PERFORMANCE-DRIVEN RESOURCE MANAGEMENT IN A DISTRIBUTED COMPUTER SYSTEM

摘要

A system and method for managing resources in a distributed computer system that includes at least one resource pool for a set of virtual machines (VMs) utilizes a set of desired individual VM-level resource settings that corresponds to target resource allocations for observed performance of an application running in the distributed computer system. The set of desired individual VM-level resource settings are determined by constructing a model for the observed application performance as a function of current VM-level resource allocations and then inverting the function to compute the target resource allocations in order to meet at least one user-defined service level objective (SLO). The set of desired individual VM-level resource settings are used to determine final RP-level resource settings for a resource pool to which the application belongs and final VM-level resource settings for the VMs running under the resource pool, which are then selectively applied.

主权项

1. A method for managing resources in a distributed computer system that includes at least one resource pool (RP) for a set of clients, the method comprising:
computing a set of target resource allocations for the clients running an application based on observed performance of the application in order to meet at least one service level objective (SLO); determining a set of desired individual client-level resource settings corresponding to the target resource allocations, the set of desired individual client-level resource settings including desired reservation and limit values for the clients; computing final RP-level resource settings for a resource pool to which the application belongs and final client-level resource settings for the clients running under the resource pool based on at least the set of desired individual client-level resource settings, current RP-level resource settings of the resource pool and available resource capacity of the resource pool, the final client-level and RP-level resource settings including final reservation and limit values for the clients and the resource pool; and applying at least one of the final RP-level resource settings and the final client-level resource settings to at least one of the resource pool and the clients running under the resource pool.