Natural Order is the future of concurrent processing

  • strict warning: Non-static method view::load() should not be called statically in /hermes/walnaweb12a/b57/moo.greydragoncom/nodsw/sites/all/modules/views/views.module on line 906.
  • strict warning: Declaration of views_handler_argument::init() should be compatible with views_handler::init(&$view, $options) in /hermes/walnaweb12a/b57/moo.greydragoncom/nodsw/sites/all/modules/views/handlers/views_handler_argument.inc on line 744.
  • strict warning: Declaration of views_handler_filter::options_validate() should be compatible with views_handler::options_validate($form, &$form_state) in /hermes/walnaweb12a/b57/moo.greydragoncom/nodsw/sites/all/modules/views/handlers/views_handler_filter.inc on line 607.
  • strict warning: Declaration of views_handler_filter::options_submit() should be compatible with views_handler::options_submit($form, &$form_state) in /hermes/walnaweb12a/b57/moo.greydragoncom/nodsw/sites/all/modules/views/handlers/views_handler_filter.inc on line 607.
  • strict warning: Declaration of views_handler_filter_boolean_operator::value_validate() should be compatible with views_handler_filter::value_validate($form, &$form_state) in /hermes/walnaweb12a/b57/moo.greydragoncom/nodsw/sites/all/modules/views/handlers/views_handler_filter_boolean_operator.inc on line 159.
Leeland's picture

David Ungar gave a Talks on Computing Systems (TOCS) at Carnegie Mellon University called "Everything You Know (About Parallel Programming) Is Wrong!: A Wild Screed About the Future." (http://www.cmu.edu/silicon-valley/news-events/seminars/2011/ungar-talk.html). It’s just over an hour long and I am so excited to have the natural model being looked at. The imposition of artificial constraints to force computations to occur in a specific ordering has always rubbed me the wrong way. The majority of applications I have dealt with over the years just do not scale due to all of the extra constraints of trying to be 100% predictable. Life is never 100% predictable, but it is predictable enough to work well.

It is so improbable that any event in reality can be done the same way twice as to be reasonably impossible. For example, not even the greatest gymnast in the world can precisely walk across a balance beam in the exact same manor, with the exact same steps, exactly placing the feet, breathing at the exact same moments, moving the arms in exactly the same way. There will always be variances. Never the less they do it close enough to the same as to achieve great performances.

You might claim this is an overly complex argument. Very well, make it simpler. Not even a robot arm can place a playing card on a table in the exact same position every single time. There will always be variances.
Why then do we require our applications to produce the precise same results every single time?
The natural operations of the world scale to the level of the billions of galaxies. How? The infinite scaling is achieved by the simple nature of independent operations, handled by individual units, operating under a limited set of instructions and forces. Computer systems can be built using the same model. In fact the Internet is an example of this, there are individual computers, each acting independently with their own purposes, yet, interacting together based on a handful of simple rules and forces.

I will be watching this presentation a few more times to pull out every grain of detail I can from the algorithms and processes presented. More to come on this topic for sure!

Thread Slivers eBook at Amazon