Analysis of time-sharing system and real system

1. Ctss

Time sharing system, which can be used to provide a computer to multiple users and improve the utilization of computers. It is often used in the query system to meet the needs of many query users. The needs of users are specific in the following areas:

(1) human-machine interaction. Every time a programmer writes a new program, it needs to be debugged on the computer. Because the newly-written procedures are inevitably some errors or mistakes need to be modified, and therefore want to be as early as the use of the computer when the same direct control, and can be modified by the side of the operation of the process, the error in the program to modify, that is, want to be able to interact with human-machine.

(2) shared host, in the 1960s the computer is very expensive, it is impossible to like now a person to monopolize a computer, and can only have multiple users to share a computer, but users in the use of the machine should be able to like their own computer, not only can at any time to interact with the computer, And you should not feel that other users are using the computer.

(3) Easy to user on the machine. In the multi-channel batch processing system, the user must send his or her homework to the computer room in person before the machine. This is very inconvenient for users, especially remote users. Users want to be able to directly transfer the work to the machine through their own terminal for processing, and can control their own operations.

It is not difficult to know from the above that the time-sharing system refers to a host on a plurality of terminals with a display and keyboard, while allowing multiple users through their own terminal, to interactively use the computer, sharing the resources in the host.

The first real time-sharing operating system was developed successfully by MIT.

Key problems in realization of 1.1 time-sharing system

In order to realize the time-sharing system, some column problems must be solved. Among them, the most important question is how to enable users to interact with their own jobs, that is, when the user enters the command on their own terminal, the system should be able to accept timely and timely processing of the command, and then return the results to the user. After that, the user can continue typing the next command, which is human-computer interaction. It should be emphasized that the system should be able to accept and process these commands in a timely manner, with multiple users entering commands at the same time through their own keyboards.

(1) Timely acceptance. It is not difficult to accept commands or data entered by the user in a timely manner, so simply configure a multiple card in the system. For example, when you want to connect 8 terminals on a host, you must configure a 8-user multi-card. The function of multi-card is to enable the host to accept the data entered by each user from the terminal at the same time. In addition, a buffer must be configured for each terminal to stage the command or data typed by the user.

(2) timely processing . the key to human-computer interaction is to enable users to type commands to control their own operations in a timely manner, or modify their own jobs. To do this, each user's job must be in memory, and should be able to get the processor to run frequently. Otherwise, the commands typed by the user will not work on their own jobs. Batch processing system is not able to achieve human-computer interaction, because usually most of the work is also mainstream in the external memory, even the operation has been transferred into the memory, but also often after a long time to wait before the run, thinking that the user typed commands difficult to timely effect on their own jobs.

Therefore, in order to realize human-computer interaction , it is necessary to completely change the original batch processing system operation mode. First, the user's job cannot enter the erase disk first and then into the memory. Because the job does not run on disk and the user is not able to interact with the machine, the job should go directly into memory . Second, a job is not allowed to occupy the processor for a long time until it runs out or an I/O request is dispatched before other jobs are scheduled to run. To do this, you should specify that each job only runs for a short period of time (for example, 0.1 seconds, usually referred to as a time slice), and then pauses the operation of the job and immediately dispatches the next program to run. If in a not long time (for example, 3 seconds) can make all the user jobs are executed once (a time slice of time), so that each user can be in time to interact with their own jobs, but the user's request to be timely response.

1.2. Characteristics of time-sharing system

(1) Multiplexing. Allows simultaneous connection of multiple online terminals on a single host, and the system serves each user on a per-tick basis. On the macro level, multiple users work at the same time, sharing system resources, while on the micro, each user's job turns to run a time slice. Multiplexing, which is simultaneous, improves resource utilization, lowers usage costs, and facilitates wider application of computers.

(2) independence. Each user occupies a terminal, independent operation of each other, non-interference. Therefore, the user feels that he is a person exclusive host.

(3) Timeliness, the user's request can be in a short period of time to get a response. This interval is determined by the waiting time that people can accept, usually close to 1-3 seconds.

(4) Interactivity. Users can use the terminal and the system for a wide range of man-machine dialogue. Its wide-ranging performance in: Users can request the system to provide a variety of services, such as file editing, data processing and resource sharing.

the time-sharing operating system rotates in accordance with the same schedule The time- sharing operating system automatically calculates the priority of the process by the scheduler, rather than the priority of the user-controlled process. Such systems cannot respond to external asynchronous events in real time.

2. Real-time system

The so-called "real-time" means "in time", while the real time system refers to the system's ability to immediately respond to requests for external events, complete processing of that time within a specified time, and control all real-time tasks to run in unison.

The real-time operating system can execute the required functions within a limited time and respond to external asynchronous events within a limited time. time-sharing system is mainly used in the case of scientific calculation and the low requirement of the general realtime. Real-time system is mainly applied to time-sensitive occasions such as process control, data acquisition, communication and multimedia information processing.

2.1 Application Requirements

(1) Real-time control

(2) Real-time information processing

Analysis of time-sharing system and real system