| 摘要 |
<p>A large-scale cluster management method, device and system, which can conduct performance management and resource scheduling on a user according to a service level, thereby improving the user experience. The method comprises: in management objects corresponding to the first service level of a plurality of service levels, determining at least one management object, wherein the management object is a resource unit in a large-scale cluster; determining the target performance of the at least one management object; acquiring the actual performance of the at least one management object; and according to the target performance and the actual performance, conducting performance management on the management objects corresponding to the first service level. By determining at least one management object in the management objects corresponding to the first service level in the large-scale cluster, and conducting performance management on all the management objects corresponding to the first service level according to the target performance and the actual performance of the at least one management object, it can be guaranteed that the performance of most or even all of users reaches the target performance, thereby improving the user experience.</p> |
