Friday Aug 30, 2013

Introducing Elastic JMS

In WebLogic 12.1.2, we enhanced the way that you can configure JMS servers, stores, and subdeployments so that the JMS subsystem can automatically scale with the Managed Servers in a cluster. We call this Elastic JMS. My friend Maciej Gruszka calls it Magic JMS!

 Here are some details:

JMS Servers: In releases before WebLogic Server 12.1.2, each JMS Server was individually configured and targeted at a single Managed Server. It didn’t matter whether or not that Managed Server was part of a cluster. Starting in WebLogic Server 12.1.2, you can target a JMS Server at a cluster. Under the covers, WebLogic spins up a JMS Server on each managed server in the cluster. If you add or remove servers from the cluster, JMS Servers are added or removed automatically.

WebLogic Persistent Stores: Like JMS Servers, in releases before WebLogic Server 12.1.2, each WebLogic Persistent Store (file store or JDBC store) was individually configured and targeted to a single Managed Server, clustered or not. In WebLogic Server 12.1.2, you can target a WebLogic Persistent Store at a cluster. Under the covers, WebLogic creates a store instance on each Managed Server in the cluster. Each instance of a file store uses the same path to either a shared file system or to a local file. Each instance of a JDBC store uses the same JDBC data source, but gets its own underlying tables.

Subdeployments: A subdeployment defines the list of JMS Servers that will host a queue or topic. In releases before WebLogic Server 12.1.2, when you defined a subdeployment for a distributed queue or topic, you listed each JMS Server in the cluster. When you scaled up the cluster by adding a Managed Server and a corresponding JMS Server, you also needed to update the subdeployment with the new JMS Server. Starting in WebLogic Server 12.1.2, subdeployments are much simpler. You can list a single JMS Server that is targeted at the cluster. When you scale up the cluster, the distributed queue is automatically extended to the new JMS Server instance without any changes to the subdeployment.

Pulling it all together: By using cluster targeted JMS Servers and Persistent Stores, you get some nice benefits:

  • Simplified configuration – Even initial JMS configuration is much simpler than it was in the past: no need for individually configured JMS Servers and related items.
  • Elastic scalability – As you scale the cluster, the JMS services automatically scale with it. 
  • Support for Dynamic Clusters – Because Dynamic Clusters require homogenous targeting of services, the new configuration options make it possible to run JMS on Dynamic Clusters. 

  • Check out the documentation at http://docs.oracle.com/middleware/1212/wls/JMSAD/dynamic_messaging.htm or see my video at for more details.

    WebLogic Active GridLink for RAC: Check out NEC results!

    Oracle WebLogic Active GridLink for RAC is the market-leading mid-tier integration solution leveraging additional Oracle RAC advancements. Oracle and NEC have jointly invested on verifying and testing the Active GridLink for RAC solution which provides a high-availability, scalability and high performance solution for helping customer building mission critical applications. In addition, WebLogic Server is the only application server which has been fully integrated and certified with Oracle Database without losing any capabilities in Java EE implementation with respect to security, transaction, connection pooling , management, and more.

    NEC has constructed high availability systems with WebLogic Server and Oracle Real Application Cluster. It investigated advantages and features of the GridLink data source focusing on high availability and flexibility that GridLink data source achieves in integrating with Oracle RAC. The purpose of these tests is to get high serviceability by diverting system integration know-how of JDBC data source already storing, in addition to using these functions. The following are functions which the GridLink data sources were tested for.

    1. Runtime Connection Load Balancing(RCLB)

    2. Web Session Affinity

    3. Fast Connection Failover (FCF)

    4. Performance Improvement by Dynamic Changes in RAC Topology

    Join us in this OpenWorld session as we discuss the detailed Active GridLink capabilities and the testing results obtained by NEC. We will also discuss the new functionality introduced with Oracle Database 12c.

    About

    The official blog for Oracle WebLogic Server fans and followers!

    Stay Connected

    Search

    Archives
    « August 2013 »
    SunMonTueWedThuFriSat
        
    1
    2
    3
    4
    5
    6
    7
    8
    10
    14
    15
    16
    17
    18
    20
    21
    22
    24
    25
    26
    27
    28
    29
    31
           
    Today