Shulou(Shulou.com)05/31 Report--

This article will explain in detail what are the differences about Oracle service models, and the content of the article is of high quality, so the editor will share it with you for reference. I hope you will have a certain understanding of the relevant knowledge after reading this article.

Undefinedundefined

First, shared server mode.

Shared server mode, also known as multithreaded server mode. In this mode, the Oracle database allows multiple user processes to share very few server processes. Therefore, the number of users who can access concurrently at the same time is greatly increased. If you use dedicated server mode, each user process needs its own dedicated server process when each user connects to the database. In other words, the client-side process corresponds to the server-side process one by one.

If the shared server mode is adopted, multiple users can share a server process, that is, the client process and the server process are many-to-one. This is mainly achieved through the scheduler. The scheduler directs multiple join database access session requests to a common queue. A shared server process that is idle in the server process sharing pool gets a request from the queue in turn. In other words, a small shared pool of server processes can serve a large number of clients.

The advantages and disadvantages of shared servers can be summarized in two sentences. The advantage is that the client process is many-to-one, increasing the number of users that the database can support. The disadvantage is that all users share a process, which has an impact on the performance of users accessing the database.

Specifically, in the following application scenarios, you can choose to use shared server mode.

1. There are a large number of front end clients.

When a large number of users need to connect to the database and effectively use the available system resources, the shared server mode needs to be considered. Specifically, such as the current use of some client / server mode of network applications, such as ERP system and so on. On the one hand, because the number of users is relatively large; on the other hand, enterprises purchase general servers because of limited funds. These servers are due to some hardware limitations, such as a unified server for the database system and the foreground application server. Therefore, the database is often required to make full use of hardware resources to reduce the conflict between the two.

In this case, the shared server mode is often adopted, which is more appropriate.

2. The memory limit of the server is relatively large.

In shared server mode, when the number of connected users increases, the memory usage does not increase much. Because they share a server process. So, from this point of view, shared server mode can reduce memory usage. However, in dedicated server mode, memory usage is almost proportional to the number of users.

Therefore, if users deploy Oracle databases on some old servers, because the motherboard has some restrictions on memory upgrade, in order to get a good database performance, they often adopt the shared server mode. In this way, even if there is an increase in the number of users accessing the database at the same time, their memory will not have much impact. Can greatly reduce the pressure on memory.

3. Some specific functions require the shared server mode.

Although, shared server mode and dedicated server mode are common in most cases and support most of the functions of Oracle database system. However, some specific features still need to be enabled by the database administrator in shared server mode. More typical, such as Oracle database server connection sharing, connection centralization and load balancing technology and so on. They must be in shared mode to run.

Load balancing is used to achieve multi-computer sharing database in a cluster environment to ensure the high availability of applications. At the same time, it can automatically realize parallel processing and load sharing, as well as fault tolerance and non-breakpoint recovery of the database in the event of failure. Therefore, in some application scenarios with high requirements for performance and stability, such as banks, load balancing technology is often used. At this point, the database administrator needs to consider the shared server mode when configuring the database.

Second, dedicated server mode.

In dedicated server mode, the database server requires each user to have a dedicated server process. That is, each server has one server process. The database server sends the existing server process geology back to the client, and then the client resends the connection request to the provided server geology and compares it with the dedicated server. Because the client-side and server-side processes are one-to-one corresponding, the dedicated server model may have an advantage in performance. However, when there are more users, it will exert great pressure on the hardware resources of the server, especially the memory.

If the enterprise application scenario conforms to the following characteristics, the use of dedicated annual server model can achieve better results.

1. There are only a few clients.

If there are fewer users connected to the database, dedicated server mode is preferred. If the front desk of an enterprise is a financial management system, the number of people using it will be relatively small. Because it often involves only one department. However, the amount of data will be more. Because we know that every transaction needs to be recorded in detail in the financial system. Therefore, although it is only used by one department of finance, the number of records is often the sum of the records of other departments. In the case of a large number of records and relatively few connected users, it can be said to be a wise choice to adopt a dedicated server mode.

2. The database system for the data warehouse.

Data warehouse is a very typical application of database technology. It is a subject-oriented, integrated, time-varying, non-volatile data set and supports the decision-making process of the management department. Like Oracle database, it is a relational database, in which records are stored in tabular form and can be queried in a unified structured language. The main purpose of adopting data warehouse is to complete the business processing in time and give customers a timely response.

Although relational database can meet the needs of large-scale database applications, it can not be simply stacked into a data warehouse to use. Because the data warehouse is mainly used to deal with multidimensional data. The way he stores records is different from the ordinary one. Data warehouse mainly stores records through array, that is, there is no unified law to follow, there is no unified multi-dimensional model to follow, he can only buy and classify according to the category to which he belongs. Although the traditional two-dimensional database can not be simply stacked to become a data warehouse. However, when designing a data warehouse, we don't mean to start all over again, but we can use the existing Oracle database system to deal with the data and synthesize the information, so as to construct a data warehouse that meets different needs.

Therefore, if the database is built to implement a data warehouse, a dedicated server mode must be adopted. This is a mandatory requirement of the Oracle database.

3. Online transaction processing system.

Online transaction processing system, abbreviated as OLTP. In this kind of application, what mode should be adopted? There is no uniform standard in the industry. Some say that because of the large number of users of the online transaction processing system, the shared connection server mode should be adopted, while others say that the online processing system requires high timeliness, and the shared server mode may have a long waiting time, so the dedicated connection server mode should be adopted. The author is more inclined to the latter opinion.

Online transaction processing systems, generally speaking, have more long transactions and large transactions. Such as some of the user's fuck, must be as a transaction and so on. At this point, if the number of user requests is greater than the number of shared server processes, queuing occurs. If a shared server process is now performing a long transaction, the request queue needs to wait all the time. Until this transaction is completed. From the client's point of view, the response time of customer requests is longer. Therefore, when the user has more connection requests than the shared process, the shared connection with such a long transaction will inevitably cause the queue of the shared process. The longer the transaction, the longer the queue time and the longer the user response time. If this vicious circle continues, many online transaction jobs will be affected to varying degrees. Therefore, if the online transaction processing system meets two conditions (the number of user connection requests is larger than the shared process and most of its transactions are long transactions or large transactions), the dedicated server connection mode is more efficient than the shared server connection mode. In order to ensure the normal operation of the online transaction processing system, the author suggests that the database administrator configure the dedicated server connection mode for it. If one of these two conditions is not met, it is more reasonable to adopt the shared connection mode.

What are the differences between Oracle service models to share here, I hope that the above content can be of some help to you, can learn more knowledge. If you think the article is good, you can share it for more people to see.

Welcome to subscribe "Shulou Technology Information " to get latest news, interesting things and hot topics in the IT industry, and controls the hottest and latest Internet news, technology news and IT industry trends.