<<< Previous speaker next speaker >>>

Joe Albahari, Egton Australia

Joe Albahari Joseph Albahari, author of C# 3.0 in a Nutshell, is a software architect and developer with 17 years' experience in the health, education and telecommunication industries. He is currently Chief Systems Architect at Egton Australia, supplier for the largest primary healthcare software provider in the UK. Joe has a keen interest in LINQ, and is author of LINQPad - the code snippet IDE for interactively querying databases in LINQ. He specializes in integrating LINQ into corporate multi-tier applications and creating extensions to maximize query composability.

Presentation: "PLINQ and Parallel FX"

Time: Friday 11:15 - 12:00

Location: To be announced

Abstract: With CPUs on the horizon boasting a core count of 8, 16, or more, the burning question for software developers is, "how could we possibly keep all those cores busy?" Anyone familiar with the multithreading APIs currently on offer is sure to have uttered the mantra, "there must be a better way". PLINQ and the Task Parallel Library promise to provide that better way.   In this talk, we discuss each of these upcoming technologies. The Task Parallel Library can be considered a modern replacement for asynchronous delegates and the ThreadPool class; PLINQ is a very cool technology that transparently parallelizes LINQ queries. We then look at plausible use cases, and how well these solutions address the hard problems of thread safety and resource utilization.

Presentation: "LINQ to SQL: Taking the Boredom out of Querying"

Time: Friday 13:00 - 13:45

Location: Ballroom Le Grand 1

Abstract:

LINQ is more than just a vehicle for query language integration and static type safety - it's also a better querying language than SQL. In this session, we'll examine how this is so, and how LINQ to SQL cuts the development time in writing a data access layer.

We'll also look at strategies for designing data context classes, patterns for query reusability, and how to push C# 3.0's limits in expression composability. We'll then take an honest look at the collateral damage - in terms of CPU overhead and query optimization - and round up the scenarios that, at present, evade this kind of technology.