You can configure a WebSphere scheduler for high availability by configuring it
at the cluster scope. A scheduler at the cluster scope means that schedulers are
replicated accross all the servers that are members of the cluster. Each of these
schedulers will have poll daemons that will poll for tasks from the same database
and this will result in collisions. How websphere reduces this is by the using the concept of
leases. At one time only one of the schedulers in the cluster will be granted a lease
by the lease manager and only that scheduler's poll daemon will poll the database and
will load and process tasks. This reduces contention and also ensures that if one of the
schedulers go down, another one will get the lease in its place and continue processing
tasks when they expire.
More information on WebSphere schedulers can be found in the WebSphere infocenter.
I am also linking to an excellent article on developerworks on WebSphere Schedulers
and I highly recommend that you read it if you plan to utilize WebSphere schedulers in