Appeal 2007-1476
Application 10/308,866
Notwithstanding the Examiner's statement (Answer 14) that Dias
discloses replacing a first failed node with a designated spare node1 and
therefore discloses the step of identifying a second member in claims 1 and
32, we find nothing in Dias to teach or suggest when a first subsystem (or
member) fails on one node, identifying a second subsystem (or member) on
a second node to replace the first subsystem, where the two subsystems
perform comparable services. Similarly, we find no teaching or suggestion
in Dias of causing the service provided by the first member to be provided
by the second member, as recited in claim 18. If anything, we find
suggestions to the contrary. For example, that "[r]ecoverable failure of any
instance of any subsystem triggers recovery actions which are taken for the
other instances of that subsystem and also for subsystems that interact with
or depend on that subsystem" (see Dias, col. 3, ll. 60-63) suggests that
recovery actions are applied to all comparable subsystems, not that one
subsystem is substituted for the failed one. Similarly, by specifying (col. 10,
ll. 1-2) the "set of nodes on which the command should execute," Dias
suggests that recovery commands are applied to multiple nodes, not a
command to a single second member to replace the failed member. Also, by
specifying (col. 10, ll. 18-19) "the maximum number of times that the
command will be executed if the command does not execute successfully,"
Dias suggests randomly trying other members rather than identifying the one
that will replace the failed member.
1 We note that the Examiner's reference to replacing one node with another
(Answer 14) suggests that the Examiner has taken Dias' nodes as the claimed
members. However, as indicated by Appellants (Reply Br. 2) a member is a
resource on a node, like Dias' subsystem, not the node itself.
5
Page: Previous 1 2 3 4 5 6 7 Next
Last modified: September 9, 2013