| 摘要 |
A method, program, and server for estimating the bandwidth of a network connection between a client and a server includes requesting the server to serve first and second objects back-to-back to the client. The first and second objects are sent to the client. The client determines the time interval between delivery of the first and second objects. The time interval is used, in conjunction with information about the size of the second object, to estimate the bandwidth. The requests for the first and second objects preferably identify the first and second objects with URL's that are unique on the network to prevent the request from being serviced by a file cache. The first and second objects may be transmitted to the client from a content distribution network server that is architecturally close to the client's ISP to improve the reliability of the bandwidth estimation.
|
