[ExI] Techie question: best/fastest WAN file system solution

Bryan Bishop kanzure at gmail.com
Sun Aug 3 02:19:18 UTC 2008


On Saturday 02 August 2008, Samantha Atkins wrote:
> One thing that I grapple with a lot is how i/we can use network
> resources efficiently in at least small steps towards extending our
> brains via hardware and the network.   One aspect of this is looking

There's a few ways that I've been considering for storing brain related 
information, especially in neurofeedback datasets and MRI datasets and 
the like. The tricky part is extracting value out of it ... as in, 
semantic value, or doing something interesting with data dumped from 
your brain. You could try training artificial neural networks to act 
like small regions of your own functionality :-) which is something 
that Thomas DeMarse has shown relevant technology for. 

"A new approach to neural cell culture for long-term studies"

"The Neurally Controlled Animat: Biological Brains Acting with Simulated 
Bodies"

"Closing the loop: stimulation feedback systems for embodied MEA 
cultures"

"Removing some ‘A’from AI: embodied cultured networks"

"Poly-HEMA as a drug delivery device for in vitro neural networks on 
micro-electrode arrays"

> for and using the most efficient means of having the equivalent of
> fast dependable cloud data storage and retrieval. By this I mean

Hey, want to implement DeMarse's work with Amazon's S3? That would be 
worth a paper or two.

> network file system equivalents fully useable by any tools that
> access filesystems.   It is very important to me and I imagine a lot
> of folks to be able to access our persistent bits from wherever we
> are given a network connection.   I have looked at WebDAV, both on my
> own linux server and the Mac iDisk.  Pretty slow, especially for
> browsing reasonable sized directories of information and opening PDF
> and media docs.  Hell, any kind of document save, even small ones, is
> a bit tedious.  I tried sshfs but it is if anything a bit slower in
> my experience thus far.  Something that automatically cached locally
> might be better (the iDisk does some of that).  I haven't pulled out
> NFS as my very dated opinion is that it isn't secure enough.  I have
> used various offerings that use the Amazon cloud S3 data.  None of
> those to date were better than WebDAV.

There's a few solutions for this specifically so that there are file 
systems extended across the network, so that cached reads/writes occur 
and so on, but generally there's nothing that is working as well as it 
really should be and you're rightly worried. I know a few individuals 
that I can put you in contact with regarding distributed file systems 
(freenet was at one point somewhat like this). 

http://en.wikipedia.org/wiki/Freenet
http://en.wikipedia.org/wiki/Project_Xanadu <-- pipe dream ;-)

(Also, WikipediaFS is totally worth it ... if it would have directory 
listings ... which by the nature of the mediawiki, it's not quite easy 
to immediately have all of the categories in some giant 
root-node-origin structure .. )

- Bryan
________________________________________
http://heybryan.org/
Engineers: http://heybryan.org/exp.html
irc.freenode.net #hplusroadmap



More information about the extropy-chat mailing list