Continued
Failure Type Probability Predictability Impact
Process High There??™s always some Could range from embarrasslevel
of predictability, be- ment to major downtime and
cause the events happen data loss.
at a fixed time.
User Usually low, Almost impossible to pre- Could range from a minor
depending on dict, although poorly trained annoyance to catastrophic.
training and staff and a poorly designed
application application may be a hint.
design
Probability, predictability, and impact together help to prioritize disaster recovery
planning. If a disaster has a low probability, is difficult to predict, and has a relatively low
impact, there??™s no sense in placing it at the top of the list of action items (or placing it on
the list at all).
I??™ll refer back to these categories and scenarios throughout the book, applying specific
technical features to each particular category and the example scenarios.
Disaster Recovery from a Technical Perspective
Up to this point, I??™ve been approaching the question of disaster recovery from an
abstract, procedural level. While it??™s important to think about the subject in an abstract
way, this is a technical book.
Pages:
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68