Introduction to Virtual Machines (vii)--OSI virtual machine model

OSI Virtual machine model



"Virtual" is to hint at a certain time, although an object or device does not exist, but people can see it



The word "virtual machine" is easy for many people to have a rich association. The word "virtual" is often used in popular media and news broadcasts when reporting on new developments in computer simulations and computer games. In either case, the use of "virtual" is suggestive of a certain time, although an object or device does not exist, but one can see it. This use of virtualization is a reasonable assumption about expanding the development of virtual machines. The creation and continued development of the extended virtual machine is the dual logic improvement of the computer operating system in ability and flexibility.





The key to understanding the extended virtual machine model is to understand how modern computers are designed and how the operating system controls it. The modern operating system consists of a series of instruction sets, which combine to form a service routine. The service routines and data are combined to become applications submitted to the computer. Combining instructions into a single service routine is like an office worker combining simple actions to accomplish a more complex task (such as an office supply catalog). For example, a clerk has only one limited set of instructions that can be executed in memory. These instructions are limited to calculations, record the results of the previous operation, and archive some information for later use. The use of these instructions alone is of little value, but these instructions can be grouped into a sequence that allows clerks to complete an office supply directory. This sequence of instructions may be: Count the number of records, record the results, count the number of pencils, record the results, count the number of paper clips, record the results, and save the archive. This sequence can be reused to complete the directory service request. Using the order "count the number of notes", it is assumed that office clerks have instincts such as identifying record paper and counting accurately. For humans, recognizing objects and counting is the natural function of the clerks ' brains and eyes. For the simulation of office clerks or computers, these functions need to be designed at the hardware level and embodied in the circuit system. When you look at the actions of office clerks in detail, it is clear that you need to implement some of the basic functions that are often used in the circuit system, rather than combining simpler instructions. Office clerks must walk around the office in person, check objects, record counting results with notepad and pencil, and so on. These features will be implemented as a series of repetitive hardware actions and software directives. For example, the "Record last results" directive would include accessing a storage part (such as disk storage), locating the space for storing data, obtaining data to record in a local register, transferring data, and writing data to a hard disk. In the circuit layer, there is a basic layer, all instructions are derived from this level.



The bottom-most instructions are the 01 code that is submitted to the computer. They generate mobile data, configure the system, or create conditions for the next action. This level can be considered machine instruction level, the operating system level of instruction is derived from this level. Now, we all know clearly that any instruction set is a simple combination of its next level of instruction set. The task that the office clerk wants to complete is actually a series of subtasks. These subtasks are grouped together to form a more complex sequence of operations. When the directory command is issued to the Office clerk, the Observer can only know the meaning of the Word directory and the overall action of the Office clerk. The underlying instruction subset is not visible to the observer. Each successive instruction layer, including machine circuitry, machine code, operating system directives, and application code, hides the underlying instructions from its next layer of users. The ability to combine instructions and build more complex operations is the key to understanding extended virtual machines.







The ultimate goal of extending a virtual machine is to build capabilities and analog devices that are not present in the computer. For example, most computers have a hard disk to store data and program instructions. This hard drive may actually be on another computer, or on a number of devices on your computer. In this case, the extended virtual machine method is to write a sequence of instructions that makes the user feel that there is only one hard drive on the computer. This single drive is modeled with the same storage capabilities as the hardware drives on all computers. With a computer, the user sees only one hardware drive. Users can use a combination of instructions to store and retrieve data as if they were on a large hardware drive. A user-invoked instruction accesses a subroutine that is composed of the underlying instruction. The details of what information is stored on which hard drive the underlying instruction in the subroutine handles.