Distributed system and decentralized system
1、 What is a distributed system
A: distributed system is more accurately a system architecture concept, not a technology. In short, it is to distribute the operating system in different computers, which are managed together
Second, what is the function? What are the benefitsA: improve safety and efficiency
distributed system generally adopts client / server mode, multi-layer, server cluster and other technologies. It's the mainstream now
each system has its own advantages. Moreover, the division of these two systems is generally from the perspective of engineering, and there is no such definition in textbooks.
advantages of distributed system:
characteristics of centralized system, distributed requirements, advantages of distributed system: easy to use; Robustness and reliability; Resource sharing; Scalability; Proction efficiency of end users; Easy maintenance
different from stand-alone operating system in process communication, resource management and system structure, process communication is different from stand-alone: it does not share memory; Low reliability; The implementation of communication depends on the communication facilities
communication protocol: semantic, syntax, timing resource management is different from single machine: multiple managers manage the same kind of resource management mode: distributed centralized: a class can have multiple resources, but each resource itself is managed by a single manager
distributed computer system is a kind of computer hardware configuration mode and corresponding function configuration mode. It is a multiprocessor computer system, each processor through the interconnection network to form a unified system. The system adopts the distributed computing structure, that is, the tasks processed by the CPU in the original system are distributed to the corresponding processors, so that the processors with different functions can coordinate with each other and share the peripherals and software of the system. In this way, the processing speed of the system is accelerated and the logical structure of the host is simplified. It is especially suitable for the automatic control of instrial proction lines and the management of enterprises and institutions. It has low cost and easy maintenance, and has become an important direction of the development of computers in the application field
centralized system, mainly refers to the system of IBM, HP and other small computers above grade, one host with multiple terminals. The terminal has no data processing ability, and all operations are carried out on the host. Most of today's banking systems are centralized systems. In addition, they are also distributed in large enterprises, scientific research institutions, the military and the government. Centralized system, mainly popular in the last century. Now the centralized system is still used, a large part of it is to use the original software, which is very expensive.
Distributed system is a system in which hardware or software components are distributed on different network computers and communicate and coordinate with each other only through message passing
because of the characteristics of software, distributed system has a high degree of cohesion and transparency. Therefore, the difference between network and distributed system lies more in high-level software (especially operating system) than in hardware
advantages of distributed system
1. Economy: microprocessor provides better performance price ratio than mainframe
2. Speed: the total computing power of distributed system is stronger than that of single mainframe
3. Inherent distribution: some applications involve spatially dispersed machines
4 Reliability: if a machine collapses, the whole system can still run
5, graally increase: the computing power can graally increase
The real driving force of the system tends to be distributed is economy. Twenty five years ago, computer authority and critic herb GROSCH pointed out that the computing power of CPU is directly proportional to the square of its price, which later became GROSCH's theorem. That is to say, if you pay twice the price, you can get four times the performance. This conclusion is very consistent with the mainframe technology at that time, which makes many organizations do their best to buy the largest single mainframe
with the development of microprocessor technology, GROSCH theorem is no longer applicable. By the beginning of the 21st century, people could buy a CPU chip for only a few hundred dollars, which could execute more instructions per second than the processor of the largest mainframe in the 1980s. If you're willing to pay twice as much, you'll get the same CPU, but it runs at a higher clock rate. Therefore, the most cost-effective way is usually to use a large number of cheap CPUs in a system. Therefore, the main reason for the preference of distributed system is that it can potentially get much better performance price ratio than a single large centralized system. In fact, distributed systems achieve similar performance at a lower price
slightly different from this view, we find that the set of microprocessors can not only proce better performance price ratio than a single mainframe, but also proce absolute performance that a single mainframe cannot achieve in any case. For example, according to the technology of the early 21st century, we can use 10000 modern CPU chips to form a system, and each CPU chip runs at a rate of 50 MIPS (million instructions per second), so the performance of the whole system is 500000 MIPs. If a single processor (i.e. CPU) wants to achieve this performance, it must be in 2 × It takes 10-12 seconds (2 picoseconds, 0.002 nanoseconds) to execute an instruction. However, no existing computer can approach this speed. It is considered theoretically and practically impossible for any computer to meet this requirement. In theory, Einstein's theory of relativity states that light travels the fastest, and it can travel 0.6 mm in two picoseconds. In fact, a computer with the computing speed mentioned above, which is contained in a cube with a side length of 0.6 mm, generates a lot of heat and can melt itself immediately. Therefore, whether it is to get ordinary performance at a low price or to get high performance at a high price, distributed system can meet the demand< On the other hand, some authors distinguish distributed systems from parallel systems. They think that distributed system is designed to allow many users to work together, and the only goal of parallel system is to complete a task at the fastest speed, just like our computer with a speed of 500000 MIPs. In our opinion, the above distinction is difficult to establish, because in fact the two design fields are unified. We prefer to use the word "distributed system" in the broadest sense to refer to any system with multiple interconnected CPUs working together
another reason for building a distributed system is that some applications are distributed. A supermarket chain may have many branches, each of which needs to purchase locally proced goods (which may come from a local farm), sell locally, or make decisions about which local vegetables have to be thrown away because they have been too long or rotten. As a result, the local computer of each store can understand that the inventory is meaningful, rather than concentrating on the company's headquarters. After all, most queries and updates are done locally. However, from time to time, the top managers of supermarket chains want to know how much cabbage they have. One way to achieve this goal is to build the whole system to be like a computer to the application program, but it is distributed in the implementation. As we described earlier, a store has a machine. This is a commercial distributed system
another inherent distributed system is commonly known as CSCW (Computer Supported Cooperative Work). In this system, a group of people who are physically far away from each other can work together, for example, to write the same report. As far as the long-term development trend of computer instry is concerned, people can easily imagine a new field CSCG (Computer Supported Cooperative Games). In this game, players who are not in the same place can play the game in real time. You can imagine playing electronic hide and seek in a multi-dimensional maze, or even playing an electronic air war together. Everyone controls their own local flight simulator to try to shoot down other players. The screen of each player shows the situation outside his plane, including other planes flying into his field of vision
compared with centralized system, another potential advantage of distributed system is its high reliability. By distributing the workload to many machines, a single chip failure will only stop one machine at most, while other machines will not be affected. Under ideal conditions, if 5% of the computers fail at some time, the system will still work, only 5% of the performance will be lost. For the key applications, such as the control system of nuclear reactor or aircraft, the distributed system is used to achieve high reliability
finally, incremental growth is also a potentially important reason why distributed systems are superior to centralized systems. Usually, a company will buy a mainframe to complete all the work. When the company prospers and expands, the workload will increase, and when it increases to a certain extent, the host will no longer be competent. The only solution is to either replace the existing mainframe with a larger machine, if any, or add another one. Both of these will cause confusion in the operation of the company. In contrast, if the distributed system is adopted, only adding some processors to the system can solve this problem, and it also allows the system to expand graally when the demand grows. These advantages are summarized in Table 1-1. Project description economic microprocessors provide better performance, price and speed than mainframes. The total computing power of a distributed system is stronger than that of a single mainframe. Some applications involve spatially dispersed machine reliability. If a machine crashes, it will be more reliable, In the long run, the main driving force will be the existence of a large number of personal computers and the need for people to work together and share information. This information sharing must be carried out in a convenient form, not affected by geography or the physical distribution of people, data and machines. Since using microprocessors is a way to save money, why not give everyone a personal computer and let them work independently? For one thing, many users need to share data. For example, the staff of the air ticket booking office need to access the main database that stores flight and existing seat information. If you back up the entire database for each staff member, it won't work in practice, because no one knows which seats other staff members have sold. Shared data is the basis of the above example and many other applications, so computers must be interconnected. And computer interconnection proces distributed system
sharing is not just about data. Expensive peripherals, such as color laser printers, photo typesetters, and large storage devices, such as CD ROMs, are shared resources
the third reason for connecting a group of isolated computers into a distributed system is that it can enhance communication between people. E-mail has more attractions than letters, phone calls and faxes. It's much faster than a letter. Unlike a telephone that requires two people at the same time, or a fax, the files it generates can be edited, rearranged and stored in a computer, or processed by a text processing program
finally, distributed systems may be more flexible than giving each user a separate computer. Although one possible model is to give everyone a personal computer and connect them over a LAN, this is not the only one. In addition, there is a mode of mixing personal computers and shared computers together (these machines may not be exactly the same model), so that the work can be done on the most suitable computer, not always on one's own computer. In this way, the workload can be more effectively distributed in the computer system. The failure of some computers in the system can also be compensated by making them work on other computers. Table 1-2 summarizes the points described above. Project description data sharing allows multiple users to access a common database device sharing allows multiple users to share expensive peripheral devices (such as color printers) communication makes it easier for people to communicate, such as distributing workload to available machines in the most effective way through e-mail flexibility
1、 The common points of distributed system and computer network system are as follows:
most distributed systems are built on computer network, so the physical structure of distributed system and computer network is basically the same
Second, the difference between computer network and distributed system is that the design idea of distributed operating system is different from that of network operating system, which determines that they are also different in structure, working mode and functionnetwork operating system requires network users to understand network resources when using network resources. Network users must know the function and configuration of each computer in the network, software resources, network file structure, etc. if users want to read a shared file in the network, they must know which computer and which directory the file is placed in
distributed operating system manages system resources in a global way. It can schele network resources arbitrarily for users, and the scheling process is "transparent". When the user submits a job, the distributed operating system can select the most suitable processor in the system according to the needs, submit the user's job to the processor, and then pass the result to the user after the processor completes the job
in this process, users will not be aware of the existence of multiple processors, and the system is just like a processor
As a whole, computer network is a large-scale and powerful system that connects computers distributed in different geographical areas with special external devices by communication lines, so that many computers can easily transfer information to each other and share hardware, software, data information and other resources. In short, the computer network is a collection of many independent computers connected by communication linesthe simplest computer network has only two computers and a link connecting them, that is, two nodes and a link
The difference between computer network and distributed system is that the design idea of distributed operating system is different from that of network operating system, which determines that they are also different in structure, working mode and function
network operating system requires network users to understand network resources when using network resources. Network users must know the function and configuration of each computer in the network, software resources, network file structure, etc. if users want to read a shared file in the network, they must know which computer and which directory the file is placed in
distributed operating system manages system resources in a global way. It can schele network resources arbitrarily for users, and the scheling process is "transparent". When the user submits a job, the distributed operating system can select the most suitable processor in the system according to the needs, submit the user's job to the processor, and then pass the result to the user after the processor completes the job
in this process, users will not be aware of the existence of multiple processors, and the system is just like a processor
READ MORE