The Artima Developer Community
Sponsored Link

Java Community News
Van Simmons on the ComputeCycles Compute Grid Project

2 replies on 1 page. Most recent reply: Sep 30, 2006 8:43 PM by Todd Blanchard

Welcome Guest
  Sign In

Go back to the topic listing  Back to Topic List Click to reply to this topic  Reply to this Topic Click to search messages in this forum  Search Forum Click for a threaded view of the topic  Threaded View   
Previous Topic   Next Topic
Flat View: This topic has 2 replies on 1 page
Bill Venners

Posts: 2284
Nickname: bv
Registered: Jan, 2002

Van Simmons on the ComputeCycles Compute Grid Project Posted: Sep 22, 2006 3:12 PM
Reply to this message Reply
Summary
The JavaPosse has published part I of an interview with Van Simmons on the ComputeCycles project, which uses Jini, JavaSpaces, Glassfish, and Groovy with the goal of creating a dynamically configurable, easy-to-administer secure compute grid.
Advertisement
In Interview with Van Simmons about Jini and ComputeCycles, Van Simmons gives a personal history of Jini, and describes the ComputeCycles project. From the ComputeCycle's home page:

One of the challenges in grid computing is dynamic configuration of the grid. Java is uniquely suited to meet this challenge because it offers the ability to combine mobile code and standard APIs with a powerful security model. These capabilities make it possible to build scalable grid computing systems which require minimal centralized configuration infrastructure and simplify "on-the-fly" deployment and removal of grid assets. We call such a minimally configured grid a Flashgrid.

The goal of the ComputeCycles project is to build a "flashgrid-capable" compute grid which:

  1. Has zero single-points-of-failure
  2. Is secure
  3. Is easily administered
  4. Is dynamically configured
  5. Is based on the Master/Worker pattern (eventually moving to a full workflow model)
  6. Can be run securely through firewalls
  7. Assumes Java and specifically Jini everywhere

What kind of problems do you have for which you think a compute grid, such as the one envisioned by the ComputeCycles project, would be an appropriate tool?


Cameron Purdy

Posts: 186
Nickname: cpurdy
Registered: Dec, 2004

Re: Van Simmons on the ComputeCycles Compute Grid Project Posted: Sep 25, 2006 4:28 AM
Reply to this message Reply
Just out of curiousity, how does this compare with what Paremus does with OSGi?

Peace,

Cameron Purdy
http://www.tangosol.com/

Todd Blanchard

Posts: 316
Nickname: tblanchard
Registered: May, 2003

Why limit it to Java? Posted: Sep 30, 2006 8:43 PM
Reply to this message Reply
>Java is uniquely suited to meet this challenge because it offers the ability to
>combine mobile code and standard APIs with a powerful security model.

I don't see where Java is any better at utility computing than any other platform. Mobile code is a red herring. Code seldom moves unless you are rolling a new deployment. The security model is also bogus. The Java security model is designed to protect against the possibility of hostile foreign code as in applets. It does nothing for server hosted code and is often an annoyance.

Lets be real. Security starts at the OS and network configuration.

In fact, there is a grid project that is real, available, and doesn't care if you run Java or not. Amazon's EC2 will let you bring up computers configured with whatever disk image you like. Java could be there, but doesn't have to be. Amazon itself uses this system to add capacity to their website and precious little of that is written in Java.

The point is, Java doesn't help and isn't uniquely suited for this sort of thing at all. You saying so just propagates propaganda.

http://www.amazon.com/gp/browse.html?node=201590011

Flat View: This topic has 2 replies on 1 page
Topic: Van Simmons on the ComputeCycles Compute Grid Project Previous Topic   Next Topic Topic: Testing Concurrent Programs

Sponsored Links



Google
  Web Artima.com   

Copyright © 1996-2019 Artima, Inc. All Rights Reserved. - Privacy Policy - Terms of Use