Dynamic capacity optimization for shared computing resources

Systems, methods, devices, and other techniques for managing computing resources shared by a set of online entities. A system may receive, from a first online entity, a request to reserve a capacity of a computing resource. The system determines a relative priority of the first online entity and identifies a reserved area corresponding to the relative priority of the first online entity. The system determines whether the request is satisfied based on comparing (i) a requested capacity of the computing resource and (ii) a portion of the unused capacity of the computing resource specified by the reservation area that is allowed to be reserved by an online entity having a relative priority equal to or lower than the relative priority of the first online entity. 用于管理由一组在线实体共享的计算资源的系统，方法，设备和其他技术。系统可以从第一在线实体接收预留计算资源的容量的请求。该系统确定第一在线实体的相对优先级，并识别与第一在线实体的相对优先级相对应的预留区域。系统基于比较(i)所述计算资源的请求容量以及(ii)允许具有相对优先级等于或低于所述第一在线实体的所述相对优先级的在线实体预留的、由所述预留区域指定的所述计算资源的所述未使用容量的一部分，来确定是否满足所述请求。