Manage Learn to apply best practices and optimize your operations.

Java EE platform changes explained by Ashesh Badani

Red Hat and JBoss made several announcements at the combined Red Hat Summit / JBoss World event in Boston earlier this month. The first developer preview of the Java EE 6 platform has been released, as well as two cloud offerings—OpenShift and CloudForms. The former is a platform as a service (PaaS) offering and the second provides infrastructure as a service (IaaS). The senior director of middleware for JBoss, Ashesh Badani, explains which announcements are most exciting for him, and gives us a look into the Java specification requests (JSRs) that are driving the changes in the upcoming edition of the Java platform.

Read the full text transcript from this video below. Please note the full transcript is for reference only and may include limited inaccuracies. To suggest a transcript correction, contact   

Java EE platform changes explained by Ashesh Badani

James Denman: Hi this is James Denman with I am here with Ashesh
Badani. He is the senior director of Middleware products for JBoss and Red
Hat. Thank you for speaking with us today.

Ashesh Badani: Sure.

James Denman: Can you talk about the new announcements, the new products that are
coming out?

Ashesh Badani: OK. So, excitingly for us, we made a couple of really big
announcements for the JBoss Middleware Portfolio. The first is our early
access for Enterprise Application Platform 6, which is based Java E6
specification and it's a big release for a variety of reasons. First off
is the fact that it's our flagship product line. Many folks know JBoss, for
it's, application servers, application platforms, we have a full portfolio
of offerings. This is the one that many folks are very familiar with. It's
also leveraging the entire Java E6 specification, which was pretty fundamental
and a large evolution of the Java E platform, for a variety of reasons.
Number one, there's a big focus on ease of use, extensibility, making the
platform more lightweight and more flexible. So what we are seeing with
Java E6 is a web profile, if there are customers who don't want to use a
full Java E6 specification, there's a way to be able to use a subset of it,
called a red profile which allows for more web facing types of
applications, and it may have some constraints from a deployment
perspective. So we're happy to support that view and provide that for our
customers. That's number one.

Number two; there were a couple of very significant GSR Java Specification
requests, that's accepted as part of this specification, both of which Red
Hat and JBoss led from an expert group perspective. So, we expect leads, from
a JSR 299 context and dependency injection. And what this did was,
really help to simplify and unify some program models around EJB, in Java
server bases, and job courses in architecture. It made it much easier
for developers to use annotation based dependency injection. In the past,
there was a lot of messing around with the XML configuration files. As a
result we had some non-standard frameworks like Spring Top Cloud, and now
we've taken a lot of the good work that Spring and many others have done
and are able to standardize that within the Java E6 specification. So that's
been a big move forward.

Also, specification JSR 303: Bean Validation, again, led by a member of the
JBoss engineering team, allows for a much more standard way to do
validation logic, essentially across the life cycle of the application,
greatly simplifying the work of developers. So, some fairly big improvements
from a Java E6 overall evolution perspective, so we're happy to support
that. There were also enhancements to the JSR, the Java Server Faces 2, was
introduced as part of this, EJBV.1, the ability to be able to deploy much
more easily instead of having the EJB files, the archive files. You can
instead deploy those in more simple WAR files. So, from an overall
usability perspective, the specification has moved forward, which will
also then, for our own product line, implement those advancements but also
focus pretty heavily on what we call OA&M, operations, administration and
management capabilities, making it much easier to use our platform, provide
an imbedded management console in addition to JBoss, operations network, that monitors
the entire application platform as well as the portfolio.

To be able to have a new domain model, it makes
it easier for administrators to manage the product line, making significant
enhancements from a performance perspective by a new implementation of
GMS. That actually leads me to come to the announcement, because that
helps increase performance over the base cloud form. That's the
announcement of Enterprise Data Group 6. That's based on the open source
community project called Infinite Span. But really what it is, is a
large scale distributed data caching technology. And we've heard a lot of
talk for the need for elastic application platforms, in cloud environments.
And with Infinite Span what we can actually do is work with very highly distributed data
environment, leveraging some of the core advances between the no SQL movement
and some of the work being done around big data to better understand large data sets,
and be ready for cloud computing. But with cloud computing we're
seeing greater demands from a transaction perspective, interaction
perspective, lots of different applications, mobile application requiring
more from the application platforms. An alternative for scaling up or
scaling out is a very difficult issue. What's happening in many cases,
difficult achieve those goals, is the economic challenges. They may
increase the cost from a database perspective, a hardware perspective. But
of course, all the associated management overhead that goes with it. But
also challenges from a technical perspective, being able to cluster, being able to
provision additional instances that come online. With Infinite
Span, are the data grade technologies that we have, what you can do is have
an in memory, persistent data store. It's not like its transient, so there's
some persistence that's happening inside the memory layer, to really scale out the data
tier, without having to constantly go back to additional databases. And
what it also allows you to do is take advantage of existing architecture.
That's all the folks using Java technologies are familiar with many of the
concepts that have led us to Infinite Span, like the patience in the work
we've done in ncaching and with jcashe, we’ve advanced that considerably
by taking advantage of some ideas that developed for Amazon, Dynamo
and other ideas in the industry, for high speed and high-performance
scaling. And by taking advantage of existing infrastructure and being able
to leverage, what we're trying, not just developing with Java, but
also with additional languages and frameworks like Ruby, or PHP or
Python and we can make it a lot more accessible to the folks out there.
There are 9 million Java developers out there today, and we believe this
technology is appropriate for them to start using, and that's why we
submitted it as a Java specification request, as far as the Java community

View All Videos

Start the conversation

Send me notifications when other members comment.

Please create a username to comment.