Appeal No. 2001-2553
Application No. 08/512,369
been obvious “to determine a popular processor so that efficient
load balancing can be performed” (answer-page 12). While it is
not clear from this rejection exactly how the alleged combination
is to be made, i.e., how the alleged teaching of Rudolph is to be
combined, exactly, with the alleged teachings of Belo or Baron,
this is not argued by appellant and is not an issue in the case.
Appellant’s argument centers on Rudolph and its alleged
teaching of a popular processor whose “unlocked local queue
contains at least a predetermined number of eligible threads.”
It is appellant’s view that “[t]he load of a workpile is measured
as the number of tasks on the workpile,” recited at page 240,
left column, of Rudolph, fails to teach “at least a predetermined
number” of threads in the workpile, even if we consider the
workpiles to be unlocked local queues. Appellant further points
to Rudolph’s recitation of a threshold value, J, at page 240 and
concludes that this compares a thread count in one workpile with
the thread count in another workpile, instead of comparing the
thread count in one workpile with some predetermined minimum
count, so that Rudolph fails to teach “at least a predetermined
number” of threads.
Since this is the only argument made by appellant relative
to the “minimum count claims,” and we do not agree with
-9–
Page: Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 Next
Last modified: November 3, 2007