Appeal No. 96-3077
Application No. 08/255,544
Any task in the cluster can potentially
communicate with any other task in the
cluster without needing to know precisely
where the other task is executing, but only
whether it is executing in the same or in
a different program.
In our view, this passage indicates that, while the sending
task in Gammage et al. does not require the exact location of
a remote task to be invoked, such sending task requires
knowledge as to whether, in the terminology used by Gammage et
al., a local rendezvous or a remote rendezvous is required.
The discussion of local and remote rendezvous on pages 13-15
of Gammage et al. indicates that a remote task identification
is required for the sending task to invoke a remote
rendezvous. The inclusion of such remote task identification
is effectively an internal message path selection included in
the sending task definition.
The Examiner further argues (Answer, page 5) that the
inclusion of “target locale and node” information in the
identification of message type in Gammmage et al. is not
precluded by the use of the transitional phrase “comprising”
in Appellant’s claim 1. On this point, while the Examiner’s
statement regarding preclusion is correct, we agree with
6
Page: Previous 1 2 3 4 5 6 7 8 9 10 11 12 13 Next
Last modified: November 3, 2007