Progress Software SonicMQ the fastest JMS server


News: Progress Software SonicMQ the fastest JMS server

  1. Progress Software, maker of the SonicMQ JMS server has released benchmark comparison between SonicMQ and Fiorano MQ, showing SonicMQ being 43% to 96% faster than Fiorano MQ. An older benchmark also shows SonicMQ performing 50% to 3584% faster than IBM MQSeries.

    Even more interesting than the bench mark figures is SonicMQ's ultra agressive anti-Fiorano marketing. The benchmark comparison page contains a "SonicMQ Migration Program for Fiorano Customers", and quotes Atul Saini, CEO of Fiorano saying "Java's a great language for productivity. It's just not fast. So, real programmers want their server written in a native language. Something like C++." SonicMQ is entirely implemented in Java.

    As more application servers integrate JMS offerings to become J2EE compliant, I think that we can expect to see an overall marketshare of the JMS implementation vendors declining. The remaining vendors will have to compete to show that their implementations are better than each others, and better than prepackaged JMS servers such as the one included with Weblogic. The overall goal of these JMS vendors is of course to gain customers, but also to have application server vendors integrate their JMS implementation with their own application server products.

    Read the press release.

    Read SonicMQ vs. Fiorano MQ.

    Read the SonicMQ vs. IBM MQ series whitepaper.
  2. Why don't they mention anything about the Softwired iBus JMS product ? They have an advanced clustering solution and really good scalability and performance, not to mention the results in wireless transaction support.
    There is a chance for the JMS market to look as crowded as the app server market looks today, but I don't think it is wise to start this stupid "wars". After all, I would wait for an independent benchmark before making all this noise ;-)
  3. Over a long time Fiorano was claiming to be the fastest JMS server, based on one very simple test. Fiorano itself was very aggresive. Now, last week, published an interview with Atul Saini, Fiorano's CEO, where he states that - in short words - the performance limit of Java for messaging is reached and that only a C++ server impl will help and that Fiorano plans to offer such server (with JMS clients) next year.

    I though, hmmmm, does this man knows anything about architectural issues? Does he plan to use separate processes for threads he uses now in Java or will he bundle the RogueWave Threads lib with his server?

    However, I guess that Progress see a chance to hit Fiorano and - maybe - to throw him out of the JMS market. Announcing a migration program which is based on that interview. You are right, that's ultra-ultra aggressive.

    For the current braggart to be the fastest, SonicMQ, let me say that this is not even more serious than Fiorano's approach to enter the JMS throne, because both are based on selfgenerated tests.

    The JMS 'marketing' field is a bad area. Strange coalitions are appearing. Richard Monson-Haefel, lead of OpenEJB which is a cousin of OpenJMS, writes a book co-authored with Dave Chappell, VP from Progress. Strange? Very strange!

    What I'm not understand is why the vendor of the fastest JMS system, SonicMQ, has the very top hitcount on our website,

    It's amusing, isn't it?

    Andreas Mueller
    CEO, IIT GmbH,
    Vendor of SwiftMQ,
    The most popular JMS system in the world, believe me!! ;-)
  4. I am curious to hear about the business strategy that JMS vendors plan to take now that consumers are looking for a single J2EE vendor.
  5. There doesn't seem to be one. Either that or it's not shareable in a public forum.

    I see two possibilities off the top of my head. Either they build upward into the enterprise with tools such as workflow engines or message broker. Or they get bought out by one of the major App Server/Integration firms like IBM/BEA/Tibco....