of distribution of information can change the way you design and expand your grid and may ultimately affect the scalability.
Read: Design a scalable grid, Part 2: Operating system and application
According this this story, and the the original BBC story, the UK population is going to be able to request a home energy meter that will tell us how much energy is being used and how much it is costing.
The meters themselves are a stop gap until we get proper smart meters, but they should help us monitor what is going on. Unfortunately these units wont be good enough that we'll be able to get and long term (or even short term) information without manually recording the details, but it's a good start.
According this this story, and the the original BBC story, the UK population is going to be able to request a home energy meter that will tell us how much energy is being used and how much it is costing.
The meters themselves are a stop gap until we get proper smart meters, but they should help us monitor what is going on. Unfortunately these units wont be good enough that we'll be able to get and long term (or even short term) information without manually recording the details, but it's a good start.
OK, a quick little humourous item that can get to with Google maps:
Alternatively, try this link.
Or view this:
Yes, it really does say ‘Swim across the Atlantic Ocean’. Why, considering the distance, it suggests swimming from New York to Pont Vauban and not to Southampton, or somewhere in the UK instead of going through France is a little bit of a mystery, but I guess if you are going to swim then you are not aversed to a little sight-seeing.
The biggest worry among geeks at the moment is not global warming. It is the ongoing shortage of Internet capacity.
Two stories today highlight the problem. The first, at the BBC website talks about the impending problems of the ever growing need for bandwidth and some of the potential solutions for fixing them. As the commentary mentions, we may well require a completely new suite of technology to replace the aging packet based solution we currently use.
Such radical ideas will be tough and expensive to implement, and it's not clear at this stage how this will work with some other solutions which are rapidly moving towards the packet switched model.
The biggest worry among geeks at the moment is not global warming. It is the ongoing shortage of Internet capacity.
Two stories today highlight the problem. The first, at the BBC website talks about the impending problems of the ever growing need for bandwidth and some of the potential solutions for fixing them. As the commentary mentions, we may well require a completely new suite of technology to replace the aging packet based solution we currently use.
Such radical ideas will be tough and expensive to implement, and it's not clear at this stage how this will work with some other solutions which are rapidly moving towards the packet switched model.
I'll admit to be thoroughly impressed with my two Apple notebooks. I have both a MacBook (13in) and MacBook Pro (17in). I have no problem on either running pretty much anything I want, and on the 17in MBP I frequently run multiple virtual machines within Parallels. Part of that is undoubtedly due to the Intel Core Duo CPU. Of course, that doesn't mean I'm satisfied - I could always use more horsepower if it was available.
Intel's next revision, dubbed Santa Rosa, is due out soon, and I'd love to see it in some new notebooks, as it allows for some better performance increases, including the ability to ramp up the GHz on an individual core (instead of both at the same time), which should mean more speed with a lower power cost. This article goes into a bit more detail.
I'll admit to be thoroughly impressed with my two Apple notebooks. I have both a MacBook (13in) and MacBook Pro (17in). I have no problem on either running pretty much anything I want, and on the 17in MBP I frequently run multiple virtual machines within Parallels. Part of that is undoubtedly due to the Intel Core Duo CPU. Of course, that doesn't mean I'm satisfied - I could always use more horsepower if it was available.
Intel's next revision, dubbed Santa Rosa, is due out soon, and I'd love to see it in some new notebooks, as it allows for some better performance increases, including the ability to ramp up the GHz on an individual core (instead of both at the same time), which should mean more speed with a lower power cost. This article goes into a bit more detail.
Sash Pachev has written a new guide on Understanding MySQL Internals for O’Reilly.
I was one of the technical reviewers on this book and can tell you that it makes a good read if you want to understand more of what is going on under the hood in MySQL. There’s a lot to cover, for example the pluggable storage engine archtiecture, how queries are parsed and executed, and how individual storage engines work.
The book is compact, but very readable, and worth a look if you any interest in how databases are programmed, or are thinking about building your own storage engine to plug into MySQL.