Hello, Martin Fowler has published an article ( http://martinfowler.com/articles/lmax.html) on the LMAX architecture, in case anyone is interested:
>From the article: -- LMAX is a new retail financial trading platform. As a result it has to process many trades with low latency. The system is built on the JVM platform and centers on a Business Logic Processor that can handle 6 million orders per second on a single thread. The Business Logic Processor runs entirely in-memory using event sourcing. The Business Logic Processor is surrounded by Disruptors - a concurrency component that implements a network of queues that operate without needing locks. During the design process the team concluded that recent directions in high-performance concurrency models using queues are fundamentally at odds with modern CPU design. -- Cordialement, Matthew. 2011/6/24 Michael Barker <[email protected]> > Hi, > > Given the recent discussion on concurrency and performance in the Scala > Persistence thread, I thought some of the readers may be interested in an > open source project that some colleagues and I launched recently: > http://code.google.com/p/disruptor/. It's a concurrent structure used at > the heart of a high performance financial exchange that we've been building > over the past 3 years (in Java of course). I also chatted to a few people > at the round-up about the work we were doing. > > It falls squarely into the exceptional case of Dick's "Don't Repeat > Yourself or Others" rule, so we've thrown it out there in open source form. > In its simplest form it's an alternative to a queue, i.e. a structure to > move data between threads. We've also added a couple of classes that give > it the feel of an actor framework (not truly actors, as it's 1 consumer per > thread). It's pretty fast, over 3 orders of magnitude lower latency when > compared to ArrayBlockingQueue. For those that are interested in the gory > details, there's a technical paper on the Google code site that does a very > deep dive into the implementation and describes the results of our > performance tests. > > Mike. > > -- > You received this message because you are subscribed to the Google Groups > "The Java Posse" group. > To view this discussion on the web visit > https://groups.google.com/d/msg/javaposse/-/DynCqAEpU0gJ. > To post to this group, send email to [email protected]. > To unsubscribe from this group, send email to > [email protected]. > For more options, visit this group at > http://groups.google.com/group/javaposse?hl=en. > -- You received this message because you are subscribed to the Google Groups "The Java Posse" group. To post to this group, send email to [email protected]. To unsubscribe from this group, send email to [email protected]. For more options, visit this group at http://groups.google.com/group/javaposse?hl=en.
