Update on Writing and Monitoring HFT systems

I gave a presentation to JAX Frankfurt on Thursday which went well with about 50 attending. My public presentations are available on
GitHub PerfromanceJUG/Presentations

I plan to update my content for JavaOne 2013 as the presentation is very dense,. I plan to cut some of the topics and focus on the aspects which got the audience's interest most. ;)

The OpenHFT project is getting closer to its first releases and I will be including mention of these in my presentation.  Chronicle 2.0 in this project appears to be about 3x faster than Chronicle 1.7.  The initial release will be have cut down functionality and you may not be able to switch immediately.

My hands on Tutorial on writing and monitoring low latency, high throughput systems in Java. (3 hours) at JAX London has been split into an afternoon and morning session as the morning session is already full.

I hope to be speaking at Devoxx Belgium and W-Jax in Munich.  If so, I might see you there.

Comments

  1. I was thinking since your implementation uses quite a bit of sun.* internals,
    (cf:
    public class ByteBufferBytes extends AbstractBytes {
    import sun.nio.ch.DirectBuffer;

    public class DirectStore
    sun.misc.Cleaner;
    ),
    how do you feel about this comment by Mark Reinhold
    that in java 9 these calls are gonna break:

    Mark Reinhold 09:33
    https://www.youtube.com/watch?v=7QSoKarQp8o&list=UUmRtPmgnQ04CMUpSUqPfhxQ#t=23

    ReplyDelete

Post a Comment

Popular posts from this blog

Java is Very Fast, If You Don’t Create Many Objects

System wide unique nanosecond timestamps

Comparing Approaches to Durability in Low Latency Messaging Queues