Track 4: java ee, Technical sessions | track four | java ee, Java ee – Google 2007 JavaOne Advance Conference Guide User Manual

Page 28

| technical sessions | track four : java EE | java.sun.com/javaone

* Content subject to change.

TECHNICAL SESSIoNS

| TRACK FOUR | JAVA EE

Java EE

TS-1419 Best oSGi Practices

BJ Hargrave, IBM
Peter Kriens, aQute
Thomas Watson, IBM

The OSGi specifications have seen widespread adoption in the industry
over the last few years. Adoption in enterprise software projects, JSR 232
for the mobile phone industry, the explosion of the number of Eclipse
users and programmers, the immense interest in the Spring OSGi project,
and the long trail of OSGi use in industrial automation and embedded
computing have firmly placed OSGi technology on the map. However, as
with any new technology, there is a period in which practitioners struggle
with how to best use the technology. Existing patterns clearly work in the
OSGi environment, but the best results are achieved when patterns are
used that work most effectively on an OSGi platform.

One consideration with OSGi design is how to decompose the system into
separate modules called bundles. Although applications can be written as
monoliths and deployed as monoliths, more advantages can be obtained
when the system is decomposed into smaller, modularized parts. Properly
applying this model allows bundles to be reusable, and it allows the reuse
of third-party bundles. Also, most systems today contain large chunks
of code from previous projects or from external providers. This code will
require bundlization to cooperate. This presentation shows you how to
bundlize legacy code and warns you about potential pitfalls.

Bundle decomposition is closely related to how to use services. OSGi
services provide a service-oriented model for interbundle collaboration
and are highly dynamic, with varying cardinalities. The whiteboard
pattern is one pattern that is very powerful and related to the inversion
of control (IoC) pattern, which is highly popular today. Additional use
patterns are also presented.

The session helps practitioners understand the advantages of OSGi
technology and properly apply it to their projects. The speakers are
people who have been using OSGi since 1998, who have heavily influenced
the specification, and would like to share their experiences with you.

TS-1743 xen and the Art of distributed virtual Machine

Management

Greg Lavender, Dept. of Computer Sciences, The University of Texas
at Austin

There is a lot of interest in virtual machine software for virtualizing
multicore hardware resources. This session looks at Xen, a hypervisor
platform that supports multiple virtual machine instantiations on the x86
and x64 hardware platforms. The Department of Computer Sciences at
the University of Texas at Austin designed a Java Message queue software
agent-based system that allows for the dynamic creation, instantiation,
monitoring, management, and automigration of Xen virtual machines
running on a rack of Sun x64 Opteron systems. The agents monitor
various system resources and can automatically determine opportunistic
strategies for load-balancing virtual machines (VMs) across multicore
CPUs and physical machines on a gigabit-switched VLAN. Java Message
queue software is used to establish multicast message channels for agent
communication as well as real-time monitoring messages sent to a Java
technology-based Standard Widget Toolkit (SWT) graphical display.

JAvA EE

Java Platform, Enterprise Edition (Java EE) is
the industry standard for developing portable,
robust, scalable, and secure server-side Java
technology-based applications. Building on
the solid foundation of the Java SE platform,
Java EE offers standard APIs for web services,
persistence, security, management, deployment,
and communications that make it the industry
standard for implementing enterprise-class
service-oriented architecture (SOA) and Web 2.0
applications.

This track includes technical sessions and tutorials
on topics such as the following:

• Present and future APIs, standards, and

specifications

• Enterprise data modeling, persistence,

databases

• Quality of service: security, manageability,

availability, scalability, reliability, portability,
and performance

• Web services and interoperability with other

platforms

• Portal and telephony technologies
• Open source innovation involving the platform
• Pragmatic testing and debugging techniques
• Case studies on applying the platform to solve

real-world problems

• Cool stuff that shows innovation on the platform