MuleCon2007 Mashup

Eugene Ciurana, Director of Systems Infrastructure at Leapfrog, is presenting on Enterprise Application Mashups with Mule ESB.

Once upon a time, Eugene did industrial robots - we'll have to make sure that he and Clarise get to talk, as she still talks about her Singapore days, doing robots for Motorola.

Case Study 1: eCommerce Site

The objective for a large, international retailer, was to make an Architecture that would last for 5 years, through a 10 months skunkworks, that wouldn't force system upgrades on business units. The prime directive was to acquire not build, and to go with "best of breed". This prime directive was in recognition that no ONE vendor is best in every area. There was also a clear separation of concerns between platform infrastructure and applications, such as ERP, eCommerce, CMS, etc. [BI, where's BI in this plan?]. &#59;)

The result was a mixed environment of over 10 vendors, 3 hardware platforms and geographically distributed. How to integrate all of this? The answer is an ESB, and they looked at commercial and open source alternatives. From the commercial, one vendor looked good, but with an insane price; their JMS component was included in the final solution. On the open source, they looked at:

  • Jboss
  • OpenESB
  • Tuscany
  • ServiceMix
  • Mule

Jboss and OpenESB requires you to eat their whole enchilada [my words], and it wasn't digestible. ServiceMix and Tuscany are built to meet JBI, not solve real world problems. Eugene wanted to keep flexible, and not commit to Java as the solution for every problem. Mule was the solution. The resulting architecture has eCommerce Suite, Order Capture, Order Management, Web Service, Single Sing-on, CRM, CMS, Product Information, B2B/EDI and BPEL hanging off of the Mule Services Backbone, bringing about a successful reference implementation. This is being used to support a customer facing web structure that has 50 MILLION hits per day. The team only had 21 days to build the environment and integrate all vendors' products. Working with the Mule team, the first Mule-enabled components were the eCommerce Suite and CRM, which only took 7 days. Integrating the other 8 vendors took 20 minutes. More about this can be found from Eugene's site.

Case Study 2: Download Store

The objective was to create a download store, with a geographically dispersed team in 60 days. The work is still ongoing, but Eugene went into some of the details and showed the architectural diagram.

We're blogging from MuleCon2007 throughout today and tomorrow.

MuleCon2007 Introduction Day 2

Dave and Ross welcomed the well-over 100 attendees for MuleCon2007 Day 2. Let's just say that Dave should leave his day job for a stint in stand-up comedy. :D

Many of the ~25 MuleSource employees are here, and Dave introduced them. The staff is distributed among almost every time zone on Earth. They're looking for recommendation on team management software [They're living the TeleInterActive Lifestyle; ServiceCycle anyone?]

Here's some Mule stats:

  1. Over 600K downloads in 300+ enterprises.
  2. 58.3% of the users have build extensions to Mule.
  3. 67% want their current vendor to be more open.
  4. 47% feel the Jeda would win over the Ninja, only 8% thought superman would be the winner
  5. Approximately 400 pages of documentation on the website, but more is needed. Everyone has write access - sign up, and write away.
  6. Mule 1.4 & 2.0 - release dates at Ross' whims [Hey, it's open source], but they are planning more scheduling

The Mule team is always interested in understanding use cases - tell them what you're doing.

Mule is going international with a growing presence in Japan, UK and EMEA.

MuleForge

The MuleForge was developed to help manage the contributed code and extensions. The forge will also be a developer resource with components and use cases that don't belong in the main distribution.

Partners

MuleSource is looking for more partners to extend their reach. Contact Mike Lewis, the man in the pink shirt.

Thanks to Travis for the link from the MuleSource Blogs.

We're blogging from MuleCon2007 throughout today and tomorrow.

MuleCon2007 Training

The first half of the afternoon session, here at MuleCon2007, were simple labs inspecting and configuring the XML files for Mule, and seeing how they relate to the Java files. My MacBookPro can now ask your name, and then greet you by name, as a web service. :D

Peter is now going through modeling of applications, using order processing as an example. The applications to be integrated include a Sales tool, in Java, the Warehouse [physical not data] in COBOL, a Logistics application, Accounts, and the Assembly process. Since the Assembly process has no application in place, we'll create a service to track this process and facilitate interaction with the COBOL warehouse application via flat files, and with Logistics via JMS.

Think about the power inherent in that ability to interact with legacy applications.

There's a lot more in the online documentation from the Mulesource wiki.

And now it's back to more labs.

Mulesource and Ricston have put together a great day. There are about 100 people here from all over the world. This is a true testament to the interest and reach that open source projects and companies are now generating.

We'll be blogging from MuleCon2007 throughout today and tomorrow.

MuleCon2007 Routers

Types of Routers

  • Inbound
  • Outbound
  • Response
  • Filters
  • catchAllStrategies

One can stack routers & transformers

Peter gave examples of the ease of creating various special purpose routers, such as a wiretap, exception based or chaining router with the familiar XML configuration, showing the ease, power and flexibility of the Mule router implementation. Also take a look at the EIP book.

We'll be blogging from MuleCon2007 throughout today and tomorrow.

MuleCon2007 Recovery Strategies

Exception Strategies

Component Exception Strategies

Connector Exception Strategies

Essentially route/log

Fatal exceptions such as a message queue breaking also can be handled

Reconnection Strategies

Configure how a connector behaves when it connection fails

  • retry # & frequency
  • other basic reconnect strategies are included
  • create your own

We'll be blogging from MuleCon2007 throughout today and tomorrow.

MuleCon2007 Messaging Style

Asynchronous Inbound & Outbound

Asynchronous Inbound & Synchronous Outbound

Synchronous Inbound only

Synchronous Inbound & Outbound

Synchronous Inbound & Outbound with transaction

Request-Response allows a UMO to make a request and await a specific response

See diagrams for each of above from the to-be-published presentation [URI to come]

In mule, one can specify how many instances of a UMO can be invoked and configure various message threads to create a complex flow

Mule can automatically match incoming & outgoing messages or fkn support event calls throug its callable interface.

One can control synchronicity through the endpoint property

We'll be blogging from MuleCon2007 throughout today and tomorrow.

MuleCon2007 Theory

Most configuration in the XML file can be done either by setting a property or using a URI

Configuring JMS, embedded ActiveMQ

Using Queues & Topics

http://mulesource.org/wiki/

Selector Configuration Options

Durable subscribers can be configured

SOAP Core Transport

Hosting a web service with Mule can be done using AXIS orXfire - Example of simple HelloWorld POJO exposed as a web service - configure lnbound router & endpoint with transport

Accessing a web service with Mule can also be done using either AXIS orXfire; the same approach is used whether local or remote for AXIS but WDSL is used for remote access via Xfire

Over 25 Transports includeded with Mule; configuring one is much like any other. Mule also provides a framework to easily add other componts

With Xfire one can use the client, but more common is to see a series of chain routers.

The Big Plcture - see diagram - Endpoint [message receiver] -> Inbound Router -> Interceptor -> Inbound Transformer -> Service Invocation -> Interceptor -> Outbound Router -> Outbound Transformer -> Endpoint [message dispatcher]

Interceptors - see diagram from the presentation [link TBS]

We'll be blogging from MuleCon2007 throughout today and tomorrow.

MuleCon2007 Overview

Peter Delia, co-founder with Ross Mason [CTO of Mulesource], of Ricston, based in Malta, is providing today's training, starting with an overview of Mule, Mulesource and Ricston.

SOA Topologies; Mule can be configured to implement any topology

  • ESB - see Mule project front page
  • ESN
  • Peer Network
  • Client Server/Hub & Spolke
  • Pipeline

Mule Architecture: Each Mule instance connects to many Applications and/or Service Components & other Mule Instances; each Mule Instance has a Mule.Manager & Service Containers; more detail on wiki; Mule supports a wide variety of transport providers.

Core Concepts - see diagram on wiki

  • Service Components - can be simple POJO
  • Endpoints [message receiver or dispatcher, connector, transformer] & Transport Providers [synchronous & asynchronous endpoints possible]
  • Inbound Routers
  • Outbound Routers
  • Transformers
  • Service Manager

Question - running threads in Mule, running Mule on slower, but multi-core machine - it depends on specific needs and processes, as to whether or not it is advisable, but it is possible.

Configuring Mule

  • extensible
  • XML file
  • DTD
  • root is UMOManager [Universal Message Object = Service Component]

XML Configuration File Structure

  • Mule Environment
  • Connectors - used to send & receive data over a particular channel: name, className, 〈ref〉 - it is possible to configure connectors on endpoints
  • Transformers: name, className, returnClass, property
  • Endpoint - used to control how events are sent & received, best configured on routers; transformers are configured on endpoints; Global endpoints, Endpoint Identifier
  • Routers - inbound Routers, outbound Routers, response Routers, filters, catchAllStrategies
  • Service Components: mule-descriptor element
  • Mule Components

We'll be blogging from MuleCon2007 throughout today and tomorrow.

Campus Technology 2007 Schedule

The schedule for the Campus Technology 2007 conference is online; a PDF of the brochure is also available for download. In addition to our session, there are several other workshops and talks related to either BI/DW or open source solutions.

We're hoping to coordinate with other speakers, so that our sessions are complementary and to avoid duplication.

Will you be going to CT 2007? What would you like to see discussed in terms of BI/DW and open source solutions? See you there.

Open Source To Be or NOT

Is a company worthy of referring to itself as open source or not and thus presenting as an open source vendor at OSCON? First, I think you must differentiate between open source projects and open source companies. And next, everyone would have to agree on what "open source" actually means.

IMNSHO, a project is open source if all of the source code is available as a download, under a license, or expressed permission, to recompile, change and use that source. Using a version control system such as CVS or subversion is great, but even just including it in the zip archive or tarball is fine too. Of course, other things are adopting the open source philosophy as well, such as Sun Microsystems open sourcing the underlying chip architecture of Niagara. So, perhaps an open source project is anything that provides enough information [system, functional, design and component specifications, for example] from which one could recreate the original project. And of course, the license to do so.

But which license? There are so very many, though there seems to be two main camps: the GPL and Berkeley/Apache. And let's not forget the badgeware wars for the MPL+attribution license. Dual-licensing or not? Commercial or not? Can the source code be available but not free [as in beer]? To me, it's open source not freeware, so yes, yes and yes. That's not the popular opinion. :>>

Nat Torkington started things off...

I think EnterpriseDB is a tricky boundary case. They've got software that helps bring people directly to OSS from closed source. That's surely worthy of exposure at OSCON -- it'll help attendees convince their bosses to use more open source. But on the other hand, it fails my "can people build it?" test.end quotation
-- Nat Torkington in O'Reilly Radar "Is "Open Source" Now Completely Meaningless?"

Allison Randal responded on her own O'Reilly blog...

Nat recently made a provocative post taking the position that companies like EnterpriseDB who don't release all of their source code don't belong in an "Open Source" conference. Really, it goes back to the age-old GPL vs. BSD licensing argument. GPL takes the perspective that proprietary versions should never be allowed, and so requires all versions of GPL'd software to be released under the GPL. The BSD license, as well as Apache, MIT, Artistic, etc., take the perspective that proprietary versions of open source software are an expected part of the open source ecosystem. They're even desirable, especially when the company involved is contributing back to the open version of the code, and employs developers to work on the open version of the code.end quotation
-- Allison Randal in O'Reilly Radar EnterpriseDB is/n't Open Source

and the debate was on.

EnterpriseDB is quite frank about what it's about. A shark is not the cute, cuddly mascot we associate with open source. You buy from an outfit with a shark as its mascot, you know what you're getting into.end quotation
-- Who decides what is open source? by ZDNet's Dana Blankenhorn

EnterpriseDB is not an open-source product. In my opinion, they're not an open-source company either.
"I happen to favor the GPL for a number of reasons but there are a number of other, perfectly valid, open-source licenses. It's interesting that some say that BSD doesn't require EnterpriseDB to publish their source code. But, BSD doesn't prevent them from publishing their source code either. For that matter, there's nothing preventing Oracle from publishing their source code save their desire to be a closed-source, proprietary product.end quotation
-- Dave Dargo's Random Thoughts "If You Can't See It, It Ain't Open"

All of which begs the question: what open source benefits do any of EnterpriseDB's customers get? A lower cost product and better quality, yes, but also the same lock-in you'd find with any proprietary product. Is that the promise of open source?end quotation
-- Matt Assay in InfoWorld Open Sources "More on what constitutes an open source company"

When Eric & Bruce originally coined the term "open source" they intended to trademark it and began the application process under Software in the Public Interest, Inc. Unfortunately, SPI forked into SPI and OSI due to political differences of opinion, and in the process the trademark was lost. So "open source" is now public domain and legally speaking any company can use it however they want.
"However, serious open source geeks like Nat do still have control over some things despite the lawyers, like who gets to keynote at OSCON. So he's started wondering what the definition of an "open source company" should be. What follows are a number open-source-ish things which companies do; I'd like you to let me know which things, or combinations of things, should be the measure of an "open source company". And I'll share my opinion tommorrow.end quotation
Josh Berkus in ITtoolbox Database Soup "Who's open source? (part 1)"

For my part, having had the model fully explained to me by Astor, I took the decision to stop referring to EnterpriseDB as an “open source database vendor” in news stories and articles, as can be seen from our recent article on VC funding:
"'The level of interest in open source is evident in the experience of EnterpriseDB, which offers an enterprise database of the same name built on the open source PostgreSQL code base… While EnterpriseDB's key database enhancements, such as Oracle compatibility, and reliability improvements, are not open source, Astor believes that the firm's involvement in the open source community was a key factor in raising the funding.'
"However, I continue to include the company in CBR’s list of open source funding deals because clearly an investment in EnterpriseDB is an investment in the strengths of the open source development model.end quotation
-- by Matthew Aslett on March 1, 2007 11:25 AM in Open Source Weblog "The EnterpriseDB debate – some insight from the top"

All of which brings us back to the question, when is a company an open source vendor? As I said at the beginning, first, I differentiate between an open source project and an open source company. If a company can dual-license, such as MySQL with their community and enterprise, cluster, & carrier versions, and still be lauded as a pinnacle of the open source movement, than I would say that a company whose main source of income is in leveraging an open source project, and that they are recognized as being a strong supporter of that project in terms of money, developer time, community participation, etc. than yes, that company is an open source company, and I think EnterpriseDB is an open source company, as is Greenplum, who like EnterpriseDB leverages PostgreSQL, but in a different way, and with a different type of participation via Bizgres. To be clear, I don't think that what any company that contributes back to, and possibly distributes, an open source project is an open source company. IBM and HP have been good for open source, and for various open source communities, and especially for Linux. Sun Microsystems has more wholly embraced the open source philosophy with openSolaris, Niagara, and now Java, and their DW appliance with Greenplum. Would you call Sun an open source company? I'm undecided. Pentaho brought in the "best of the best" of the open source BI/DW open source components: Mondrian, jFreeReport, jPivot, KETTLE, WEKA as well as added framework and workflow, and they have a very interesting model in how those projects are still independently run, and how the Pentaho project is run and its community developed, and how the company has evolved [no longer having a professional edition]; but at first, some argued that they were taking advantage of open source and doubted their sincerity [private conversations, no web links, sorry]; I don't think anyone would argue against Pentaho anymore, and the nay-sayers were proven wrong. These [EnterpriseDB, Greenplum and Pentaho] are three examples of commercial open source companies, and I think they all deserve to be called so. Some of these doubts and questions have also been raised about Ingres and Actuate [BIRT for Eclipse]. I'll leave the decision on those two as an exercise for the reader.

It will be interesting to see how this debate goes, though I think we'll be having it for years to come.

Updated with more quotes, better links and some clarifications 2007.03.01;18h33PST

May 2019
Mon Tue Wed Thu Fri Sat Sun
    1 2 3 4 5
6 7 8 9 10 11 12
13 14 15 16 17 18 19
20 21 22 23 24 25 26
27 28 29 30 31    
 << <   > >>

At the beginning, The Open Source Solutions Blog was a companion to the Open Source Solutions for Business Intelligence Research Project, and book. But back in 2005, we couldn't find a publisher. As Apache Hadoop and its family of open source projects proliferated, and in many ways, took over the OSS data management and analytics world, our interests became more focused on streaming data management and analytics for IoT, the architecture for people, processes and technology required to bring value from the IoT through Sensor Analytics Ecosystems, and the maturity model organizations will need to follow to achieve SAEIoT success. OSS is very important in this world too, for DMA, API and community development.

37.652951177164 -122.490877706959

Search

  XML Feeds