Chapter 98. Capacity Planning
Capacity planning is complicated and hard, especially for CPU planning. It’s difficult to predict how much resource an application that you’ve never run in real-life before is going to consume. Fortunately, we live in the age of the elastic cloud, in which vendors like Amazon and Oracle can sell you the amount of CPU you need to get going, and then they can adjust your CPU capacity to your needs without having to physically move your application from one computer to another.
Even with this tremendous technology at your disposal, it is still important to be able to see a program you care about in a systemwide context. Here is a model that helps you do that.
In this model, the shaded cells are where you’d type your input, and the white cells are all computed. It’s just Mrs. Utley’s problem, but with different nouns.
| Item | Qty | Price | Total |
| WENUS Report | 1 | 3,600 | 3,600 |
| TPS Report | 4 | 200 | 800 |
| Book Order | 1,200 | 0.8 | 960 |
| Other | 1 | ? | ? |
| Total | 5,360 | ||
| Change | 9,040 |
-
Instead of having the capacity of a dollar, this system has a capacity of 4 CPU × 60 sec/min × 60 min/hr = 14,000 CPUsec/hr (cell D13).
-
The run/hr column (B) contains the quantity data. These values are how often the business wants to run each business function during the Monday 2:00 p.m. peak hour.
-
The CPUsec/run column (C) contains the price data. You learn these prices by ...
Become an O’Reilly member and get unlimited access to this title plus top books and audiobooks from O’Reilly and nearly 200 top publishers, thousands of courses curated by job role, 150+ live events each month,
and much more.
Read now
Unlock full access