watermark,size_16,text_QDUxQ1RP5Y2a5a6i,color_FFFFFF,t_100,g_se,x_10,y_10,shadow_90,type_ZmFuZ3poZW5naGVpdGk=

watermark,size_16,text_QDUxQ1RP5Y2a5a6i,color_FFFFFF,t_100,g_se,x_10,y_10,shadow_90,type_ZmFuZ3poZW5naGVpdGk=


Q           The title is as follows :

stay Oracle in , What is high availability ?


     
A           The answer is as follows :          


High availability (High Availability,HA) It is one of the factors that must be considered in the design of distributed system architecture .HA Usually refer to , Reduce the time that the system cannot provide service by design . Assume that the system is always able to provide services , Then we can say that the availability of the system is 100%. If every time the system runs 100 Time units , There will be 1 Time units can't provide services , So the availability of the system is 99%. Many companies ( For example, three operators 、 Baidu 、 Jingdong etc. ) Our high availability goals are 4 individual 9, That is to say 99.99%.

Baidu's search home page , Is the industry recognized high availability guarantee very excellent system , Even people will pass through www.baidu.com Can you visit to judge “ Network connectivity ”. Baidu's highly available services leave a legacy “ The Internet is clear , Baidu can visit ”,“ Baidu can't open , It should be that the Internet can't be connected ” The impression of , This can be said to Baidu HA The highest praise .

as everyone knows , Single point is the biggest risk and enemy of system high availability , We should try to avoid single point in the process of system design . In methodology , The principle of high availability assurance is “ clustering ”, Or call it “ redundancy ”. There is only one single point , Hung up service will be affected ; If there are redundant backups , There are other backups that can continue to provide services when they are down . and RAC Provides instance level redundancy ,DG Provides data storage level redundancy .

To ensure high availability of the system , The core criterion of architecture design is : redundancy . With redundancy , Is not enough , Every time there is a fault, the need for human intervention recovery is bound to increase the system's unserviceable practice . therefore , And often through “ Automatic failover ” To achieve high availability of the system .