Appistry Blogs

  With the U.S. presidential elections in the not-too-distant past, I thought it would be a timely opportunity to highlight a few of the results of a survey we conducted at the recent CloudCamp Silicon Valley.

As part of an ongoing series, we are discussing design principles that influence how ready our code is for distributed  computing in the cloud, as well as for multi-core utilization. Today, we talk about…

Parallelism

"Parallel code?" you ask, "I don't need no stinkin' parallel code!" In the past, writing parallel code was an area of specialization, and most of us did not have a parallel processing computer with specialized hardware in our basements.

However, most new workstations, servers and laptops today have multi-core CPUs. However, even If you do have a dual core, quad core (or more) setup and your software doesn't utilize it, then you are leaving computing power lying, literally, on your desk, your rack, or your lap.

I can’t believe it’s taken me so long to post about the amazing CloudCamp Silicon Valley event we held a couple of weeks ago. But there’s no better time than the present, especially since it gives me a chance to plug next week’s CloudCamp Chicago.

As part of an ongoing series, we are discussing design principles that influence how ready our code is for distributed computing in the cloud, as well as for multi-core utilization. 

Today we'll talk about idempotence…today we'll talk about idempotence…today we'll talk about idempotence…. Ya, all this and cheap humor too. If you don't get the lame humor, no problem, just read on.... :-)

Idempotence

If code is atomic and stateless (as we discussed in part 2 and part 3) it may take on other attributes like idempotence. Idempotence says I can execute the same method repeatedly and get the same outcome without adversely affecting anything else, like state.

I remember the first few days of what became Appistry very well. Not simply well as in “cool we’re starting something that’ll change the world” well, but really, really well as in “day 2 of our awesome new venture was ... 9/11”.

Not just some 9/11, the 9/11.

So as our team watched that morning in the same horror and disbelief that is etched so indelibly in our collective cortex, we were faced with a very practical question – what do we do now?

Sure we had a pretty strong hunch that we could actually make the world of commodity infrastructure safe, easy, and cheap for enterprise software. Probably even creating the most reliable computing infrastructure in the planet, and sure we had been able to raise seed funds fairly easily, but that was yesterday, even early this morning ... what about today?

As part of an ongoing series, we are discussing design principles that influence how ready our code is for distributed  computing in the cloud, as well as for multi-core utilization.

A friend of mine used to say, "All generalizations are false, including this one." Generally, the rule of thumb in distributed computing is that stateful objects are bad, and stateless is good. Today we'll explore the advantages of being stateless on the cloud and in concurrent programming.

(with apologies to the good Dr. Seuss on the title - sorry, I just couldn't help it)

Last week I was privileged to participate in a panel on Parallel Computing at emTech08, a cool event put on by the MIT Technology Review publication.

The participants included

I'm looking for two students in the St. Louis area for a part time internship here at Appistry. This position will report to me, Appistry's Chief Architect.  During the internship we will work on prototypes of distributed algorithms to use in upcoming releases of Appistry software.

Requirements:

• Currently a senior or graduate student in Computer Science or related field
• Able to work at least 15 hours a week
• Familiar with C/C++, Java
• Familiar with TCP/IP and network programming

If you are interested or have any questions about this position, please contact me at michael@appistry.com.

It's CloudCamp time again. Please join us for CloudCamp Silicon Valley as we return to the Bay Area after a successful event in London.

----------------------------------

CloudCamp Silicon Valley – September 30, 2008

Call for Participants

CloudCamp was formed in order to provide a common ground for the introduction and advancement of cloudcomputing. Through a series of local CloudCamp events, attendees can exchange ideas, knowledge and information in a creative and supporting environment, advancing the current state of cloud computing and related technologies.

As part of an ongoing series, we are discussing design principles that influence how ready our code is for distributed computing in the cloud, as well as for multi-core utilization. Today, we talk about...

Atomicity

An atomic piece of code is code with a specific, clearly defined purpose. In object-oriented terminology, it has "cohesion." It's the pepperoni pizza and not the garbage pizza with everything. Likewise, atomic code can stand alone from the order of execution of other code. We’ll explore different aspects of atomic code below.