In today’s competitive market, choosing the right platform to run your mission-critical applications is a key to success. For my employer, the platform of choice for new Java development became WebSphere Application Server on z/OS V6.1 (WASz). We’re a large corporation, with high-volume transaction processing and high-availability requirements. In the past two years, we’ve implemented a solid, scalable, high-availability WebSphere on z/OS infrastructure that satisfies our data integrity, system performance, and system availability objectives.
WASz addresses our need to run very high-volume, business-critical Java and Web services applications. We architected and established a “best practice” WebSphere on z/OS implementation using a Network Deployment world-class cluster configuration, crossing LPARs, with proven failover capabilities. We used WebSphere MQSeries Queue Sharing, DB2 Data Sharing, and SYSPlex Distributor to maximize system availability. This robust, scalable design allows us to quickly adapt to new business requirements and growth.
Having a solid infrastructure in place helps us maintain high availability. However, from time to time, we also have applications that don’t always follow proper coding practices and sometimes cause problems. We proactively address application issues using several free tools from IBM:
• IBM Support Assistant (ISA)
• First Failure Data Capture (FFDC) logs
• Tivoli Performance Viewer (TPV)
• IBM Pattern Modeling and Analysis Tool for Java Garbage Collector (PMAT)
• The Svcdump.jar utility
• Interactive Problem Control System (IPCS)
• WebSphere z/OS display commands.