As CTO and founder of GigaSpaces, Nati Shalom is looking to solve the level of complexity of large-scale distributed applications - particularly with the concept of the space model. In this tech talk, Shalom explains how the space model works and how it differs from JMS. He then goes on to discuss how grids are affecting developers used to working with the classic model of tier-based applications. Other grid topics he covers include: - The grid as a replacement for the database - Network topology's effect on grid performance - Security issues - How grids affect service-oriented architecture - Scaling up - Measuring grid performance - Features customers should look for - Discovering data inside of a space with data grids - How developers should think in terms of grids Watch Nati Shalom on Java Spaces and Grids.
- Posted by: Regina Lynch
- Posted on: August 07 2006 11:50 EDT
- Re: Tech Talk: Nati Shalom on Java Spaces and Grids by Jesse Kuhnert on August 08 2006 00:32 EDT
- Re: Tech Talk: Nati Shalom on Java Spaces and Grids by Cameron Purdy on August 24 2006 09:58 EDT
Sounds like good stuff. I've used GigaSpaces commercially with great success. Keep up the good work guys :)
- The grid as a replacement for the databaseOn this point, I prefer to think of the DataGrid as the meeting point for live data that is used across any number of applications. The database is still going to be the preferred model for managing persistent data, and the amount of infrastructure (reporting, serviceability, security, storage management, migration, etc.) available around the database makes it much more valuable than just as a simple data store. However, over the past 10 years we've saddled databases with tasks that they are not particularly well suited to, and that's why in-memory coherence clustered caching and Data Grids have become popular. Peace, Cameron Purdy Tangosol Coherence: The Java Data Grid