fredag, augusti 22, 2008

5 Top Products we Acquired from BEA

Having spent the last weeks looking into the products that we got added to our product stack with the acquisition of BEA, I now feel that I'm starting to get a grip on them and can see how they will fit into our product portfolio. So far some of the products feels very strong and I will elaborate a bit on why I think so in this post. So, I'd like to rank the 5 strongest products, in my opinion, that we got from BEA.


#5 Oracle BPM aka AL BPM

This product will certainly fill a gap in our portfolio for creating business process. It will enable you to quickly & efficiently create business processes. It will also assist with the monitoring of them according to KPI, which will assist in improving the processes according to business needs. It does in no way replace our other products for creating processes, like BPEL, for example. Rather it complements it in a very good way.

Sure, there is a danger here that users starts to generate & implement processes on their own, similar to what we see in the database area where users stores business data in spreadsheets or small databases outside of their main databases. One could perhaps compare this risk of generating "process islands" similar to the risk of generating "information islands" for data storage. However, as we will not be able to prevent users from storing data outside of the main databases we will likewise not be able to stop them from creating processes outside of the main SOA architecture, thus we need to be able to provide a good tool for this purpose and this is BPM.


#4 WebLogic Server

You might be a bit surprised that I put WebLogic only as number four. The main reason is that WebLogic plays in an area where I think that we already have a very strong product. For sure, WebLogic is a darn good Application Server too, but we got other products from BEA that imho better fills gaps in our product portfolio, thus WebLogic only ends up in spot four.

The two main things like about WebLogic, and where I think that Oracle AS could be better, is in administration and JMS. The administration tools for WebLogic I find superior to the ones we provide with Oracle AS. The concept of the administration server, which you use to push configuration out to the managed servers is very nice, and will ensure that all servers in a cluster has the same configuration. This was not so easy to do in Oracle AS. Also I find the JMS implementation in WebLogic superior to the one we have in Oracle AS, however, I will not go into the details in this post about this.

There are a few things that I like in Oracle AS, that WebLogic don't have, for example that ability to run the OC4J container standalone just to mention one.


#3 AL Data Services Platform

Also perhaps a surprise. In this case it's not so much the product itself, it's rather the area that it's targeting that is interesting. As of today the common way of writing software is in a three-tier architecture: GUI-Middleware-Database. I believe that this will change in the near future and that we will see that four-tier architectures will become more and more common where we introduce a data & computing grid between the middleware and the database, so that we will look into architectures that looks like: GUI-Middleware-Grid-Database.

We already have a few products in our product stack that targets this domain, like ODI and Coherence. But we still miss a few pieces to be able to provide a complete stack of products for this domain and here is where the functionality from the Data Services Platform comes into the picture. WebLogic perhaps provides a better choice for an existing product, while DSP provides us with functions that will be crucial for the next big thing. That is why DSP is higher on my list.


#2 Tuxedo

Before the middleware era, C, C++ and Cobol were the common languages for accessing the Oracle database. When Oracle entered the middleware area it was all Java, Java and some more Java. Sure, we provide modules for Pearl & PHP, but Java is THE middleware language in the Oracle stack. I guess the C & Cobol developers may have felt a bit stepmotherly treated along the road. That is, until now.

Tuxedo can be described as a middleware platform for C, C++ and Cobol. Sure, developers for these languages have previously had the option to license Tuxedo from BEA, but to have it in-house is imho a big bonus for Oracle. This will hopefully give the large group of people developing in C, C++ and Cobol the message that they are not forgotten by Oracle. It will also show that Java is no longer the only option that we provide for developing middleware.


#1 JRockit

Before the buy of BEA we could provide all pieces of software, from the operating system to the developer IDE, except for one piece; a JDK. With BEA we now have JRockit, however, this is not the main reason why it is on the number spot on my list.

The Liquid VM version of JRockit will enable you to run Java applications directly on top of a virtual machine without any operating system. This is a very cool feature that will boost many Java applications. Still, this is neither the reason why JRockit ends up in the number one spot.

The reason I have JRockit on place number one is the Real Time version of JRockit. This will enable users to run their Java applications with minimal and deterministic interruption for garbage collection. This is a very crucial feature for existing Java applications that needs to be highly performing, but even more important will the combination of JRockit RT and Coherence be. This combination will enable Java developers to write the next generation of Java applications that will perform on the same level as C / C++ applications. This is something that has been missing in the Java world for a long, long time. Hence, JRockit ends up in my number one spot.



Inga kommentarer:

Skicka en kommentar