Hi Stewart,

Yes, the database could be a SPOF, but if you need fault tolerance, there is a 
way to configure MySQL to have a failover server that will take over if the 
primary one fails. 

Running all the workers on a single machine it's all moot though - if that 
machine fails your SOL :). Swarming was designed for cluster use originally, so 
having no central controller and being able to leverage a MySQL server with a 
backup were helpful in that context. 

And yes, swarming is best run on a powerful machine or cluster of machines. You 
can then take the best model found from swarming and run it on a smaller 
system. 


On Sep 9, 2013, at 2:58 PM, stewart mackenzie <[email protected]>
 wrote:

> Apologies, I forgot languages like python use shared state concurrency.
> Multiple collaborating agents is a cinch given the right language features.
> 
> This approach introduces a single point of failure, the database.
> 
> Sensor networks consisting of small microchip type devices are too
> tiny to run MySQL.
> Hypersearch will need to be made light weight if it's to get out into
> sensor networks.
> I suppose target deployment are data warehouse/high performance
> cluster style setups.
> 
> Nevermind, given the tradeoffs it's a sensible decision, and it works.
> 
> Thanks for clearing that up for me.
> 
>> The advantage of using MySQL is that the co-operating workers have a joint
>> noticeboard. MySQL is specifically designed to handle issues like race
>> conditions and multiple collaborating agents. It's also ubiquitous and
>> easily installed, it's practically an OS function.
> 
> _______________________________________________
> nupic mailing list
> [email protected]
> http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org


_______________________________________________
nupic mailing list
[email protected]
http://lists.numenta.org/mailman/listinfo/nupic_lists.numenta.org

Reply via email to