Page Visitors: 267
As promised this is my second part of my blog. The first part of the blog can be accessed from here.
As part writing this blog i browsed through various blogs on the same topic. As i told in my first blog, this really is a holy war and over coming few months and years, this will rage even harder.
I bust out into laugh after reading one of the blogs, the link for which is as given below:-
The point that i burst outlaughing when the author stated as below:-
…… Java EE 5 took a few pages out of Spring’s book and simplified the development of enterprise components with…….
Yes, JEE sure did take some very crucial and important pages out of Spring book and because of which it is becoming superior day-by-day if not by huge margin. Spring with full support from its community is not far behind and with foundation in place, its building up components one after another. Its also adding in more and more support for open source libraries with ever growing sub-projects under the Spring Framework umbrella.
Also now its is slowly becoming a norm when considering a new open source for development to see how strongly it can be integrated into Spring and how seamless Spring configurations can be used. In my web service framework comparison, you wouldn’t believe, this was considered as one of the requirements and it will be a bonus for any open source to state that “they can be integrated into Spring” with some few lines of Java or XML configuration.
Spring has an edge by keeping it open for good open source libraries to plug on at any time and replace these libraries with a new one by mere configurations (at-least theoretically possible… :)). Java theoretically is “Write once run anywhere”, but is it truly that easy for a a production application to move from one environment to a new one… Hmmm… I don’t think so… 🙂
I am sort of exhausted already writing this blog. I will have to put you all in some suspense and put everything else that i plan to write in a new blog part. Hold on tight and i am sure i wouldn’t make you unhappy. Expect third part in another 3 days time. Its weekend here and i am really looking forward to it in a big way.
Page Visitors: 378
Am I playing with fire here. Perhaps Yes.
As many say, this truly classify itself to be a holy war for predominance and will surely stir some dirty stuff out into the air.
One of my colleagues recently came to me and had a small debate as to why the our organization selected Spring over JEE. He was very much in favor of JEE especially JEE 6.
We sometime back (more than an year) had an exercise in which the task was to choose organization wide Java framework. As you guessed right, we did have Spring Framework and JEE as strong candidates. Our approach was really methodical and a methodology which was devised internal to the organization was used.
In this methodology the two candidates was put into test. First of all, various architecture quality attributes were put in such as Performance, Scalability, Testability etc. and then various requirements in these attributes were classified/grouped together. Once we have that, we get the various stakeholders involved and get their buy in and their stand on the overall acceptance of these quality attributes and rank it. We then went on with POC (Proof of Concept) and each requirement was rated against each of the chosen frameworks.
Finally after a grueling exercise of rating various requirement for each of the framework, a final rating comes out which give good indication as to which on quality attributes gets favored for each of the framework and according to the stakeholders, which one to choose out of these two. The choice was made and there was no surprise, it was Spring Framework.
For our company now the chosen default framework is Spring and anything else if it is taken, it goes as an exception.
For me when i say that i use Spring Framework, it doesn’t in any way mean that i don’t use JEE. They cannot be mutually exclusive, rather when combined together, it gives immense advantage to the project that you are developing. We do use Spring as our choice of framework but haven’t just taken out JEE in any sense out of the window, rather when the right time comes another evaluation as detailed above will be done to asses the situation even further. Till that time i feel choosing Spring as the default framework is a good one in all aspects.
In my next blog (Click here), I would try to give some of the major points, which made us choose Spring as a default framework of choice for developing any application within the organization.
Earlier i have done couple of comparisons as detailed below:-
Page Visitors: 1029
The Blog is still under construction but i would like to publish it and then keep on updating with more content. Apologies for incomplete blog post.
The project is a merger between two open source projects namely Celtix developed by IONA technologies and XFire developed by Codehaus. As you would have already figure why the name CXF – Celtix XFire. Rather than me plagiarizing, I would suggest reading the Wikipedia for more details http://en.wikipedia.org/wiki/Apache_CXF.
Some of the advantages associated with Apache CXF is as follows:-
- Good cohesive support for the Spring Framework.
- Good OSGi support (How much this is a valid constraint for the evaluation is questionable, i agree).
- Good support for more strong web service functionalities like WS-Security, WS-Policy etc. to name a few.
- It also has a compliant JAX-RS implementation allowing it to expose web service as REST and SOAP based.
- It also has W3C compliant SOAP/JMS implementation if they are required.
- Much easier to use and implement.
- Less verbose when it generates basic classes for WS/WS-Client from the WSDL.
- More “standards based” API’s. The API’s are slim and trim with good performance and ease of use.
- Considered more “embeddable” as opposed to other WS frameworks, especially with the Spring Framework.
- CXF provides support for code first and contract first development, and provides ability to plug-in various flexible binding mechanisms.
- The front end programming API’s is spectated from the core code making it really flexible.
Some of the disadvantages associated with Apache CXF is as follows:-
- Doesn’t support Orchestration and WS transactions yet (I really don’t know how relevant this is, i plagiarized it from Ankur Kumar blog, refer reference section of this blog post. As he is well-versed with WS, i think i should believe him. Might be Daniel Kulp will help me with this.)
- WSDL 2.0 support is not yet there.
- Again, CXF is very good with SOAP, so for web services based on REST, due to its very good support for SOAP, CXF brings in added weight and complexity (I am expecting Daniel Kulp to help on this as well)
As explained earlier is a completely re-written version of the famous Apache Axis SOAP stack. While doing so it has to some extend looked upon various lessons learnt and corrected it. Some of the advantages of using Axis2 are as detailed below:-
- Good support for more strong web service functionalities like WS-Security, WS-Policy etc. to name a few.
- Good integration support for binding frameworks like XMLBeans, making it suitable for creating web services having complex schema definitions.
- Somewhat better integration support with the Spring Framework. More details can be found at http://axis.apache.org/axis2/java/core/docs/spring.html.
- Considering performance aspects, it performs a bit faster compared to Apache Axis2 due to use of proprietary API’s.
- Supports bothe Contract-First and Contract-Last approach, which gives good flexibility and choice to the users. This is key to some extend.
Some of the disadvantages attributed to Axis2 are as detailed below:-
- Lack of cohesive support with JAX-WS. the support is detailed in the following URL: http://axis.apache.org/axis2/java/core/docs/jaxws-guide.html.
- Again good cohesive support between Axis2 and Spring Framework as opposed to other WS frameworks.
- Good amount of classes, a bit hard to understand and difficult to refactor in case if it is required.
- Churns out more proprietary API’s and you as a developer would have to handle these.
- Quick turnaround with fixes and releases. You could read Part I of this blog which compared these aspects in detail.
- Slowly Apache is phasing out Axis2 in favour of Apache CXF (Please dont quote me on this, but this is slowly happening)
- It’s a web application by its own with all the services deployed in WEB-INF/services folder as AAR (Axis 2 Archives). This structure as a whole can be problematic while deploying, maintaining, upgrading, versioning and testing to a greater extend.
Some of the advantages associated with Spring WS is as follows:-
- Full support of the Spring programming model and this can be a very good advantage to take this framework on. The Spring Framework is leading the Java web framework race, now almost single handedly and this advantage is key according to me.
- Clean API’s (same as above because Spring having clean API’s has influenced this to a larger extend)
- Clean mapping of the incoming SOAP messages to the methods
- If your web application framework is based on Spring MVC, Spring WS gives easy and clean mixing between WS endpoints and Spring MVC’s controllers. Also coding of endpoints is as easy as coding a new controller in Spring MVC.
- Easy unit/integration testing capabilities.
- Support for different XML marshallers.
- No generated code/deployment descriptors.
Some of the disadvantages associated with Spring WS is as follows:-
- Supports only Contract-First approach, which could result in lack of flexibility to the users.
- Most of WS-* support might be questionable. By the way how much of WS-* things we use?
Technologies grow faster and with open community development, issues identified gets fixed quickly. Having said that, there is an article which i would like to share; http://www.ibm.com/developerworks/java/library/j-jws19/index.html; which rates Apache CXF more as compared with Apache Axis2 in security aspects. Many issues identified would have got corrected over the period of 3 years, but its worth noting these.
Although my comparison is not that well structured, i found a blog released on the same day as i was typing mine, which details the set of criteria which should be used for evaluating the web service frameworks. If i have to do this extensive comparison, it would take me at least 3 months. Since i don’t have much time at this point, i still don’t want the reader not to see these evaluation criteria, which is well structured and though through:
Things which I think (Purely my own)
- For almost 80% of the projects the capabilities required can be covered in a good way by using either Apache CXF or Apache Axis2. If already you application is based on Spring Framework, I would suggest using Apache CXF along with Spring Framework OR using Spring WS with use of CXF as the web service engine.
- Spring WS similar to the Spring programming model has good and clean API’s, making it easy for developers (juniors) to code web services in Spring WS fast and easy. If you would like more standards, i think Apache CXF is a good choice over Apache Axis2. Again Apache Axis2 is slowly getting itself eaten up by Apache CXF. Fro an enterprise application point of view longevity and community support id required and here CXF scores well compared to Axis2.
Spring WS > Apache CXF > Apache Axis2
I cannot really say that Spring WS is above Apache CXF, because they are very very good fraewmorks. According to your application requirement, i would suggest using either Spring WS or Apache CXF.
Finally to finish off the blog, i would again like to share a picture, which gives Google Search Trends for these Web Service Frameworks in the year 2013 – WorldWide. How much it gives an edge to Apache CXF is questionable, nevertheless gives an indication as to what people are looking for in Google with regards to these frameworks and which is more active.
There is a beautiful page maintained in Apache which compares the various WS frameworks as a whole. The page is edit last in the year 2011, but I felt the page is a good share in my blog (note Spring WS is missing in the list because it is not an Apache project).
- http://forum.springsource.org/archive/index.php/t-105424.html (post by Grzegorz Grzybek)
Page Visitors: 6905
I have done a similar blog post earlier, namely Spring Integration Vs. Apache Camel, surprisingly it received good review from the community and i thought i will do a similar exercise on one of the other big topic of comparison these days, Web Services Framework.
Again, i am not biased on selecting any of the frameworks. The view expressed are solely mine. To be honest i don’t have a great deal of experience in the field of web services, but again as part of one of my projects i came into a scenario where i have to select on these frameworks and i thoughts to share the details that i collected. Again, i don’t want to give a verdict, as i don’t consider myself capable of doing so. Just so that anyone in situation can go through the blog and get details, compile it, present it to the team and based on it take the final decision for your project.
Also i have taken three web service framework, which i felt was top in the industry as of now. If anyone disagrees, please feel free to comment on the blog and looking at its relevance i will correct it, as i don’t want this blog to a misleading one for the community as a whole.
Apache CXF is an open source services framework. CXF helps you build and develop services using frontend programming APIs, like JAX-WS and JAX-RS. These services can speak a variety of protocols such as SOAP, XML/HTTP, RESTful HTTP, or CORBA and work over a variety of transports such as HTTP, JMS or JBI.
Apache Axis2™ is a Web Services / SOAP / WSDL engine, the successor to the widely used Apache Axis SOAP stack. There are two implementations of the Apache Axis2 Web services engine – Apache Axis2/Java and Apache Axis2/C.
Spring Web Services is a product of the Spring community focused on creating document-driven Web services. Spring Web Services aims to facilitate contract-first SOAP service development, allowing for the creation of flexible web services using one of the many ways to manipulate XML payloads.
Comparison from ohloh.net for the two projects:-
I think the comparison figure that is shown above doesn’t convey the exact comparison that we expect. Just by looking the figure above, Spring WS scores very low and can be taken out immediately from comparison. BUT that’s not the case IMHO. Lets get dirty and try the various comparison in some detail.
This is something new in this post, job trends on these technologies from indeed.com is as shown below. What it shows? It might not be relevant, but it shows what companies are looking at hiring when it comes to these technologies. It might be misleading but it shows that what companies are thinking and which technology is popular in various organization’s mind.
The above comparison figure on the various web service framework’s job opening’s doesnt really give any technical comparison between them but is a good to know fact. It also gives some trends followed by various companies.
Note:- Spring Web Service project was added by me in ohloh.net, solely for comparison perspective. The code location was already added in the main Spring Framework project. If anyone feels that this is not the right way, please feel free to send me a comment and i will do the needful. Also, if it is violating any of the legal terms doing so, please don’t sue me.. :). If given proper directions, i will correct it as early as possible.
In Part II of the post, i will be covering the comparison of these frameworks in detail, looking at various aspects and trying to get some conclusion as to which framework has an edge over the other and what is the reason for me to think so.
Let me have my short and sweet weekend and by next week you can read my blog post, which will be more useful and informative.
Page Visitors: 1647
The debate on which integration technology to use namely Spring Integration versus Apache Camel is one of the prominent topic now a days similar to mobile web application versus native mobile apps.
I don’t have any bias towards either Camel or Spring but for judgmental reasons I wanted to know which one to choose when given a chance in one of my projects. While doing so I did good amount of net surfing, and this blog post is an effort from my end to consolidate all the points that I could see in the internet. I have indeed plagiarized certain very good points from various blogs while doing so and some of the prominent blogs I have given as reference. Over the course of time I will be updating the blog post with more details as and when I find time.
Note: – I still haven’t been able to find an answer to my question and I am still using both co-jointly in my application. When using a connector I still don’t have a clear picture whether to use Spring connector or Camel connector.
Apache Camel – is a powerful open source integration framework based on known Enterprise Integration Patterns.
Spring Integration – It provides an extension to the Spring Framework to support the well-known Enterprise Integration Patterns while using the principles of Spring programming model.
Apache Camel as well as Spring Integration provides a lightweight alternative to ESB (Enterprise Service Bus), focusing on the programming model in such a way that the developer spends less time with infrastructural boilerplate.
Comparison from ohloh.net for the two projects:-
- Camel has a total connector counts of >130 in all which does outscore Spring Integration by a great deal with Spring Integration catching up with every release. +1 Camel
- Both are using Apache License version 2.0 which states, you can do whatever you want with the code. +1 Camel +1 Spring Integration.
- Both projects have good activity with good amount of commits from good amount of committers. Camel does have an edge though over the number of commits with respect to Spring Integration. +1 Camel
- Both started in the year 2007. The effort put for building Camel outscores Spring Integration by large. The lines of code with respect to Camel also outscore Spring Integration by a large distance. +1 Camel for the features it has with respect to lines of code. +1 Spring Integration as code lines cannot be said as a plus point (sometimes less code can be more effective and non-buggy).
- Vibrant response community of Camel does outscore Spring Integration by a great deal. Might be because of Camel’s focused approach on integration as opposed to Spring’s all in one approach. What it means that most of the problems you face using Camel can be solved quickly by Googling as opposed to Spring Integration. Also the amount of stuff available in the case of Camel does outscore Spring Integration by a great deal as well. + Camel
- JIRA activity is healthy in the case of Camel as opposed to Spring Integration. Again I see this because of the focused approach of Camel as well as attributed to the vibrant community along with good amount of code lines as compared to Spring Integration. Healthy JIRA activity does mean that turnaround time for a bug is quick in the case of Camel as compared to Spring Integration. +1 Camel
- CamelOne conference for Camel and none for Spring Integration. Again this can be attributed to non-focus nature of Spring Framework as a whole as compared to Camel. +1 Camel
- Both has good amount of tooling support in the form of STS and Fuse IDE. +1 Camel +1 Spring Integration
- Both have commercial support and good amount of 3rd party consultants who are ready to help if you have more money to shell. +1 Camel +1 Spring Integration
- Camel supports other DSL’s (Domain Specification Language) to a greater extend as compared with Spring Integration (ease of use and learn). +1 Camel
There is a very good presentation from Kai Wähner (Twitter: @KaiWaehner), which he presented in CamelOne Conference 2012, which does cover good amount of points and below is the slideshare details which he has shared. A well-documented comparison between various integration options available today on the Java (JVM) Platform. The author has played it safe similar to me by not giving a final verdict, but it will definitely help you to take into considerations and according to your application requirement take a right choice.
Some conclusions which can help the final decision:-
- If you already have a Spring application which requires an integration layer, definitely Spring Integration is a good choice to make. If you have a non-spring application, using/introducing Camel as the integration layer would be really the way to go.
I don’t want to take any credit away from authors who has taken pain in doing this research. That’s the reason I would like to put the reference as the first section in my blog. I have gone through various blogs, tutorials and then compiled a blog in a focused and easy to understand manner.
One of the post from Mark Fisher will surely give an insight into the reason behind developing Spring Integration – http://www.coderanch.com/t/568364/Spring/Camel-Spring-Integration-Part
Page Visitors: 8906