Thirteen characteristics of embedded systems
Embedded computing not only requires fast and consistent computation of the network, but also requires the system to be able to orderly store its execution code and data in a “common” “small” space. This paper introduces the thirteen characteristic rules of embedded system.
1. Law of limited resources
Embedded computing not only requires fast and consistent computation of the network, but also requires the system to be able to orderly store its execution code and data in a “common” “small” space.
2 Robustness rule
Embedded computing not only requires the system to calculate quickly and efficiently, but also requires that the system can continue to work normally when some computing units are wrong.
3. Real-time rule
The calculation result of embedded system depends not only on the correctness of the system logic operation, but also on the calculation time of the calculation result.
4 Redundancy rule
After the embedded system has sufficient redundancy, the “initial sensitivity” of the system has little impact on its “final calculation result”.
5 Structural rules
For embedded systems, trends in their structural complexity indicate that:
a The simplest is the best;
b The More complex is more stable.
6 The Law of parsimony
When simplifying an embedded system, the interaction between the remaining functions of the system will become stronger and stronger; When the function of the system is reduced, the probability of success of external invaders becomes greater and greater.
7 Conservation law
If you want to preserve a certain system function in an embedded system, it is best to regard all other functions as “sacrosanct”; The removal (extinction) or generation (invasion) of a system’s functionality must result in a major change in the overall (cluster) structure and its dynamic performance.
8 Rules of organization
The most important thing in embedded Internet is often not the characteristics of individual devices in the network, but the overall order that exists in the network, that is, the network order.
In a highly redundant network, the single role of the device can no longer constitute the main factor affecting the overall performance of the system, but the main role is all the nodes and their connection characteristics.
9 Rules of network-ness
The properties of a network composed of a group of embedded Internet nodes interacting with devices as a whole often have no significant relationship with the properties of individual nodes.
10 The Law of Informality
Guaranteed search messages: It has strict time-sensitive or basic normal system operation requirements, such messages require a time guarantee from the system. That is, once the activities or tasks caused by such messages are executed, their time limit must be guaranteed by the system within a specified time interval.
Best effect message: It has a typical soft time limit, that is, its time limit is defined by the time series of the activity or task itself, and can meet its time limit requirements without system guarantees.
11 Immunity Rules
The immune system of embedded Internet should be a life-like mechanism, and the immune function is a “feedforward” system, so it is required that the system should have foresight, so that it can “attack (big) poison with (small) poison”.
12. The Law of integration
Embedded Internet is a complex network, the complex network structure with a simple “composition” to analyze, so that the system can be isolated “group” to interpret the “whole”, or so that the system can be “node” to express the “global”.
13 Value for Money Rule
If system A is an embedded system in system B, that is, B (a), then the cost of system A should not exceed 10% of the cost of system B, and the cost of system B (a) should be greater than the sum of the costs of system A and system B, and the cost performance of system B (a) should be increased by 30%.