Applications that gracefully degrade and re-enable as possible are easier
to manage and maintain. They change the mode of application instability to monitoring,
instead of scrambling to reboot and provide service. In some ways, we are talking about
building smarter systems. At the heart of any autonomy is a core system that can indicate
for it and others that a problem exists.
Future.Trends
Just as software design and architecture continue to evolve, underlying infrastructure and
middleware continue to provide an ecosystem within which applications can execute. Key
to maximizing future infrastructure advances is the ability for an application to tolerate
and to exploit many of these advances. Recently, autonomic computing technologies have
advanced to the point where grid-like technologies can be delivered, without the need for
disruptive approaches to be embedded within application designs. By having a middleware
environment that provides the needed abstraction layer around the underlying physical
deployment topology, applications can be managed against a set of predetermined policies.
These policies, while enforcing against a breach of their predefined goals, will provision
and de-provision instances of an application, all without having the application or pre-mediation
of the event.
Pages:
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341