Demonstration data (Demo Data)
: great for getting to grips with Sage Accounts before you move on to real data because any mistakes you make don't affect your live company
Structure of Policy Rules
Understanding how a policy rule is structured is helpful in understanding other Policy Management concepts. A policy rule is defined in an if-then structure, consisting of a set of conditions that the MPE device compares to information extracted from protocol messages or obtained from subscriber records, and a set of actions that are executed (or not executed) when the conditions match. Many conditions can be tested for existence or non-existence (by optionally selecting the logical operator NOT or using, where available, the policy condition operator is or is not).
Policy Parameters
When you define a policy rule, you select from a list of available conditions and actions. Most of the conditions and actions have parameters (that is, they contain placeholders that may be replaced with specific values to allow you to customize them as needed).
For example, consider the following policy rule, which has one condition and two actions:
where the device will be handling "greater than 100 upstream reserved flows"
apply profile Default Downstream Profile to request
continue processing message
- An operator (greater than)
- A value (100)
- The flow direction (upstream)
- The bandwidth reservation type (reserved)
The first action, apply profile..., specifies a single parameter that is the name of a traffic profile to be applied to the request.
The second action, continue processing message, instructs the MPE device to evaluate the remaining rules within the policy rules list (as opposed to immediately accepting or rejecting the request). The conditions and actions that are available for writing policies are discussed later in this section.
Reference
'AI > Data Science' 카테고리의 다른 글
| [선형대수학] Norm, 행렬곱, 내적 (0) | 2023.05.30 |
|---|---|
| SVM (0) | 2023.05.16 |
| [ML] 빅데이터 메모리 사용량 줄이기 (0) | 2022.12.05 |
| [이론] The Kalman filter, 칼만 필터 (1) | 2022.10.26 |
| [pandas, python] 시계열/날짜 데이터 전처리 (0) | 2022.10.14 |