such as the title.
what"s the difference between purchasing SLB services and writing upstream yourself? can"t all meet the needs of load balancer?
or buying a SLB service is like a stupid one-click operation. Omitting the process of writing code?
the final result is the same, except that Aliyun cloud load balancer provides a configuration operation interface, which is simpler and easier to manage, just like Aliyun security group and server configure firewall ports
I think the most critical difference is that Aliyun's cloud load balancer promises high availability. If you write a upstream by hand, it should still belong to the category of single point deployment, that is, you cannot guarantee the availability of the load balancer node itself. To put it bluntly, once this node dies, all the services behind it will die collectively. But if you deploy multiple points to ensure high availability, the cost will be higher than the cost of using SLB. Of course,
also has relatively intuitive differences such as visualization, which I won't repeat.
the company has a very basic, very important function. Now want to turn it into a service, so the question is how to make the service highly available. Because the foundation is important, it must not be hung up because it is important, and there must be...
there are three servers, one is a Nginx reverse proxy server, and it is also a static server, and two are web servers. Now, if you use a proxy server to visit a website, it is found that it takes longer to access a website than to visit a web server dir...
when we want to use nginx to manage static resources, start nginx on the 10.xx.xxx.xx machine (the code part is part of the nginx configuration) where server_name can only be the address of the machine 10.xx.xxx.xx if the 10.xx.xxx.xx machine does not h...
problem description I have a back-end service that takes time to start, and it takes about 5 minutes from process startup to normal access. The slow_start parameter of the nginx reverse proxy can be buffered for a period of time after the back-end ser...