Appeal No. 2003-0186
Application No. 09/033,529
criteria of the device which are outlined in columns 19 and 20, the criteria include
whether the device is a low priority “no-impact” (column 20, lines 66-67), or a
“real time (point of failure)” (see column 20, lines 36-44) and bandwidth required
(see column 18, lines 55 to 64) . The priority curve is divided into three regions
(see column 17, lines 34-35). Both the region of the device’s priority curve and
the value of the priority counter are used to determine which device’s request will
be processed (see column 17, lines 54-67). The priority of a request increases
as time goes on (see figures 2 and 4).
The appellants have argued, on page 8 of the brief, that Barnaby does not
disclose monitoring the non-expedite requests by a non-expedite counter. We
agree. On page 10 of the answer the examiner provides the following
explanation:
Barnaby explicitly discloses that all clients are measured for memory
usage in a window of time and that for all clients latency and bandwidth
requirements are calculated at col. 17 lines 16-23 and col. 19 lines 47-52.
Barnaby then discloses that the window of time where all clients are measured
for memory usage is defined as the interval of bandwidth and latency
measurements col. 19 lines 47-52. Therefore, memory usage is defined by
bandwidth and latency which is in turn measured in terms of time and number of
clock cycles. Also measuring implies counters; and as all clients including those
issuing non-expedite requests are measured, non-expedite requests are
measured with counters that may be termed non-expedite counters.
As stated above we find that the scope of claim 1 includes that the non-
expedite counter is monitoring total time to perform non-expedite requests. The
examiner has not shown where Barnaby teaches this limitation. We find that
7
Page: Previous 1 2 3 4 5 6 7 8 9 10 11 12 Next
Last modified: November 3, 2007