Enterprise Integration–The Seriously Nasty Stuff
Recorded at:
- Share
-
- |
Read later
My Reading List
Bio
John Davies is co-founder and CTO of Incept5, which is involved in implementing Visa's new capabilities and initiatives around the world. John's past includes Global Chief Architect at JP Morgan and BNP Paribas, co-founder and CTO of C24, later sold to Iona and then Progress Software where he was Technical Director. John specializes in high performance, low latency enterprise architectures.
SpringOne 2GX is a one-of-a-kind conference for application developers, solution architects, web operations and IT teams who develop business applications, create multi-device aware web applications, design cloud architectures, and manage high performance infrastructure. The sessions are specifically tailored for developers using the hugely popular open source Spring technologies, Groovy & Grails, and Tomcat. Whether you're building and running mission-critical business applications or designing the next killer cloud application, SpringOne 2GX will keep you up to date with the latest enterprise technology.
Re: not worth the time
by
peter lin
One could argue "why didn't john davies go into technical details?" Honestly, it's hard to get into details without diving into the specs. Do you really think that's going to appeal to the general public? I had to work with those specs for past jobs and honestly it wasn't joyful. Just about everyone that works with those specs finds them painful once you get into the nitty gritty. For example, try running .Net xsd.exe on accord spec and see what happens.
watched 40 min of video and almost 30 slides out of 43
by
Harsh Gupta
Will watch rest of the video and update
Re: not worth the time
by
Harsh Gupta
Also he mentioned using C++ at one point. Increasing permgen size to load 5000 classes is eventually required. JaxB uses sax parser and is very fast but the object itself will require memory.
Anybody's opinion on this part? Was expecting more on this front....
Re: not worth the time
by
John Davies
This presentation lasts 90 minutes, 60 of that was about integration in banking and the other 30 was a demo of a tool that anyone can download and use for free (from www.C24.biz). To use the SWIFT, ISO-20022, SEPA, FIX & FpML etc. in production you need to part with money, there is NOTHING in the open source realm that implement the SWIFT rules so I'm afraid you're stuck with sales pitches. This one is about as near as you can get to the open world as it was integrated with Mule, Service Mix, Spring Integration & Camel.
Re: not worth the time
by
John Davies
It's always a dilemma to know whether to dive into detail and walk through the code or cover the problem at a higher level hoping that the people who want details will engage afterwards, these integration issues are so specific to large enterprises (HL7, Accord, SWIFT, ASN.1, ISO-8583, ISO-20022 etc.) that unless you know what the problem space is you're going to end up walking out after the first 15 minutes and that's not good for the conference.
I have had a number of emails (to John dot Davies at Incept5 dot com) asking about this talk and I'm always happy to delve into details, supply code examples or reference to production examples.
I'm doing a shorter (60 minutes) version of this talk (without the "sales pitch") at SpringIO in Madrid on the 17th Feb if anyone's in the area.
-John-
Re: watched 40 min of video and almost 30 slides out of 43
by
John Davies
I've had a chance to do a little work outside the financial services and I found it very useful being able to apply my knowledge in these areas. The "Integration Objects" product was created as a result of our pain and those of people I worked with over the years (and decades). We must have done something right because it's been very successful and now used by the majority of the world's largest investment banks.
It is free to download and use but you will need a runtime license for production use of the complex libraries.
Regards,
-John-
Re: not worth the time
by
John Davies
If we get something seriously complex to process then we often "pre-parse" the messages and then distribute them, an example would be processing a large number of complex SWIFT messages where we can only parse, validate & transform a few thousand per second per server. If we need a higher throughput we can simply distribute them but sometime it's more efficient to parse part of the message first and then distribute them.
When we need seriously low latency then we simply have to move to C/C++, our very fastest FIX engines, used by the lowest latency traders are all C/C++, it's the predictability as well as the latency that rules out Java. By the time we get to this stage we're into FPGA, RabbitMQ, ZeroMQ and LMAX etc.
Send me an email if I can be of further help,
-John-
Hello stranger!
You need to Register an InfoQ account or Login to post comments. But there's so much more behind being registered.Get the most out of the InfoQ experience.
Tell us what you think