Method and system FPOR transferring data to improve responsiveness when sending large data sets

• US 9,509,802 B1
• Filed: 03/15/2013
• Issued: 11/29/2016
• Est. Priority Date: 03/15/2013
• Status: Active Grant

• Current Assignee: PME IP Pty., Ltd.
• Original Assignee: PME IP Pty., Ltd.
• Inventors: Stalling, Detlev, Westerhoff, Malte
• Primary Examiner(s): Avellino, Joseph E
• Assistant Examiner(s): NGANKAM, PATRICK F
• Application Number:
  US13/831,982
• Time in Patent Office: 1,355 Days
• Field of Search: 709/203, 370/230, 370/235, 370/401
• US Class Current: 1/1
• CPC Class Codes: H04L 43/0894 Packet rate
• CPC Class Codes: H04L 45/121 by minimising delays
• CPC Class Codes: H04L 45/125 based on throughput or bandwidth
• CPC Class Codes: H04L 47/10 Flow control; Congestion control
• CPC Class Codes: H04L 47/12 Avoiding congestion; Recovering from congestion
• CPC Class Codes: H04L 47/127 by using congestion prediction
• CPC Class Codes: H04L 47/27 Evaluation or update of window size, e.g. using information derived from acknowledged [ACK] packets
• CPC Class Codes: H04L 67/01 Protocols
• CPC Class Codes: H04L 69/04 Protocols for data compression, e.g. ROHC
• CPC Class Codes: H04L 69/163 In-band adaptation of TCP data exchange; In-band control procedures

Abstract

Most of the internet traffic today is carried out via the Transmission Control Protocol (TCP). The main advantage of TCP is that it provides reliable data transfer to the application layer and simplifies programming. The protocol maximizes data throughput but may also lead to noticeable transmission delay in wide area networks (WAN). A client-server based medical image viewing system is disclosed that achieves high data throughput over TCP without impacting responsiveness. Special timestamp messages inserted into the data stream allow the system to detect situations where network latency increases noticeably and to obtain a reliable estimate of sustained transfer bandwidth. The system applies a feedback scheme that avoids network delays by limiting send bandwidth. In addition other parameters, in particular image compression settings, are dynamically adjusted depending on current network quality.

Claims

1. A method of minimizing network latency comprising the steps of:

sending a request for an image data file from a client computer, including inserting one or more first timestamp messages into the request for the image data file at an application level;

transferring the request and the one or more first timestamp messages to a server using a standard Transmission Communications Protocol (TCP) connection;

receiving the one or more first timestamp messages at the server, where a timestamp is inserted into the one or more first timestamp messages;

calculating one or more first times for the one or more first timestamp messages and a current send bandwidth (b_send) in bytes/msec;

sending one or more second timestamp messages from the server to the client computer, where a timestamp is inserted into the one or more second timestamp messages;

sending the one or more second timestamp messages from the client computer to the server;

receiving the one or more second timestamp messages from the client computer at the server;

calculating one or more second times for the one or more second timestamp messages, where the one or more second timestamp messages are used to compute read bandwidth (b_read), where b_read=C/T, where C=c_i−

c<sub>i−

1</sub>, where C is an amount of data that was read in a time between a last two timestamp messages (c_i, c<sub>i−

1</sub>), and T=t_i−

t<sub>i−

1</sub>+d_i−

d<sub>i−

1</sub> is the time elapsed on the client computer between the last two timestamp messages, where t_i−

t<sub>i−

1</sub> is the time elapsed on the server between the last two timestamp messages, and d_i and d<sub>i−

1</sub> are differences between client computer time and server time when the messages arrived at the client computer;

calculating a best estimate of network bandwidth (b_est) in bytes/sec based on one or more of the one or more first times and the one or more second times;

determining situations where network latency in sending the image data file based on one or more of b_send, b_read and b_est has increased; and

using one or more of b_send, b_read and b_est to one or both reduce and modify client computer requests to minimize network latency.

2. The method of minimizing network latency of claim 1, where the b_est is calculated from a difference between the one or more first times and the one or more second times.

3. The method of minimizing network latency of claim 2, further comprising determining a running average.

4. The method of minimizing network latency of claim 1, where samples of b_read are included in the running average.

5. The method of minimizing network latency of claim 4, where samples are included in the running average when b_read is greater than b_est.

6. The method of minimizing network latency of claim 5, where samples are included in the running average when b_read is greater than 1.4*b_est.

7. The method of minimizing network latency of claim 5, where a bandwidth limit (b_limit) is applied on the server in order to avoid network delays, where the b_limit is computed using a feedback scheme.

8. The method of minimizing network latency of claim 7, where bandwidth is not reduced if there are less than 10 KB of data in the line (bytes_inline), where bytes_inline can be estimated by c−

c_i, where c is the current total number of bytes that were sent to the client at the time and c_i is the total number of bytes that were sent to the client at the time the current timestamp message was sent.

9. The method of minimizing network latency of claim 8, where the feedback scheme uses a pseudo code which includes an expression

if (e>

e_max)

then

if (bytes_inline>

e_max)

then

b_limit;

=max(Factor1*b_est, Factor2*b_limit)

end

else

if (b_read>

Factor3

then

d;

=Factor4 (e_max−

e)/e_max

b_limit;

=b_limit+d

end.

10. The method of minimizing network latency of claim 9, where the b_limit is used to compute a lossy compression rate.

11. The method of minimizing network latency of claim 10, where the lossy compression rate is calculated in order to achieve a desired interactive speed.

12. The method of minimizing network latency of claim 11, where the lossy compression rate is used to stream compressed images with a compression ratio.

13. The method of minimizing network latency of claim 12, where images are streamed to the client computer using a buffering system.

14. The method of minimizing network latency of claim 13, where the buffering system is based on monitoring user interaction and anticipating a next image that will be requested by the client computer.

15. The method of minimizing network latency of claim 14, where the b_limit is used to compute the lossy compression rate.

16. The method of minimizing network latency of claim 15, where the lossy compression rate is calculated in order to achieve a desired interactive speed.

17. The method of minimizing network latency of claim 16, where the lossy compression rate is used to calculate a compression ratio, where one or more compressed images are streamed with the compression ratio.

18. The method of minimizing network latency of claim 16, where a target compression quality is defined by a user, where a first image is streamed with a first compression quality, where the first compression quality minimizes network latency during interaction based on bandwidth monitoring and where the first image is streamed with a second compression quality when the user stops interacting, where the second compression quality is greater than the first compression quality if the first compression quality is lower than a target compression quality.

19. A method of identifying network latency comprising the steps of:

sending a request for an image file from a client computer, including inserting a first timestamp message into the request for the image data file at an application level;

transferring the request and the first timestamp message to a server using a standard Transmission Communications Protocol (TCP) connection;

receiving the first timestamp message at the server, where a timestamp is inserted into the first timestamp message;

calculating a first time for the first timestamp message to be sent from the client computer to the server and a current send bandwidth (b_send) in bytes/msec;

sending one or more second timestamp messages from the server to the client computer, where a timestamp is inserted into the second timestamp message;

returning the one or more second timestamp messages from the client computer to the server;

receiving the one or more second timestamp messages from the client computer at the server;

calculating a second time and subsequent times taken for the one or more second timestamp messages to be sent from the client computer to the server, where the one or more second timestamp messages are used to compute read bandwidth (b_read), where b_read=C/T, where C=c_i−

c<sub>i−

1</sub>, where C is an amount of data C that was read in a time between a last two timestamp messages (c_i, c<sub>i−

1</sub>), and T=t_i−

t<sub>i−

1</sub>+d_i−

d<sub>i−

1</sub> is the time elapsed on the client computer between the last two timestamp messages where t_i−

t<sub>i−

1</sub> is the time elapsed on the server between the last two timestamp messages and d_i and d<sub>i−

1</sub> are differences between client computer time and server time when the messages arrived at the client computer;

calculating a best estimate of network bandwidth (b_est) based on one or more of the first time, the second time and the subsequent times; and

using one or more of b_send, b_read and b_est to determine network latency in sending the image data file.

20. The method of identifying network latency of claim 19, where the estimate of current network bandwidth is calculated from a difference between the first time and the second time or the second time and the subsequent times.