[racket] File buffering

Is there a way to control the size of the buffer used when opening an input
file?  Looking around collects/racket/port.rkt (don't know if I should have
been looking elsewhere), it appears buffers of 4K and 1K are used.  I'd
like to have much larger buffers.  Is my main option to give a large number
to read-bytes and then gradually read from that result?  If I go that
route, will I actually bypass the 4K buffer (if that is indeed the buffer
size)?   For context, I'm reading lots of large files (10's of GB each) and
want to reduce filesystem load by reading bigger chunks of each file at at
a time.  Thanks in advance for any advice!  --Wayne
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.racket-lang.org/users/archive/attachments/20150127/d95dba28/attachment.html>