Tuesday, 2 February 2010
IBM WebSphere eXtreme Scale 6 by Anthony Chaves, my closing comments
As per my previous posts here and here, I completed this book a few days ago, and would summarise my review as follows. This is an excellent book that really digs into the depths of In-Memory Data Grids (IMDGs), with specific focus on the WebSphere product.
The book is broken down into discrete and logical chapters: -
Chapter 1 What is a Data Grid
A good overview of the overall Data Grid approach, with an explanation of the architecture, and its benefits when compared to relational databases and In-Memory Data Bases (IMDBs).
Chapter 2 The ObjectMap API
A deep dive into the core API behind the IMDG concept, with relevant Java code samples
Chapter 3 Entities and Queries
Using the SQL-like Query API to interact with the Data Grid
Chapter 4 Database Integration
Comparing the Data Grid with relational databases, including the costs/benefits e.g. latency of memory access vs. disk access etc.
Chapter 5 Handling Increased Load
How an IMDG can scale up and out
Chapter 6 Keeping Data Available
How availability non-functional requirements can be met with replication
Chapter 7 The DataGrid API
How IMDG breaks the traditional mould of bringing the data to the application, by moving application code back out into the grid, in a manner similar to, but not the same as SQL Stored Procedures
Chapter 8 Data Grid Patterns
How Extreme Transaction Processing (XTP) requirements can be met using the IMDG concept
Chapter 9 Spring Integration
Using the Spring framework to interact with the IMDG, where Javabeans etc. are instantiated by Spring rather than within the Java code itself
Chapter 10 Putting It All Together
Taking a "real world" example application for bookmark storage, management and access, and moving it from a somewhat "kludged" pattern-light model to something a lot more formalised and structured
As per my previous comments, as one might expect, the book is heavy on the Java code, and is likely to be of more use to an enterprise application developer, especially one working with highly-performance transaction/compute-intensive data models, *BUT* the overall architecture, benefits and implementation best practices are clearly established within the book. As an infrastructure architect, I've learnt some important lessons that, although not specifically relevant to the projects on which I'm working right now, are of value now, and almost certainly, in the future.
In conclusion, I'd strongly recommend this book to anyone wishing to explore In-Memory Data Grids, whether they intend to utilise the WebSphere eXtreme Scale platform or not.
Mr Chaves, I thank you :-)
# posted by Dave Hay : 19:53
Subscribe to Posts [Atom]
