Global Trends: Unified SOA Performance Management Matters

A comprehensive solution is needed for managing all of the different complexities involved with software-oriented architecture (SOA) environments.

SOA application performance is unsustainable without end-to-end visibility into the entire transaction path. This means 24/7 awareness of all integration points, services, components, and connected back-end systems. To ensure performance levels that satisfy customers, IT operations staff needs to know about SOA applications’ linkages to other services, applications, and tools. They need real-time metrics like number of concurrent users, throughput, processing time, messages per hour, queries per day, latency, availability, rejected transaction counts.
Such monitoring must also be low-profile. Highly available SOA apps cannot afford the extra overhead of monitoring code. In this white paper, you’ll learn that these sorts of requirements mean that monitoring, measurement, and management of an SOA application environment is best handled with a single performance management toolset that provides

• Automatic discovery of services/components
• Always-on performance and availability monitoring in production
• An ability to monitor all transactions at the component level, 24/7, across multiple processes and applications
• An ability to monitor HTTP traffic, portals to track customer experience
• Root-cause analysis that maps and models configurations, relationships, and interactions
• An ability to establish performance baselines and leading indicators, so problems are identified
• Early on Integration of diverse management capabilities