Modifying network site behavior using session-level performance metrics

摘要

Disclosed are various embodiments for modifying network site behavior. At least one session-level performance metric associated with a client is determined. The one or more session-level performance metrics are determined from one or more latency times. Each one of the latency times represents a time elapsed between a sending of a network page request in the client and a rendering in the client of a network page received from a network page server in response to the network page request. A response to a next network page request from the client is modified according to the one or more session-level performance metrics in order to adjust a next latency time for the client.

主权项

1. A system, comprising:
at least one computing device; and at least one application executable in the at least one computing device, wherein when executed the at least one application causes the at least one computing device to at least:
determine a session-level latency performance observed by a client across a session of accessing a network site, the session-level latency performance being determined from a plurality of latency times, individual ones of the plurality of latency times representing a time elapsed between a sending of a network page request in the client and a rendering in the client of a network page received from a network page service in response to the network page request;modify a response to a subsequent network page request from the client according to the session-level latency performance in order to adjust a subsequent latency time for the client;wherein modifying the response further comprises selecting a different network page service to respond to the subsequent network page request received from the client, the different network page service being selected according to the session-level latency performance in order to adjust the subsequent latency time for the client; and wherein the different network page service is associated with a slower response time than a network page service that otherwise would have been selected to respond to the subsequent network page request.