System and method for automatic and adaptive use of active network performance measurement techniques to find the fastest source

• US 7,827,257 B2
• Filed: 06/19/2001
• Issued: 11/02/2010
• Est. Priority Date: 06/19/2001
• Status: Expired due to Fees

• Current Assignee: Intel Corporation
• Original Assignee: Intel Corporation
• Inventors: Chu, Jim, Hady, Frank T.
• Primary Examiner(s): Lee; Philip C
• Application Number:
  US09/884,674
• Publication Number:
  US 20020194310A1
• Time in Patent Office: 3,423 Days
• Field of Search: 709/223, 709/227, 709/200, 709/241, 709/201, 707/2, 725/13, 395/200.63
• US Class Current: 709/223
• CPC Class Codes: H04L 41/5009 Determining service level performance parameters or violations of service level contracts, e.g. violations of agreed response time or mean time between failures [MTBF]
• CPC Class Codes: H04L 43/00 Arrangements for monitoring or testing data switching networks
• CPC Class Codes: H04L 43/0852 Delays
• CPC Class Codes: H04L 43/0888 Throughput
• CPC Class Codes: H04L 67/1001 for accessing one among a plurality of replicated servers
• CPC Class Codes: H04L 67/1008 based on parameters of servers, e.g. available memory or workload monitoring of computer activity G06F11/30
• CPC Class Codes: H04L 67/101 based on network conditions
• CPC Class Codes: H04L 67/1012 based on compliance of requirements or conditions with available server resources

Abstract

Systems and methods are provided through which automatic and adaptive use of active network performance measurement techniques identifies a fastest download source. Before a download source is selected, empirical measurements of the download speed are performed. For small files, the first source to acknowledge an open socket connection is used as an indication of the fastest download source. For large files, a download test is performed as the empirical measurement to determine the fastest download source.

Claims

1. A method for managing a plurality of sources comprising:

selecting a type of empirical measurement of performance from a plurality of types of empirical measurement, wherein the type of empirical measurement is selected according to a comparison of a size of data to be obtained from at least one of a plurality of sources with a threshold size;

selecting a download source from the plurality of sources, the selection made according to empirical measurements of performance having the selected type for each of the plurality of sources; and

initiating a download of data from the selected download source of the plurality of sources.

2. The method of claim 1, further comprising:

obtaining an empirical measurement of performance of each of the plurality of sources from a local source.

3. The method of claim 1, wherein the performance further comprises a throughput speed.

4. The method of claim 1, wherein the performance comprises latency.

5. The method of claim 2, wherein obtaining the empirical measurement further comprises:

measuring the elapsed time of a transmission involving each of the plurality of sources.

6. The method of claim 2, wherein obtaining the empirical measurement further comprises for each of the plurality of sources:

recording transmission time from the current time and date;

initiating a transmission to a download source of the plurality of sources;

receiving a response to the transmission from the source;

recording the receipt time from the current date and time; and

determining the throughput speed of the source from the difference between the receipt time and the transmission time.

23. The method of claim 1, wherein the threshold size is determined according to a communications bandwidth of a computer system initiating the download of data.

7. A non-transitory computer-readable storage medium having stored thereon executable instructions for managing a plurality of sources, said executable instructions capable of directing a processor to perform:

selecting a type of empirical measurement of performance from a plurality of types of empirical measurement, wherein the type of empirical measurement is selected according to a comparison of a size of data to be obtained from at least one of a plurality of sources with a threshold size;

selecting a download source in reference to the empirical measurements of the throughput speed of each of the plurality of sources; and

initiating a download of data from a download source of the plurality of sources.

8. The non-transitory computer-readable storage medium of claim 7, wherein the throughput speed further comprises a download speed.

9. The non-transitory computer-readable storage medium of claim 7, wherein said instructions for selecting a type of empirical measurement further comprises an instruction capable of directing the processor to perform:

measuring a throughput speed of each of the plurality of sources.

10. The non-transitory computer-readable storage medium of claim 9, wherein said instructions for measuring further comprises instructions capable of directing the processor to perform for each of the plurality of sources:

recording transmission time from the current time and date;

initiating a transmission to a download source of the plurality of sources;

receiving a response to the transmission from the source;

recording the receipt time from the current date and time; and

determining the throughput speed of the source from the difference between the receipt time and the transmission time.

11. The non-transitory computer-readable storage medium of claim 7, wherein the download source further comprises a source in a peer-to-peer network.

12. The non-transitory computer-readable storage medium of claim 7, wherein said instructions for selecting a type of empirical measurement further comprises instructions capable of directing the processor to perform:

recording transmission time from the current time and date;

initiating a transmission to a download source of the plurality of sources;

receiving a response to the transmission from the source;

recording the receipt time from the current date and time; and

determining the round-trip timing of the download source from the difference between the receipt time of the response and the transmission time of the transmission.

24. The non-transitory computer-readable storage medium of claim 7, wherein the threshold size is determined according to a communications bandwidth of a computer system initiating the download of data.

13. A computerized method for managing a plurality of sources comprising:

obtaining a list comprising a plurality of identification of sources;

initiating a plurality of socket connections, the plurality of socket connections further comprising one socket connection for each of the plurality of sources, yielding a plurality of initiated socket connections;

selecting a type from a plurality of types of empirical measurement of performance of each of the plurality of sources, the type of empirical measurement selected according to a comparison of a predetermined file size with a predetermined threshold file size;

receiving a response for the each of the plurality of initiated socket connections, yielding a plurality of responses;

selecting a download source of the plurality of sources in reference to an empirical measurement of performance having the selected type; and

initiating a download of data from the selected download source of the plurality of sources.

14. The computerized method of claim 13, wherein the predetermined file size is less than the predetermined threshold file size and wherein the type of empirical measurement of performance comprises a first response and wherein the selecting further comprises:

selecting the source associated with the response that is received first.

15. The computerized method of claim 13, wherein the predetermined file size is greater than the predetermined threshold file size and wherein the type of empirical measurement of performance comprises a latency and wherein the selecting further comprises:

measuring the latency of each of the plurality of sources; and

selecting a source in reference to the download speed of each of the plurality of sources.

16. The computerized method of claim 15, wherein measuring the latency further comprises:

storing a time and date of each of the plurality of initiating socket connections;

storing the time and date of each of the plurality of responses; and

determining the download speed of each of the plurality of sources from the differences in time between the time and date of each of the plurality of the responses and the time and date of each of the plurality of the initiating socket connections.

25. The computerized method of claim 13, wherein the predetermined threshold file size is determined according to a communications bandwidth of a computer system initiating the download of data.

17. A system for managing sources in a peer-to-peer network comprising:

a hardware processor;

software means operative on the processor for selecting a type of empirical measurement of a plurality of types of measurements of throughput speed of each of the plurality of sources, the type of empirical measurement selected according to a comparison of a size of data to be obtained from at least one of the plurality of sources with a threshold size;

the software means including obtainer means to obtain an empirical measurement having the selected type, the empirical measurement comprising of a throughput speed of each of the plurality of sources from at least one third-party source;

the software means selecting a download source in reference to the empirical measurements of the throughput speed having the selected type for each of the plurality of sources and the at least one third-party source; and

a transmitter to initiate a download of data from the download source of the plurality of sources.

18. The system of claim 17, wherein the throughput speed further comprises a round-trip time.

19. The system of claim 17, wherein the throughput speed further comprises a latency.

26. The system of claim 17, wherein the threshold size is determined according to a communications bandwidth of the system.

20. A computerized system comprising:

a hardware processor;

a determiner of empirical measurements of a throughput speed of each of a plurality of download peer-to-peer network sources, the empirical measurements having a type;

the determiner operable to select a type of empirical measurement and to select an empirical measurement of a throughput speed having the selected type for each of the plurality of sources from at least one third-party source, the type of empirical measurement selected according to a comparison of a size of data to be obtained from at least one of the plurality of download peer-to-peer network sources with a threshold size;

a selector of a download source in reference to the empirical measurement of the throughput speed of each of the plurality of peer-to-peer network sources and the at least one third-party source; and

a transmitter to initiate a download of data from the download source of the plurality of peer-to-peer network sources.

21. The computerized system of claim 20, the determiner further comprising:

a transmitter to transmit a message to a download source of the plurality of sources;

a recorder of the time of a transmission of a message, operably coupled to the transmitter;

a receiver of a response to the transmission from the source, operably coupled to the transmitter;

a recorder of the time of receipt of a response; and

a determiner of the throughput speed of the source, from the difference between the receipt time and the transmission time.

22. The computerized system of claim 20, wherein:

the message further comprises a TCP/IP synchronized idle message; and

the response further comprises a TCP/IP acknowledgment message.

27. The computerized system of claim 20, wherein the threshold size is determined according to a communications bandwidth of the computerized system.