Preventing oscillatory load behavior in a multi-node distributed system

摘要

The disclosed embodiments provide a system that prevents oscillatory load behavior for a multi-node distributed system. During operation, the system uses a load-balancing policy to distribute requests to nodes of the distributed system. The system determines operational characteristics for the nodes as they process a set of requests, and then uses these operational characteristics to compute machine queuing models that describe the machine state of each node. The system then uses this machine state for the nodes to determine whether the load-balancing policy and the distributed system are susceptible to oscillatory load behavior.

主权项

1. A method comprising:
using a load-balancing policy to distribute requests to nodes of a multi-node distributed system; determining operational characteristics for two or more nodes of the multi-node distributed system as the two or more nodes process a set of requests, wherein determining operational characteristics for a node comprises tracking an incoming request rate and a request service rate for the node; using the operational characteristics to compute machine queuing models that describe a machine state of the two or more nodes; and determining, by computer, from the machine state for the nodes whether the load-balancing policy and the multi-node distributed system are susceptible to oscillatory load behavior.