Results tagged “architecture” from manAmplified

How can you measure how far a given complex System is from the simplest System it can be? Here's a hint: distance = sqrt( Tp^2 + Tm^2 + Tv^2 + Tl^2 )

I've released a new version of JStress. Many libraries have been updated. Dead code removed. Packaged it into a single jar file, libraries included, so it can simply be pushed to a remote system and run. And lastly, provided R scripts for generating profile plots from a a LinearStepRun.

Interesting presentation given at C4, Actors: A Unifying Design Pattern for Scalable Concurrency.

Recently Jnan Dash of Foldera made a great comment regarding persistence. He said there are two ways to park your car. The first is to disassemble it and store it on a shelf, then reassemble it when you need it again. Or just put it somewhere fully assembled.

There has been alot of talk around S3 for storage and backup. I find EC2 much more interesting as a foundation for a perpetual computing cluster.

Even though NetKernel is shrouded with new edge terms like 'Transreptor' and 'Fragmentor', it is quite a practical piece of software. Listen to Jon's 'Innovator' podcast with Pete at 1060 (yeah Tony, your an Innovator too) for a audible view into the NetKernel architecture.

Correlating my recent experiences with Progress Software, the company behind Sonic MQ, and the current market trend of their stock, I can only assume big trouble ahead. A shame as their recent acquisition of Actional filled a large hole in their product offering. I would like to downgrade their status to 'replace' pending identification of a more suitable product.

I don't think I can stress enough the utility of having a staging environment that you can clone the relevant bits of and then surgically implant into an ailing production system.

One value, for me, of an ESB framework is to improve the reliability and robustness of a given set of integrated applications. If a message/event producer can communicate with one or more consumers in an asynchronous way, it should. It really makes things simpler. But one thing to avoid in your routing configurations is injecting synchronous calls into an overall asynchronous process.

One of the most powerful tools available in managing change and complexity is by sandboxing applications. This is simply achieved by reducing the number of external dependencies to an applications runtime environment, creating the most natural level of isolation as possible.

I've put together a simple NetKernel module that tends to greatly simplify service and application development. mod-e4x provides helper functions allowing for rapid construction of services using the ECMAScript for XML syntax provided in the Rhino JavaScript interpreter. With mod-e4x, JavaScript becomes a viable alternative to the NetKernel DPML language.

Had a recent conversation with a Palo Alto stealth-mode startup member. Seems they want a flat organizational structure and thusly have hired 4 architects. I've had difficulty interesting developers in positions here unless they are offered an Architect title, even though they don't have more than 5 years professional experience.

If you are observant, you might have noticed that the latest ECMA Script for XML (aka JavaScript) specification defines, and the latest Rhino release includes, support for literal XML manipulation. If you are inquisitive, you would have discovered that the latest NetKernel release supports E4X on the server-side.

This older article on SAML myths myths enumerates a number of SAML implementations.

If you work with distributed systems, HTTP referenceable services, and specifically caching systems, you probably should check out On the Intrinsic Locality Properties of Web Reference Streams.

This blog entry has inspired me to make a few notes to work out some ideas through this rather meandering essay.

In very simple terms (read convenient simplification, see below), a queue is a serialized list of pending requests. And it is typically used to broker/filter/throttle requests to a naturally concurrent system behind it. What's interesting is that systems are typically compositions of other naturally concurrent (sub)systems. Turtles all the way down, if you will...

Just returned from what I thought was intended to be a discussion over patterns and best practices within an ESB deployment, likely sourced with examples from this new book, since the meeting was hosted by its author. Sadly it proved to be ESB "as" a best practice, where they handed out copies of the book, leaving the interesting discussion as homework.

A Distributed Lock Manager can be found on the IBM OSS site.

Seems that the internet is about to crash, if you believe MS and Intel. Some of the woes stem from the very nature of HTTP request/response model, and is aggravated by the rise of RSS/Atom news aggregators. So, is a push model the solution?

Lost Boy asks what RDF storage and retrieval solutions people are using. I offer some quick notes on our setup.

Part of my staging environment is built on User Mode Linux. Even this website is hosted in a FreeBSD Jail. This of course allows many OS instances to run simultaneously on one machine.

In the book "Wisdom of Crowds", the author (James Surowiecki) describes the observation that a group of sufficiently independent individuals are better decision makers (are more accurate) than the individual (or even the expert). If you aggregate a collection of independent opinions/answers, the 'local knowledge' of each individual (whether they be wrong or right) improves the performance of the group. If the individuals are not independent, that is they communicate with one another and share knowledge, you risk that influence leading the group astray.

In response to 'test driven development', I propose 'stress driven architecture'.

When developers are asked to load test a system, most will start up a million threads to create a load on the server showing that it will eventually become un-responsive, but they have no numbers that allow them to create a load profile so they can properly plan for peak times in a production environment.

In an interview, Victoria Livschitz touches on a problem that has been bothering me for some time. Notably, "The sequence of the routine itself -- what comes before what under what conditions based on what causality -- simply has no meaningful representation in OO, because OO has no concept of sequencing, or state, or cause."

Obviously, in any industry, a given vendor will have strengths and weaknesses compared to her competitors. The BPM space being no different seems to have two opposing ends. Back-end integration and human-to-human workflow, which are met in the middle with a process language.

Inappropriate Abstractions walks on network transparency, which Voyager proved fallacious, and befuddles the issues around transparent object persistence. Be sure to read the ongoing discussion.