Java Performance Management for Large-Scale Systems

There are many classes of enterprise applications that have stringent performance and scalability requirements. TheServerSide.com has assembled a collection of resources to help you better design, develop, test and manage high performance, large-scale systems - learn new and innovative approaches for performance tuning, memory management, concurrent programming, JVM clustering and more.

Format:
mp3

Extreme Transaction Processing, Low Latency and Performance

Speaker: John Davies, Technical Director and Head of Research, IONA

In this podcast, John, who has over 30 years of experience in investment banking and integration technology, John will cover several case studies of extreme transaction processing, low latency and high performance systems and offer insight into what we might expect to see in mainstream in the near future.

Format:
mp3

Java Performance Tooling

Speaker: Dr. Holly Cummins, Tool Developer, IBM Java Technology Centre

No one writing Java applications should be without an understanding of how to fix things when they go wrong. Diagnosing performance problems can be difficult; sometimes even knowing where to start is hard. In this podcast, Holly will introduce a number of tools for identifying and fixing common Java performance problems.

Introduction to the Real Time Performance Monitor (RTPM)

Speaker: Azul

The Azul REALTime Performance Monitor (RTPM) is highlighted in this 12-minute introductory presentation and demonstration. Take a virtual tour of the tool that lets you analyze your application's thread stacks, CPU, I/O, GC and memory usage on-the-fly in both development and production.

Real Time Performance Monitor

Speaker: Cliff Click, Azul's Chief JVM Architect

Cliff Click, Azul's Chief JVM Architect, gives a hands-on demonstration of how to use the REALTime Performance Monitor (RTPM) tool to do performance and bottleneck analysis. He examines the popular Apache open-source struts2 framework under load, and uses RTPM to pinpoint a hot lock preventing efficient multicore scaling. After modifying the source code fingered by RTPM, he demonstrates a dramatic performance imporovement.

Using RTPM to Find an I/O Performance Bottleneck in Apache Struts 1

Speaker: Azul Systems

The Apache struts1 package is analyzed under load in this demonstration of Azul's REALTime Performance Monitor (RTPM) tool. Using the profiling tool's abiility to measure system call rates, we find and fix an I/O-related bottleneck.

Compute Pool Manager

Speaker: Azul Systems

Azul's Compute Pool Manager (CPM) offers both application owners and IT operators a window into application behavior. See how you can use the perfbar application built in to CPM to watch CPU core activity in real-time.

Format:
On24

Kirk Pepperdine, Holly Cummins, John Davies and Gil Tene

Topic: Garbage Collection: Reducing Pauses for Greater Java Performance

This TechTalk details how the pauseless garbage collection is now being utilized to overcome such performance challenges. Discover various methods of garbage collection, how it is being improved and effectively managed. Explore how pauses are being removed from garbage collection, improving Java application throughput and performance. Learn how problems such as compaction and marking are being overcome in garbage collection and gain a comprehensive understanding of the key factors impacting pause time in collectors.

Format:
On24

Gil Tene, Vice President of Technology and CTO, Co-Founder, Azul Systems

Topic: Optimizing Java Application Performance with Compute Appliances

This Techtalk discusses how compute appliances can be used to optimize the performance for demanding Java applications. Learn how these appliances can be used to power Java Virtual Machines and existing servers, delivering greater CPU and memory capacity. Discover how to allow enterprise Java applications to transparently deliver greater scalability, throughput and consistent response times without changes to the existing code.

Format:
mp3 and slides

Gregg Sporar, Technical Evangelist on the NetBeans project, Sun Microsystems

Topic: Memory Leaks in Java Technology - Based Applications: Different Tools for Different Types of Leaks

In this presentation, recorded at TSSJS 2007, Gregg examines some of the tools and techniques available for detecting memory leaks and uses from real-world Java applications. Gregg shows how to track down a memory leak where multiple object instances of the same class are created over time, some of which are the source of a leak and others of which are not. Gregg also examines how to debug a memory leak where only a single object instance is the source of the problem. Each of these examples involves a brief examination of the source code and a demonstration using a monitoring/profiling tool. Slides for this presentation can be viewed here.

Format:
mp3 and text

Dr. Holly Cummins

Topic: Holly Cummins on Java Garbage Collection and Improving Application Performance

If you think garbage collection is a bad thing, you haven't heard Dr. Holly Cummins, noted IBM garbage collection tuning specialist, who gives her opinion on the subject. In this podcast interview, she discusses why garbage collection is good and a number of other topics all to do with Java heap space, garbage collection and other matters that can be seen a activities that degrade application performance and throughput.

In this interview, Brian Goetz and Cliff Click detail the best practices necessary for creating highly concurrent software. They discuss the meaning and purpose of concurrent software, threads and parallel computing, describe how parallel computing affects the design of programs and examine the challenges surrounding writing acceptable parallel systems. John Davies and Gil Tene sat down with Kirk Pepperdine to describe Azul's multi-processor hardware from a consumer's standpoint, especially with respect to SOA.

Format:
mp3 and text

Cliff Click

Topic: Where the metal meets the JVM

While at TSSJS Barcelona, Cliff Click from Azul systems spoke with Kirk Pepperdine about his work at Azul, his involvement with the -server compiler team, and a whole host of hardware issues as they relate to the Java memory model, thread safety, and program correctness.

In this interview, Gil Tene talks about Azul's high capacity boxes, which run from 192 to 768 cores (with up to 750GB of memory) and provide that power to other systems as a Java "compute appliance." Gil also describes some of the problems applications might find in using such massive compute appliances, and how Azul works around it. Nati Shalom of Gigaspaces, Bob Lozano of Appistry, Gil Tene of Azul, John Davies of IONA and Kirk Pepperdine of TSS discuss high performance computing on a panel at TheServerSide Symposium 2007. Azul Systems is at the forefront of building large-scale, long running applications in Java. In this tech talk, given at JavaOne 2006 in San Francisco, Vice President and Deputy CTO Bob Pasker describes how Azul overcomes obstacles such as the locking problem in Java through consolidation and application behavior.

TSSJS Show Coverage - Gil Tene on Pauseless Garbage Collection

By Frank Cohen

Frank Cohen interviewed Gil Tene at JavaOne on Azul's pauseless garbage collection feature. According to Tene, a recent performance benchmark of the product yielded 35 Gigabytes/second of continuous hashmap allocation rates, with no garbage collection pause.