??? Have replacement hardware on hand: Hardware fails. Death, taxes, and hardware
failure are life??™s only truths. Yes, there is the occasional exception, but you should
always focus on extremes when planning disaster recovery. Having a spare hard
drive or power supply can save you a world of time, even if you have the Diamond-
Studded Platinum level of support from the hardware vendor.
CHAPTER 7 n CLUSTERING 193
Summary
While useful and efficient for dealing with specific hardware issues and software upgrades,
clustering leaves a lot to be desired when it comes to disaster recovery. It doesn??™t mitigate
much in the way of risk, and it??™s only truly effective in response to a hardware or software
problem on one of the nodes. The benefit of easing the upgrade process is solely in the
realm of high availability, so that doesn??™t help us much either.
So why even use clustering? Well, it??™s ubiquitous, and several applications and
resources can take advantage of it. After all, clustering isn??™t a function of SQL Server
but, rather, a feature provided by Windows. Clustering offers some (limited) level of
risk mitigation, so why not take advantage of it? Clustering isn??™t difficult to install, nor
is it difficult to apply service packs.
Pages:
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375