Comments on: A Scalable Architecture for Low Latency Pricing http://www.cdegroot.com/blog/2006/01/16/a-scalable-architecture-for-low-latency-pricing/ Everything and the kitchen sink Thu, 20 Nov 2008 18:13:53 +0000 http://wordpress.org/?v=2.0.3 by: Geva Perry http://www.cdegroot.com/blog/2006/01/16/a-scalable-architecture-for-low-latency-pricing/#comment-123 Fri, 17 Feb 2006 00:25:00 +0000 http://www.cdegroot.com/blog/2006/01/16/a-scalable-architecture-for-low-latency-pricing/#comment-123 We at GigaSpaces have done a lot to give you the benefits of Jini/JavaSpaces/Rio with what we hope is a much more friendly and easy to use implementation, especially in our just released version 5.0. You're welcome to download it and play around with it. By the way, we are co-publishers of the white paper cdegroot is referring to. Geva www.gigaspaces.com We at GigaSpaces have done a lot to give you the benefits of Jini/JavaSpaces/Rio with what we hope is a much more friendly and easy to use implementation, especially in our just released version 5.0.
You’re welcome to download it and play around with it. By the way, we are co-publishers of the white paper cdegroot is referring to.

Geva
www.gigaspaces.com

]]> by: Mike http://www.cdegroot.com/blog/2006/01/16/a-scalable-architecture-for-low-latency-pricing/#comment-108 Sat, 28 Jan 2006 02:49:53 +0000 http://www.cdegroot.com/blog/2006/01/16/a-scalable-architecture-for-low-latency-pricing/#comment-108 Oh my the GreenTea trolls again. cdegroot, these guys used to spam the old jini and javaspaces mailing lists all the time. Just ignore them.. Oh my the GreenTea trolls again. cdegroot, these guys used to spam the old jini and javaspaces mailing lists all the time. Just ignore them..

]]> by: cdegroot http://www.cdegroot.com/blog/2006/01/16/a-scalable-architecture-for-low-latency-pricing/#comment-89 Fri, 20 Jan 2006 07:35:10 +0000 http://www.cdegroot.com/blog/2006/01/16/a-scalable-architecture-for-low-latency-pricing/#comment-89 Jini heavyweight? Gosh, that's the first time I've heard it being described this way :-) It is true that the out-of-the-box experience of Jini at least used to be less than ideal. But that, of course, is not a very valid reason to stay away from such a product. What I do know, is that Jini probably approaches the absolutely minimal set of functionality you need to build a reliable distributed system. It all depends, of course, if you subscribe to the ideas the Jini people had on distributed computing, like very much taking into account <a href="http://today.java.net/jag/Fallacies.html" rel="nofollow">Deutsch' Fallacies of Distributed Computing</a> and their <a href="http://research.sun.com/techrep/1994/abstract-29.html" rel="nofollow">ideas on transparency</a><a>. I will look at GreenTea - thanks for the pointer, I am always interested in this sort of stuff even though I don't do Java these days - but I fear that anything they have made it simpler means that they go below this minimum set of functionality for <em>reliable</em> distributed computing. Building a distributed system is trivial. Building a distributed system that works in the real world is <em>hard</em>. Maybe I'm biased because I've been so deeply involved with Jini, but I think that this is one of the very few products out there that is "rightsized".</a> Jini heavyweight? Gosh, that’s the first time I’ve heard it being described this way :-)

It is true that the out-of-the-box experience of Jini at least used to be less than ideal. But that, of course, is not a very valid reason to stay away from such a product.

What I do know, is that Jini probably approaches the absolutely minimal set of functionality you need to build a reliable distributed system. It all depends, of course, if you subscribe to the ideas the Jini people had on distributed computing, like very much taking into account Deutsch’ Fallacies of Distributed Computing and their ideas on transparency.

I will look at GreenTea - thanks for the pointer, I am always interested in this sort of stuff even though I don’t do Java these days - but I fear that anything they have made it simpler means that they go below this minimum set of functionality for reliable distributed computing.

Building a distributed system is trivial. Building a distributed system that works in the real world is hard. Maybe I’m biased because I’ve been so deeply involved with Jini, but I think that this is one of the very few products out there that is “rightsized”.

]]> by: Gary http://www.cdegroot.com/blog/2006/01/16/a-scalable-architecture-for-low-latency-pricing/#comment-88 Fri, 20 Jan 2006 04:19:57 +0000 http://www.cdegroot.com/blog/2006/01/16/a-scalable-architecture-for-low-latency-pricing/#comment-88 I've heard people complain about the complexity of Jini and it's difficult to set up and use , and quite a lot of people go with a light weight solution called GreenTea: http://www.greenteatech.com . We are a startup-like software team and would prefer a light weight solution so that we can develop a distributed application right away. Have you tried it out? It would be interesting to see your thoughts on that. Gary I’ve heard people complain about the complexity of Jini and it’s difficult to set up and use , and quite a lot of people go with a light weight solution called GreenTea: http://www.greenteatech.com . We are a startup-like software team and would prefer a light weight solution so that we can develop a distributed application right away. Have you tried it out? It would be interesting to see your thoughts on that.

Gary

]]>