Architectural Influence on Java Enterprise Application's Performance

摘要

The influence of architecture is immense on the performance of Enterprise Applications. A very careful analysis and design is required when we discuss Java Enterprise applications because of the unique memory model of JVM and the requirements for mission critical applications. Java has proved its wide acceptance by its efficient implementation of object oriented programming concepts and its unique memory management model. Also being a platform independent language cuts down the cost of software development and deployment to a drastic level. As the enterprise applications are meant to be working 24*7, under a variety of stress conditions (peak and off-peak times), and are used for mission critical operations, the performance benchmarks must be very carefully defined and analyzed. This paper presents different aspects of architecture. This paper presents techniques for delivering high performance applications to production, managing and measuring the performance of applications, and diagnosing the toughest Java enterprise problems throughout the entire application lifecycle. There are various factors that impact the overall performance and scalability of the system. These factors include, but are not limited to application design decisions, style and efficiency of user written application code, system topology, database configuration and tuning, disk and network input/output activity, operating system configuration, and application server resource throttling knobs. Different component of Java Virtual Machine (JVM) technology, such as new memory management strategy and Garbage Collection (GC) are discussed in order to explain how an application is influenced by the JVM architecture. This paper recommends techniques for selecting performance benchmarks to achieve the optimized output from the J2EE application. These benchmarks must be quantifiable, representing and must be reproducible, in order to get the most realistic representation of the system. These benchmarks include load size, throughput, transaction time, connection time, response time, processing time, waiting time (average latency) and receiving Time.