JMS Performance Comparison

Performance Comparison for Publish Subscribe Messaging

Entire contents © 2002 – 2009, Fiorano Software and Affiliates. All rights reserved. Reproduction of this document in any form without prior written permission is forbidden. The information contained herein has been obtained from sources believed to be reliable. Fiorano disclaims all warranties as to the accuracy, completeness or adequacy of such information. Fiorano shall have no liability for errors, omissions or inadequacies in the information contained herein or for interpretations thereof. The opinions expressed herein are subject to change without notice.

1 Topic Scalability -------------------------------------------------------------------------------------------------------------.3 1.4 1.5 1.5 2. System Configuration ----------------------------------------------------------.4 Environment Setup---------------------------------------------------------------------------------------------------------.2 Software Configuration --------------------------------------------------------------------------------------------------.10 Copyright © Fiorano Software and Affiliates.2 Server Scalability------------------------------------------------------------------------------------------------------------. All Rights Reserved .8 2.6 2.3 Test Duration ----------------------------------------------------------------------------------------------------------------.5 Measurement ---------------------------------------------------------------------------------------------------------------.5 1.Table of Contents Executive Summary ------------------------------------------------------------------3 1. Test Methodology --------------------------------------------------------------3 1.3 Persistent Publisher.10 3.1 Hardware Configuration -------------------------------------------------------------------------------------------------.2 Test Scenario’s --------------------------------------------------------------------------------------------------------------.9 3.7 2.5 1.6 Topology ----------------------------------------------------------------------------------------------------------------------.1 Test Conditions--------------------------------------------------------------------------------------------------------------. Non-Durable Subscribers --------------------------------------------------------------. Performance Results ----------------------------------------------------------6 2.10 3. Durable Subscribers----------------------------------------------------------------------------.4 Non-Persistent Publisher.

ActiveMQ 5. This analysis provides a head-to-head comparison of these products designed to illustrate the products’ relative performance characteristics for several messaging scenarios. The final section provides a brief description of the hardware and software configurations.GA and Sun JAVA MQ The tests examine performance under load. This section begins with a brief description of test conditions which are created to test the JMS server.0.ssp The testing tool used for these performance tests is highly configurable and can be used to test any JMS broker. The test scenarios represent stress level conditions for real world applications. This is followed by a section that describes the tests and their results. Copyright © Fiorano Software and Affiliates. Inc.  All tests were run multiple times to assure repeatability. Sonic MQ 7.3. Do note that the different configurations or performance tuning of any JMS broker may potentially yield throughput gains (or losses) for any of these tests. and are available at http://www. All Rights Reserved . It’s observed from the detailed results that the relative performance of the message brokers varies under various conditions. All the JMS brokers were configured with out-of-the-box default values and no performance specific product tuning was carried out for any of them. where a single message broker is required to support many publishers and subscribers.  All test results are recorded after the client connections have been established and publishers/subscribers and other objects had been created.2.Executive Summary This paper presents a performance analysis of publish/subscribe messaging throughput of FioranoMQ 9. While performance analysis should always be conducted for a particular messaging environment. Tibco EMS v4. Changes to the test definitions will produce different throughput rates and this should be considered when attempting to map these results to expected performance of any particular JMS application.1 Test Conditions All the tests were conducted under the following conditions:  Each client runs on a separate JMS connection. the results of these tests suggest that FioranoMQ will deliver messages more efficiently in demanding messaging environments in today’s real-time enterprises. Jboss Messaging 1. 1. The testing methodology and driving program were the ones developed by Sonic Software. This tool allows running and measurement of a wide range of test definitions. 1.4. Test Methodology All the tests described in this section were carried out using a highly configurable testing tool. Also.0.sonicsoftware.3. this tool allows running and measurement of a wide range of test

2 Test Scenario’s The tests were conducted for the most popular messaging models employed using Topics in JMS. d. The results illustrate the scalability of JMS server as more clients (each working on independent JMS Topics) are employed.2 (default configuration mode). Multiple Non-Durable Consumers: These tests observe the performance characteristics of JMS server when a single non-persistent publisher is used to publish messages to multiple non-durable subscribers. Non-Persistent Publishers & Non-Durable Subscribers: This model is typically used by applications which are exchanging high volume of messages and have a requirement of minimum latency. FioranoMQ and others in normal production ready (non-HA) mode. The following tests were conducted based on typical customer use-cases: a. Copyright © Fiorano Software and Affiliates. Multiple Durable Consumers: These tests observe the performance characteristics of JMS server when a single persistent publisher is used to publish messages to multiple durable subscribers. b.  Dups_ok was used by all consumers  All servers were tested in the default mode . ActiveMQ 5. Tibco EMS in "Evaluation" (non-HA) mode. no processing time is introduced at either side of the client message exchanges. Non-Persistent Producer. Allowing publishers to send messages as fast as possible in this manner enables these tests to expose the maximum message throughput rates. Topic Scalability Tests: These tests observe the performance characteristics of JMS server with varying # of Pub/Sub clients on a fixed number of topics. c. Persistent Publishers & Durable Subscribers This model is typically employed by applications which need maximum level of redundancy and need once and only once guarantee of message delivery irrespective of the client or server failure. 1. All Rights Reserved .  During the test.which meant running SonicMQ. Server Scalability Tests: These tests observe the performance characteristics of JMS server with varying # of Topics with fixed # of Pub/Sub clients per topic. In order to generate the highest amount of message load. Performance was measured under maximum load by publishing as many messages as possible using default settings of the server. The test message size was chosen to reflect use cases observed in typical customer proof of concept scenarios. Persistent Producer. no other applications were running and using resources on the system under test. The results illustrate the scalability of JMS server as more clients (all working on same JMS Topic) are employed.

Each product’s message store.6 Topology The topology contains two machines: One for running the clients and the other for running the server. Steady-state is the condition in which message rates exhibit negligible change. log files. No data was collected during ramp-up and ramp-down intervals. sixty-second intervals. Ramp-down intervals are times in which the systems are decreasing their capacity in response to decreased client loads that result from test completion. Ramp-up intervals are times during which the systems are increasing their message handling capacities. All Rights Reserved . publishers and subscribers were established before any testing ramp-up periods were started. 1.1. Tests were run twice. The first two and last intervals were considered “ramp-up” and “ramp-down” intervals. The remaining five intervals were considered “measurement” intervals during which steady-state performance was achieved.5 Measurement Performance data was collected during the five-minute measurement intervals only.3 Test Duration All test scenarios were executed for a total of eight minutes. queues. The system configurations are detailed later in this document.4 Environment Setup All client connections. 1. 1. and topics were deleted and recreated therefore the broker stopped and restarted between each test. These systems having 1GB NIC cards were interconnected using a 1 Gbps peer to peer connection. respectively. typically via resource allocation in response to the newly introduced client load. and measurements were averaged to obtain final results. Each test execution comprised of eight. Copyright © Fiorano Software and Affiliates.

0 14341 23809 21230 17281 Sonic MQ 7.4.3. 2.1 Topic Scalability Performance Results Message P/S/T Message Type Subscriber Type Size (bytes) Fiorano MQ 9.3 16924 17474 12247 662 1/1/1 10/10/1 25/25/1 50/50/1 NonPersistent NonPersistent NonPersistent NonPersistent Non-Durable Non-Durable Non-Durable Non-Durable 1024 1024 1024 1024 Copyright © Fiorano Software and Affiliates. GA 16448 35886 38917 37349 Sun MQ 4.0 12246 22177 24331 19212 Active MQ 5.0.1 45262 81712 88181 86104 Subscription Rate (messages / sec) Tibco EMS 4.4. All Rights Reserved .2 16137 29239 33180 32224 Jboss 1.2.

4.0 14341 12472 12444 10278 Sonic MQ 7.3 16926 17984 12860 569 Subscriber Type Size (bytes) 1/1/1 10/10/10 25/25/25 50/50/50 Non-Durable Non-Durable Non-Durable Non-Durable 1024 1024 1024 1024 Copyright © Fiorano Software and Affiliates.4.2 Server Scalability Message P/S/T Message Type NonPersistent NonPersistent NonPersistent NonPersistent Subscription Rate (messages / sec) Fiorano MQ 9.2 16137 21672 21412 21049 Jboss 1.3.1 45262 32587 30735 29192 Tibco EMS 4. GA 16448 14867 14622 16001 Sun MQ 4. All Rights Reserved .0.2.0 12246 10261 10322 7239 Active MQ 5.

All Rights Reserved .2 1990 4512 4680 4756 Jboss 690 9470 11671 9121 Active MQ 5. Durable Subscribers Message P/S/T Message Type Subscriber Type Size (bytes) 1/1/1 1/10/1 1/25/1 1/50/1 Persistent Persistent Persistent Persistent Durable Durable Durable Durable 1024 1024 1024 1024 Fiorano MQ 9.1 3321 21647 37353 14653 Subscription Rate (messages / sec) Tibco EMS 4. GA 2419 6363 7093 7000 Sun MQ 4.4.3 Persistent Publisher.0 985 8708 12215 10424 Sonic MQ 7.3 2786 6946 9714 805 Copyright © Fiorano Software and Affiliates.0.

GA 16448 35596 38479 37344 Sun MQ 4.3. All Rights Reserved .0 14341 25329 26219 22128 Sonic MQ 7.0.3 16926 17027 12239 696 1/1/1 1/10/1 1/25/1 1/50/1 NonPersistent NonPersistent NonPersistent NonPersistent Non-Durable Non-Durable Non-Durable Non-Durable 1024 1024 1024 1024 Copyright © Fiorano Software and Affiliates.4.4.0 12246 23103 24348 19223 Active MQ 5. Non-Durable Subscribers Message P/S/T Message Type Subscriber Type Size (bytes) Fiorano MQ 9.2 16137 29565 34685 33387 Jboss 1.4 Non-Persistent Publisher.1 45262 82187 88193 84398 Subscription Rate (messages / sec) Tibco EMS 4.2.

Lockheed Martin.4. POSCO.0 (In persistent tests.fiorano. Fiorano's network-centric solutions set a new paradigm in ROI.GA  Sun JAVA MQ v 4.1 Hardware Configuration  Server System  GNU/Linux 2. System Configuration 3.0. performance. the TIBCO topics were set to failsafe to ensure persistence to disk)  ActiveMQ v 5. Credit Agricole Titres.2 Software Configuration JavaTM 2 Runtime Environment. British Telecom.00GHz  64 bit 8 GB RAM  Client System  Microsoft Windows Server 2003 R2  4 CPU Intel(R) Xeon(R) CPU 5160 @ 3.00GHz  64 bit 8 GB RAM  Network Settings  Client and Server were on the same network.  Network Speed: 1GBPS. All Rights Reserved . interoperability and scalability. Schlumberger and Vodafone among others have used Fiorano technology to deploy their enterprise nervous systems.3. 3.6.0 Tibco EMS v 4. Standard Edition (build 1.5.2  Jboss 1.4.3     About Fiorano Software Fiorano Software (www. NASA. Global leaders including Fortune 500 companies such as Boeing. Copyright © Fiorano Software and Affiliates.0_05-b05) Fiorano MQ v 9.1 Sonic MQ v is a leading provider of enterprise class business process integration and messaging infrastructure technology. Qwest Communications.18-92.el5xen  4 CPU Intel(R) Xeon(R) CPU 5160 @ 3.3.