Engineering papers on google (map/reduce; linux clusters; google file system)

[Subbarao Kambhampati <rao@asu.edu>]

Last class we talked about the fact that the main engineering feat in 
Google may well be the use and maintenance of the thousands of linux machines.

Here is a paper from google labs (appeared in the top Operating system 
conference-- actually OSDI is ranked as number 1 for
impact of publication venues according to citeseer 
http://citeseer.ist.psu.edu/impact.html )

It explains their programming platform--called map/reduce (those of you 
lisp geeks know what "map" and "reduce" are
for example  to get the sum of squares of first 10 integers you just do
(reduce #'+  (map #'square '(1  2 3 4 5 6 7 8 9 10)))

Check out the paper and slides at:

http://labs.google.com/papers/mapreduce.html

A high level paper on google linux cluster is at:

http://rakaposhi.eas.asu.edu/gcluster.pdf

Finally, a paper on Google file system is at:

http://labs.google.com/papers/gfs-sosp2003.pdf

happy reading
Rao