System and method for improving an aggregated throughput of simultaneous connections
A client and content provider are connected by a plurality of simultaneous transport connections. The number of the transport connections that are used to transfer data is selected based on the size of the data to be transferred and may change after transfer of data has commenced based on the amount of data left and the attributes of the transport connections. In another aspect, data to be transmitted over the transport connections is organized into frames such that each frame includes data from only one data stream. The frames are sized to be less than or equal to a control window of the transport connection over which they are transmitted. Each frame may be assigned to a transport connection in a round robin fashion or based on the size of the frame and the sizes of the control windows of the transport connections.
1. A method for increasing an aggregated throughput
comprising:
establishing, by one of a first computer and a second computer, a plurality of simultaneous transport connections between the first computer and the second computer over a computer network;
receiving, by the first computer from the second computer, two or more groups of requests for data delivery to the second computer through the plurality of simultaneous transport connections with the first computer such that a first group of requests the two or more groups of requests is constrained to be finished before a second group of requests of the two or more groups of requests;
determining, by the first computer, that delivery of requested data corresponding to each request for data delivery of the first group of requests is finished;
setting, by the first computer, a number of the plurality of simultaneous transport connections used to deliver requested data corresponding to each request for data delivery of the second group of requests to the first computer such that a number of simultaneous transport connections used to deliver the requested data corresponding to each request for data delivery of the first group of requests is set to be different than the number of the plurality of simultaneous transport connections used to deliver the requested data corresponding to each request for data delivery of the second group of requests;
wherein the second computer is a client computer, the method further comprising:
opening, by the client computer, one or more client transport connections to request and receive requested data corresponding to each request of each group of the two or more groups of requests through the plurality of simultaneous transport connections opened between the second computer and the first computer;
closing, by the client computer, the one or more client transport connections without invoking closing of the plurality of simultaneous transport connections between the first and the second computers; and
wherein the method further comprises;
determining that delivery of the requested data corresponding to each request for data delivery of a same group of the two or more groups of requests is finished comprises determining that the client computer has closed the one or more client transport connections; and
changing the number of the plurality of simultaneous transport connections used to deliver the requested data corresponding to a same group of requests of the two or more groups of requests by at least one of pausing and resuming delivery of at least some of the requested data corresponding to the same group of requests through one or more previously opened simultaneous connections, without closing existing connections or opening new connections.
2. The method of claim 1 , wherein each transport connection of the plurality of simultaneous transport connections uses a transport protocol that includes generating one or more acknowledgements to confirm delivery of the requested data corresponding to each request for data delivery of the first and second groups of requests; and
wherein the method further comprises determining that delivery of requested data corresponding to each request for data delivery of a same group of the two or more groups of requests is finished in response to receiving the one or more acknowledgements that the requested data corresponding to the each request in the same group is delivered.
3. The method of claim 2 , where each transport connection of the plurality of simultaneous transport connections uses a Transport Control Protocol (TCP).
4. The method of claim 1 , wherein the second computer hosts a proxy programmed to terminate the one or more client transport connections.
5. The method of claim 1 , wherein at least two of the plurality of simultaneous transport connections are opened over a same physical data link.
6. The method of claim 1 , wherein at least two of the plurality of simultaneous transport connections have identical source and destination network addresses.
7. The method of claim 1 , further comprising:
determining, by one of the first computer and the second computer, an estimated size of data delivery for at least one group of the two or more groups of requests; and
setting, by one of first computer and the second computer, a number of simultaneous network connections used to transport requested data corresponding to requests of a same group of requests of the two or more groups of requests such that the number of simultaneous transport connections used to transport the requested data corresponding to the requests of the same group of requests increases with increase in the estimated size.
8. The method of claim 7 , wherein the estimated size is determined by obtaining one or more network identifiers from the requests of the same group of requests, each identifier comprising at least one of a domain name and a destination network address.
9. The method of claim 1 , wherein the number of the plurality of simultaneous transport connections used to deliver the requested data corresponding to each request of the second group of requests is decreased after delivery of the requested data corresponding to the first group of requests is followed by a period of inactivity exceeding a pre-defined time interval before the second group of requests is submitted.
10. The method of claim 1 , further comprising changing the number of the plurality of simultaneous transport connections used to deliver the requested data corresponding to each request for data delivery of the second group of requests after completing delivery of the requested data corresponding to each request of the first group of requests and before receiving the second group of requests.
11. The method of claim 1 , further comprising setting a number of the plurality of simultaneous transport connections used for data transfer from the first computer to the second computer to be different from a number of the plurality of simultaneous transport connections used for data transfer from the second computer to the first computer.
12. The method of claim 11 , further comprising changing the number of the plurality of simultaneous transport connections used for the data transfer from the first computer to the second computer while maintaining constant the number of the plurality of simultaneous transport connections used for the data transfer from the second computer to the first computer.
13. The method of claim 1 , where the first computer comprises at least one of a proxy server and a virtual private network (VPN) server.