[tor-dev] Speeding Up Tor with SPDY

[Christian Grothoff]

Dear all,

Andrey Uzunov's Master's thesis on "Speeding Up Tor with SPDY"
is now available at https://gnunet.org/content/speeding-tor-spdy

My personal conclusions are that SPDY PUSH should not be used with
Tor, and that modest performance gains with SPDY are attainable
for typical websites.  Aside from deployment considerations (how
to detect exits supporting SPDY2HTTP translation), a major area
for further work will be adding HTTP pipelining support to the
SPDY2HTTP proxy to avoid performance regressions seen with some
sites.

For those that prefer watching over reading, we'll post the
defense talk on https://gnunet.org/ once we've finished the
conversion.

Feedback is of course welcome. However, as Andrey is now finished
with his thesis and looking for work (hire him!) -- and as I don't
have any funding to support further work on this now -- it is unlikely
that we'll be doing any further major implementation work on this
in the near future.  Naturally, the existing code is freely
available (and does work, if one is willing to go through the
pains to set it up properly).


Happy hacking!

Christian

-------------- next part --------------
A non-text attachment was scrubbed...
Name: signature.asc
Type: application/pgp-signature
Size: 259 bytes
Desc: OpenPGP digital signature
URL: <http://lists.torproject.org/pipermail/tor-dev/attachments/20131118/60253ec8/attachment-0001.sig>