US 9516144 B2 - Cut-through Forwarding Of Ccnx Message Fragments With Ip Encapsulation - The Lens

Cut-through Forwarding Of Ccnx Message Fragments With Ip Encapsulation

Published: Dec 6, 2016
Family: 5
Cited Works: 99
Cited by: 3
Cites: 107
Additional Info: Cited Works Full text

Abstract

One embodiment provides a system that facilitates cut-through forwarding of message fragments over a network. During operation, the system receives a packet that corresponds to a fragment of a message and contains a fragment header. The system stores in a data structure a mapping relationship between the fragment header and a hierarchically structured variable-length identifier (HSVLI), wherein the HSVLI comprises contiguous name components ordered from a most general level to a most specific level. The system makes a forwarding decision for the packet based on the fragment header prior to receiving all fragment packets associated with the message, thereby facilitating cut-through forwarding of the message fragments over a network.

Claims

A computer-implemented method for forwarding packets, the method comprising:
determining, by a content-requesting node, a maximum transmission unit of a network;

fragmenting, by the content-requesting node, a message into one or more packets, wherein the size of each packet is less than or equal to the maximum transmission unit of the network;

generating, by the content-requesting node, a flow identifier which indicates that a respective packet corresponds to the message, wherein the flow identifier corresponds to an identifier of the message,

wherein the identifier of the message is a hierarchically structured variable length identifier which comprises contiguous name components ordered from a most general level to a most specific level;

generating, by the content-requesting node, a fragment header which identifies the one or more packets as corresponding to the message, wherein the fragment header includes:
the flow identifier;

a fragment count that indicates a total number of fragments which correspond to the message; and

a fragment number that indicates an index of the fragment;

receiving, by a computing device, a first packet that corresponds to a fragment of the message and contains the fragment header;

storing, by the computing device, in a data structure a mapping relationship between the fragment header and the identifier of the message; and

making, by the computing device, a forwarding decision for the first packet based on the fragment header prior to receiving all fragment packets associated with the message,

thereby facilitating cut-through forwarding of the message fragments over a network.
The method of claim 1, wherein the data structure further comprises a fragment stream identifier which corresponds to a tuple including:
an Internet Protocol (IP) source address;

an IP destination address;

a protocol field that indicates a higher-layer protocol carried in an IP packet; and

a flow identifier that indicates that the fragment corresponds to the message.
The method of claim 1, further comprising maintaining a pending interest table, wherein a respective entry in the pending interest table indicates a pending interest, the flow identifier, a hierarchically structured variable-length identifier that identifies the interest and corresponds to the flow identifier, and a fragment bitmap that indicates a number of corresponding content fragments that have been provided for the interest, wherein the identifier that identifies the interest is a hierarchically structured variable length identifier which comprises contiguous name components ordered from a most general level to a most specific level.
The method of claim 1, further comprising:
encapsulating the packet using an outer Internet Protocol (IP) header to transmit the packet across a mixed network, wherein the mixed network is comprised of nodes associated with one or more of:
a content centric network;

a network running a version four of IP; and

a network running a version six of IP;

thereby facilitating cut-through forwarding of the message fragments using IP encapsulation over the network.
The method of claim 1, wherein the packet contains a content object in response to an interest in the content, and wherein the size of the packet containing the content object is equal in size to or smaller than a largest fragment corresponding to an interest in the content, based on a same flow identifier in the fragment header.
The method of claim 3, further comprising:
adding an entry in the pending interest table, wherein the entry corresponds to the message;

in response to receiving the first packet, updating the fragment bitmap in the entry based on the fragment count and the fragment number of the fragment header; and

in response to determining that the fragment bitmap is zeroed out, removing the entry from the pending interest table.
A computer system for facilitating forwarding of packets, the system comprising:
a processor; and

a storage device storing instructions that when executed by the processor cause the processor to perform a method, the method comprising:
determining, by a content-requesting node, a maximum transmission unit of a network;

fragmenting, by the content-requesting node, a message into one or more packets, wherein the size of each packet is less than or equal to the maximum transmission unit of the network;

generating, by the content-requesting node, a flow identifier which indicates that a respective packet corresponds to the message, wherein the flow identifier corresponds to an identifier of the message;

wherein the identifier of the message is a hierarchically structured variable length identifier which comprises contiguous name components ordered from a most general level to a most specific level;
generating, by the content-requesting node, a fragment header which identifies the one or more packets as corresponding to the message, wherein the fragment header includes:
the flow identifier;

a fragment count that indicates a total number of fragments which correspond to the message; and

a fragment number that indicates an index of the fragment;

receiving, by a computing device, a first packet that corresponds to a fragment of the message and contains the fragment header;

storing, by the computing device, in a data structure a mapping relationship between the fragment header and the identifier of the message; and

making, by the computing device, a forwarding decision for the first packet based on the fragment header prior to receiving all fragment packets associated with the message,

thereby facilitating cut-through forwarding of the message fragments over a network.
The computer system of claim 7, wherein the data structure further comprises a fragment stream identifier which corresponds to a tuple including:
an Internet Protocol (IP) source address;

an IP destination address;

a protocol field that indicates a higher-layer protocol carried in an IP packet; and

a flow identifier that indicates that the fragment corresponds to the message.
The computer system of claim 7, wherein the method further comprises maintaining a pending interest table, wherein a respective entry in the pending interest table indicates a pending interest, the flow identifier, a hierarchically structured variable-length identifier that identifies the interest and corresponds to the flow identifier, and a fragment bitmap that indicates a number of corresponding content fragments that have been provided for the interest, wherein the identifier that identifies the interest is a hierarchically structured variable length identifier which comprises contiguous name components ordered from a most general level to a most specific level.
The computer system of claim 7, wherein the method further comprises:
encapsulating the packet using an outer Internet Protocol (IP) header to transmit the packet across a mixed network, wherein the mixed network is comprised of nodes associated with one or more of:
a content centric network;

a network running a version four of IP; and

a network running a version six of IP;

thereby facilitating cut-through forwarding of the message fragments using IP encapsulation over the network.
The computer system of claim 7, wherein the packet contains a content object in response to an interest in the content, and wherein the size of the packet containing the content object is equal in size to or smaller than a largest fragment corresponding to an interest in the content, based on a same flow identifier in the fragment header.
The computer system of claim 9, wherein the method further comprises:
adding an entry in the pending interest table, wherein the entry corresponds to the message;

in response to receiving the first packet, updating the fragment bitmap in the entry based on the fragment count and the fragment number of the fragment header; and

in response to determining that the fragment bitmap is zeroed out, removing the entry from the pending interest table.
A non-transitory computer-readable storage medium storing instructions that when executed by a computer cause the computer to perform a method, the method comprising:
determining, by a content-requesting node, a maximum transmission unit of a network;

fragmenting, by the content-requesting node, a message into one or more packets, wherein the size of each packet is less than or equal to the maximum transmission unit of the network;

generating, by the content-requesting node, a flow identifier which indicates that a respective packet corresponds to the message, wherein the flow identifier corresponds to an identifier of the message,

wherein the identifier of the message is a hierarchically structured variable length identifier which comprises contiguous name components ordered from a most general level to a most specific level;

generating, by the content-requesting node, a fragment header which identifies the one or more packets as corresponding to the message, wherein the fragment header includes:
the flow identifier;

a fragment count that indicates a total number of fragments which correspond to the message; and

a fragment number that indicates an index of the fragment;

receiving, by a computing device, a first packet that corresponds to a fragment of the message and contains the fragment header;

storing, by the computing device, in a data structure a mapping relationship between the fragment header and the identifier of the message; and

making, by the computing device, a forwarding decision for the packet based on the fragment header prior to receiving all fragment packets associated with the message,

thereby facilitating cut-through forwarding of the message fragments over a network.
The storage medium of claim 13, wherein the data structure further comprises a fragment stream identifier which corresponds to a tuple including:
an Internet Protocol (IP) source address;

an IP destination address;

a protocol field that indicates a higher-layer protocol carried in an IP packet; and

a flow identifier that indicates that the fragment corresponds to the message.
The storage medium of claim 13, wherein the method further comprises maintaining a pending interest table, wherein a respective entry in the pending interest table indicates a pending interest, the flow identifier, a hierarchically structured variable-length identifier that identifies the interest and corresponds to the flow identifier, and a fragment bitmap that indicates a number of corresponding content fragments that have been provided for the interest, wherein the identifier that identifies the interest is a hierarchically structured variable length identifier which comprises contiguous name components ordered from a most general level to a most specific level.
The storage medium of claim 13, wherein the method further comprises:
encapsulating the packet using an outer Internet Protocol (IP) header to transmit the packet across a mixed network, wherein the mixed network is comprised of nodes associated with one or more of:
a content centric network;

a network running a version four of IP; and

a network running a version six of IP;

thereby facilitating cut-through forwarding of the message fragments using IP encapsulation over the network.
The storage medium of claim 13, wherein the packet contains a content object in response to an interest in the content, and wherein the size of the packet containing the content object is equal in size to or smaller than a largest fragment corresponding to an interest in the content, based on a same flow identifier in the fragment header.
The storage medium of claim 15, wherein the method further comprises:
adding an entry in the pending interest table, wherein the entry corresponds to the message;

in response to receiving the first packet, updating the fragment bitmap in the entry based on the fragment count and the fragment number of the fragment header; and

in response to determining that the fragment bitmap is zeroed out, removing the entry from the pending interest table.