Hi, just to acknowledge Zooko's comment, it's clear to me that there are multiple more important topics than adjusting Tahoe-LAFS to minimal RAM.
Maybe I should take the way of least resistance and try to make an upgrade to some more modern, powerful and fatter hardware.... such as the Rasperry Pi *g*. The Rasperry is going to ship with 256 MB RAM but its CPU is said to be relatively slow. I also tried (so far unsuccessfully) to install Tahoe on a Nokia N900 Phone which has a OMAP3 SoC with ARM Cortex A8 CPU and, for purposes of cryptography much more interesting, an integrated TMS320C64x DSP. This platform is certainly more powerful than my NAS, so I think it's completely realistic to run a Tahoe-LAFS client on a N900 or comparable open platforms. I would love to help a bit, it's interesting enough for me. I'll try to dig a little into the code at least. I think if I can do something, it would be rather highly focused changes in some hot spots than general overhauls. OTOH I currently also have to address some job-related issues... I would love to get paid for work at that level, heck. What Brian does around Ed25519 looks great.... What are the main speed bottlenecks in Tahoe-LAFS now? Could it be grid latencies / network bandwidth for fast systems and encoding for slow processors? And would a parallelized variant of the encoding make sense for a typical multicore consumer laptop? Regards, - Johannes On Mon, 5 Mar 2012 15:49:43 -0700 "Zooko Wilcox-O'Hearn" <[email protected]> wrote: > I wrote: > > On Mon, Mar 5, 2012 at 12:24 PM, Zooko Wilcox-O'Hearn > <[email protected]> wrote: > > > > Just to be explicit about this, I'm not going to start doing these > > experiments myself anytime soon. I would love for Tahoe-LAFS to get > > slimmer on memory usage, and to continue to fit into more and more > > cute little ARM devices. (I just ordered my Raspberry Pi!) But, I > > have higher priorities in Tahoe-LAFS development right now. I'll > > offer > ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ > > encouragement, advice, feedback, and patch review. > > For the record, my highest priorities in life nowadays are my family > and Least Authority Enterprises. LAE work often contributes directly > or indirectly to Tahoe-LAFS development (we have a policy of > contributing all of the source code that we write, plus trying to give > back in terms of documentation, knowledge we've gained, etc.). But it > isn't primarily about Tahoe-LAFS development. > > In the context of Tahoe-LAFS development, my first priority right now > is unblocking Brian's accounting work by releasing a version of > pycryptopp with his Ed25519 port in it: > > https://tahoe-lafs.org/trac/pycryptopp/milestone/0.6.0 > > I'm the limiting factor in that, since we decided to use pycryptopp as > the vehicle for tahoe-lafs to use Ed25519, and since I want to do the > API and the new pycryptopp release my way. > > Once Brian is sufficiently unblocked, my next priority within > Tahoe-LAFS development will be helping Kevan and others address the > regressions and bugs in the new 1.9 code: > > https://tahoe-lafs.org/trac/tahoe-lafs/milestone/1.9.2 > > Regards, > > Zooko > _______________________________________________ > tahoe-dev mailing list > [email protected] > http://tahoe-lafs.org/cgi-bin/mailman/listinfo/tahoe-dev _______________________________________________ tahoe-dev mailing list [email protected] http://tahoe-lafs.org/cgi-bin/mailman/listinfo/tahoe-dev
