Method and apparatus to evaluate and measure the optimal network packet size for file transfer in high-speed networks

Claim 3

The method of claim 1 , wherein the optimal packet size is identified based on a policy.

Claim 4

The method of claim 2 , wherein the policy is to identify the optimal packet size based on a selected event.

Claim 5

The method of claim 3 , wherein the selected event is a periodic event.

Claim 6

The method of claim 3 , wherein the selected event is an initiation of a data transfer to the target.

Claim 7

The method of claim 1 further comprising: sending the data to the target using the optimal packet size.

Claim 8

The method of claim 1 , wherein the service times for the first set of packets and the service times for the second set of packets are measured for a slowest processing node in a path between the data processing system and the target.

Claim 9

The method of claim 1 further comprising: determining an average round trip time and an average service time for the first packet size and an average round trip time and an average service time for the second packet size using the round trip times and the service times for the set of transferred packets and wherein the selecting step comprises selecting the optimal packet size as: p ~ = ( aW - d ) + ( aW - d ) 2 + 4  bcW 2  c wherein {tilde over (p)} is the optimal packet size, a = S  ( p 1 ) - S  ( p 2 ) p 1 - p 2 , b=S(p 1 )−ap 1 , c = RTT  ( p 1 ) - RTT  ( p 2 ) p 1 - p 2 , d=RTT(p 1 )−cp 1 , p 1 is the first packet size, p 2 is the second packet size, RTTP(p 1 ) is the average round trip time for the first packet size, RTTP(p 2 ) is the average round trip time for the second packet size, S(p 1 ) is the average service time for the first packet size, S(p 2 ) is the average service time for the second packet size.

Claim 10

A data processing system for identifying an optimal packet size for transferring data to a target, the data processing system comprising: first selecting means for selecting a first packet size; first sending means for sending a first set of packets to the target using the first packet size; first measuring means for measuring round trip times and service times for the first set of packets; second selecting means for selecting a second packet size; second sending for sending a second set of packets to the target using the second packet size; second measuring means for measuring round trip times and service times for the second set of packets; and identifying means for identifying the optimal packet size based on the round trip times and service times for the first set of packets and the round trip times and service times for the second set of packets.

Claim 11

A data processing system for identifying an optimal packet size for transferring data to a target, the data processing system comprising: first selecting means for selecting a first packet size; first sending means for sending a first set of packets to the target using the first packet size; first measuring means for measuring round trip times and service times for the first set of packets; second selecting means for selecting a second packet size; second sending for sending a second set of packets to the target using the second packet size; second measuring means for measuring round trip times and service times for the second set of packets; and identifying means for identifying the optimal packet size based on the round trip times and service times for the first set of packets and the round trip times and service times for the second set of packets.

Claim 12

The data processing system of claim 9 , wherein the optimal packet size is identified based on a policy.

Claim 13

The data processing system of claim 10 , wherein the policy is to identify the optimal packet size based on a selected event.

Claim 14

The data processing system of claim 11 , wherein the selected event is a periodic event.

Claim 15

The data processing system of claim 11 , wherein the selected event is an initiation of a data transfer to the target.

Claim 16

The data processing system of claim 9 further comprising: third sending means for sending the data to the target using the optimal packet size.

Claim 17

The data processing system of claim 9 , wherein the service times for the first set of packets and the service times for the second set of packets are measured for a slowest processing node in a path between the data processing system and the target.

Claim 18

The data processing system of claim 9 further comprising: determining means for determining an average round trip time and an average service time for the first packet size and an average round trip time and an average service time for the second packet size using the round trip times and the service times for the set of transferred packets and wherein the selecting step comprises selecting the optimal packet size as: p ~ = ( aW - d ) + ( aW - d ) 2 + 4  bcW 2  c wherein {tilde over (p)} is the optimal packet size, a = S  ( p 1 ) - S  ( p 2 ) p 1 - p 2 , b=S(p 1 )−ap 1 , c = RTT  ( p 1 ) - RTT  ( p 2 ) p 1 - p 2 , d=RTT(p 1 )−cp 1 , p 1 is the first packet size, p 2 is the second packet size, RTTP(p 1 ) is the average round trip time for the first packet size, RTTP(p 2 ) is the average round trip time for the second packet size, S(p 1 ) is the average service time for the first packet size, S(p 2 ) is the average service time for the second packet size.

Claim 19

A data processing system comprising: a communications interface configured for connection to a network, wherein the communications interface is configured to send and receive packets; and a processing unit, wherein the processing unit executes instructions to select a first packet size and a second packet size, sends a set of packets using the first packet size and the second packet size to a target data processing system to form a set of transferred packets, identifies round trip times and service times for the set of transferred packets, and selects an optimal packet size using the round trip times and service times for the set of packets.

Claim 20

A data processing system comprising: a communications interface configured for connection to a network, wherein the communications interface is configured to send and receive packets; and a processing unit, wherein the processing unit executes instructions to select a first packet size and a second packet size, sends a set of packets using the first packet size and the second packet size to a target data processing system to form a set of transferred packets, identifies round trip times and service times for the set of transferred packets, and selects an optimal packet size using the round trip times and service times for the set of packets.

Claim 21

The data processing system of claim 17 , wherein the set of packets are sent using an Internet Protocol.

Claim 22

The data processing system of claim 18 , wherein the instructions are part of an Internet Protocol layer used to send the packets to the target data processing system.

Claim 23

The data processing system of claim 17 , wherein the data processing system selects the optimal packet size as: p ~ = ( aW - d ) + ( aW - d ) 2 + 4  bcW 2  c wherein {tilde over (p)} is the optimal packet size, a = S  ( p 1 ) - S  ( p 2 ) p 1 - p 2 , b=S(p 1 )−ap 1 , c = RTT  ( p 1 ) - RTT  ( p 2 ) p 1 - p 2 , d=RTT(p 1 )−cp 1 , p 1 is the first packet size, p 2 is the second packet size, RTTP(p 1 ) is the average round trip time for the first packet size determined from the round trip times for the set of transferred packets having the first packet size, RTTP(p 2 ) is the average round trip time for the second packet size determined from the round trip times for the set of transferred packets having the second packet size, S(p 1 ) is the average service time for the first packet size determined from the service times for the set of transferred packets having the first packet size, and S(p 2 ) is the average service time for the second packet size determined from the service times for the set of transferred packets having the second packet size.

Claim 24

The data processing system of claim 17 , wherein the service times are identified for a slowest processing node in a path between the data processing system and the target.

Claim 27

The method of claim 22 , wherein the selecting, sending, and identifying steps are based on a policy.

Claim 28

The method of claim 23 , wherein the policy is to identify the optimal packet size based on a selected event.

Claim 29

The method of claim 24 , wherein the selected event is a periodic event.

Claim 30

The method of claim 24 , wherein the selected event is an initiation of a data transfer to the target.

Claim 31

The method of claim 22 further comprising: determining an average round trip time and an average service time for the first packet size using the round trip times and the service times for the set of transferred packets; determining an average round trip time and an average service time for the second packet size using the round trip times and the service times for the set of transferred packets and wherein the selecting step comprises selecting the optimal packet size as: p ~ = ( aW - d ) + ( aW - d ) 2 + 4  bcW 2  c wherein {tilde over (p)} is the optimal packet size, a = S  ( p 1 ) - S  ( p 2 ) p 1 - p 2 , b=S(p 1 )−ap 1 , c = RTT  ( p 1 ) - RTT  ( p 2 ) p 1 - p 2 , d=RTT(p 1 )−cp 1 , p 1 is the first packet size, p 2 is the second packet size, RTTP(p 1 ) is the average round trip time for the first packet size, RTTP(p 2 ) is the average round trip time for the second packet size, S(p 1 ) is the average service time for the first packet size, S(p 2 ) is the average service time for the second packet size.

Claim 32

The method of claim 22 , wherein the service times are identified for a slowest processing node in a path between the data processing system and the target.

Claim 33

The method of claim 22 , wherein the sending, identifying, and selecting steps are performed in an Internet Protocol layer in the data processing system.

Claim 34

The method of claim 22 , wherein the data is sent in packets over a high-speed network.

Claim 35

The method of claim 22 , wherein the high-speed network is a asynchronous mode transfer network.

Claim 36

A distributed data processing system comprising: a network; a first data processing system; and a second data processing system, wherein the second data processing system selects a first packet size and a second packet size, sends a set of packets using the first packet size and the second packet size to the first data processing system to form a set of transferred packets, identifies round trip times and service times for the set of transferred packets, and selects an optimal packet size using the round trip times and the service times for the set of packets.

Claim 37

A distributed data processing system comprising: a network; a first data processing system; and a second data processing system, wherein the second data processing system selects a first packet size and a second packet size, sends a set of packets using the first packet size and the second packet size to the first data processing system to form a set of transferred packets, identifies round trip times and service times for the set of transferred packets, and selects an optimal packet size using the round trip times and the service times for the set of packets.

Claim 38

The distributed data processing system of claim 32 , wherein the second data processing system sends data to the first data processing system using the optimal packet size.

Claim 39

The distributed data processing system of claim 32 , wherein the network includes an asynchronous transfer mode network.

Claim 40

The distributed data processing system of claim 32 , wherein the optimal packet size is an optimal packet size for Internet Protocol packets.

Claim 41

The distributed data processing system of claim 32 , wherein the second data processing system selects the optimal packet size as: p ~ = ( aW - d ) + ( aW - d ) 2 + 4  bcW 2  c wherein {tilde over (p)} is the optimal packet size, a = S  ( p 1 ) - S  ( p 2 ) p 1 - p 2 , b=S(p 1 )−ap 1 , c = RTT  ( p 1 ) - RTT  ( p 2 ) p 1 - p 2 , d=RTT(p 1 )−cp 1 , p 1 is the first packet size, p 2 is the second packet size, RTTP(p 1 ) is the average round trip time for the first packet size determined from the round trip times for the set of transferred packets having the first packet size, RTTP(p 2 ) is the average round trip time for the second packet size determined from the round trip times for the set of transferred packets having the second packet size, S(p 1 ) is the average service time for the first packet size determined from the service times for the set of transferred packets having the first packet size, and S(p 2 ) is the average service time for the second packet size determined from the service times for the set of transferred packets having the second packet size.

Claim 42

The distributed data processing system of claim 32 , wherein the service times are identified for a slowest processing node in a path between the first data processing system and the second data processing system.

Claim 43

A computer program product in a computer readable medium for maximizing a transfer of data to a target, the computer program product comprising: first instructions for sending a set of packets using a first packet size and a second packet size to the target to form a set of transferred packets; second instructions for identifying round trip times and service times for the set of transferred packets; and third instructions for selecting an optimal packet size using the round trip times and the service times for the set of packets, wherein the optimal packet size is used in transferring the data to the target.

Claim 44

A computer program product in a computer readable medium for maximizing a transfer of data to a target, the computer program product comprising: first instructions for sending a set of packets using a first packet size and a second packet size to the target to form a set of transferred packets; second instructions for identifying round trip times and service times for the set of transferred packets; and third instructions for selecting an optimal packet size using the round trip times and the service times for the set of packets, wherein the optimal packet size is used in transferring the data to the target.

Claim 45

A computer program product in a computer readable medium for identifying an optimal packet size for transferring data to a target, the computer program product comprising: first instructions for selecting a first packet size; second instructions for sending a first set of packets to the target using the first packet size; third instructions for measuring round trip times and service times for the first set of packets; fourth instructions for selecting a second packet size; fifth instructions for sending a second set of packets to the target using the second packet size; sixth instructions for measuring round trip times and service times for the second set of packets; and seventh instructions for identifying the optimal packet size based on the round trip times and service times for the first set of packets and the round trip times and service times for the second set of packets.

Claim 46

A computer program product in a computer readable medium for identifying an optimal packet size for transferring data to a target, the computer program product comprising: first instructions for selecting a first packet size; second instructions for sending a first set of packets to the target using the first packet size; third instructions for measuring round trip times and service times for the first set of packets; fourth instructions for selecting a second packet size; fifth instructions for sending a second set of packets to the target using the second packet size; sixth instructions for measuring round trip times and service times for the second set of packets; and seventh instructions for identifying the optimal packet size based on the round trip times and service times for the first set of packets and the round trip times and service times for the second set of packets.

Claim 47

A data processing system for maximizing a transfer of data to a target, the data processing system comprising: sending means for sending a set of packets using a first packet size and a second packet size to the target to form a set of transferred packets; identifying means for identifying round trip times and service times for the set of transferred packets; and selecting means for selecting an optimal packet size using the round trip times and the service times for the set of packets, wherein the optimal packet size is used in transferring the data to the target.

Claim 48

A data processing system for maximizing a transfer of data to a target, the data processing system comprising: sending means for sending a set of packets using a first packet size and a second packet size to the target to form a set of transferred packets; identifying means for identifying round trip times and service times for the set of transferred packets; and selecting means for selecting an optimal packet size using the round trip times and the service times for the set of packets, wherein the optimal packet size is used in transferring the data to the target.

Claim 49

The data processing system of claim 40 , wherein the selecting, sending, and identifying means are initiated by a policy.

Claim 50

The data processing system of claim 41 , wherein the policy is to identify the optimal packet size based on a selected event.

Claim 51

The data processing system of claim 42 , wherein the selected event is a periodic event.

Claim 52

The data processing system of claim 42 , wherein the selected event is an initiation of a data transfer to the target.

Claim 53

The data processing system of claim 40 further comprising: determining means for determining an average round trip time and an average service time for the first packet size and an average round trip time and an average service time for the second packet size using the round trip times and the service times for the set of transferred packets and wherein the selecting step comprises selecting the optimal packet size as: p ~ = ( aW - d ) + ( aW - d ) 2 + 4  bcW 2  c wherein {tilde over (p)} is the optimal packet size, a = S  ( p 1 ) - S  ( p 2 ) p 1 - p 2 , b=S(p 1 )−ap 1 , c = RTT  ( p 1 ) - RTT  ( p 2 ) p 1 - p 2 , d=RTT(p 1 )−cp 1 , p 1 is the first packet size, p 2 is the second packet size, RTTP(p 1 ) is the average round trip time for the first packet size, RTTP(p 2 ) is the average round trip time for the second packet size, S(p 1 ) is the average service time for the first packet size, S(p 2 ) is the average service time for the second packet size.

Claim 54

The data processing system of claims 40 , wherein the service times are identified for a slowest processing node in a path between the data processing system and the target.

Claim 55

The data processing system of claim 40 , wherein the sending, identifying, and selecting means are located in an Internet Protocol layer in the data processing system.

Claim 56

The data processing system of claim 40 , wherein the data is sent in packets over a high-speed network.

Claim 57

The data processing system of claim 40 , wherein the high-speed network is an asynchronous mode transfer network.