Multiplex communications – Network configuration determination – In a bus system
Reexamination Certificate
1997-11-05
2002-07-09
Marcelo, Melvin (Department: 2663)
Multiplex communications
Network configuration determination
In a bus system
C370S475000
Reexamination Certificate
active
06418124
ABSTRACT:
BACKGROUND OF THE INVENTION
1. Field of the Invention
This invention relates generally to computer networks and, more particularly, to a method and apparatus for routing a packet in a computer network.
2. Description of the Related Art
Local area networks (“LANs”) are now commonplace. A LAN is a computer network that spans a relatively small area. Most LANs connect workstations and/or personal computers in a way that enables a user to access data and devices at any of the locations on the LAN so that many users can share expensive types of devices such as laser printers and can share information.
There are many types of LANs. Common examples include token ring networks and Ethernets. A token ring network is a type of computer network in which the computers are schematically arranged in a circle. A packet of digital information, called a “token,” travels around the circle. To send a message, a computer attaches its message to the token and lets the token continue traveling around the network. Each computer checks the message when it receives a token to see if it is the intended destination and, if so, removes the message from the token. An Ethernet is a bus network in which packets of digital information travel up and down the bus. Although these are the most common type of LANs, there are still others, such as ARCnet, in use.
FIG. 1
depicts a prior art computer network
10
that will be used to illustrate several concepts associated with the invention. The computer network
10
generally comprises a first LAN
12
, a second LAN
14
, and a router
15
. A “router” is a device that connects two or more LANs as shown in FIG.
1
and in a manner described more fully below. The first LAN
12
includes a server
16
and two workstations
18
. The second LAN
14
similarly includes a second server
20
and two workstations
22
. Each of the router
15
, the server
16
, the workstations
18
, the server
20
, and the workstations
22
may be generally referred to as a “LAN device.” The various components of the computer network
10
are interconnected by communication links
23
,
24
, and
25
, which may be any type of media such as twisted pair wires, coaxial cables, fiber optic cables, or some combination thereof. Some LANs even use wireless communications links, such as radio or infrared waves.
The workstations
18
and workstations
22
can send messages to one another through the network
10
in “packets” of digital information. Each device on the network
10
is assigned a physical address. Each packet contains a destination address for one of the other workstations and message data. Thus, for one workstation
18
to communicate with the other workstation
18
, the first workstation
18
composes a packet which is then electronically transmitted to the other workstation
18
over the LAN
12
through the communications links
24
. The workstations
22
use this same process. However, for a workstation
18
to transmit a message to a workstation
22
, the workstation
18
must compose a data packet that is transmitted via the router
15
.
The manner in which the LAN devices communicate is determined by a “protocol.” A protocol is an agreed upon format for transmitting data between and among devices. Thus, the network's protocol determines the composition, transmission, receipt, and decomposition of the packets. Most networks use some form of the transport control protocol/Internet protocol (“TCP/IP”). The TCP/IP protocol is actually a group of protocols. Note that, although the term “Internet” appears in the name of the protocol, its application is not limited to the Internet, other wide area networks (“WANs”), or any other type of network.
The TCP/IP protocols are typically used to implement computer networks, such as the computer network
10
, dictate that the destination address for any individual packet have at least two parts. One part of the address is very general and one part of the address is very specific. The general part of the address is known as the Internet protocol (“IP”) address and determines whether the packet is sent to its destination through the router
15
. The router
15
then determines, or “resolves,” the specific physical address of the packet's destination from the IP address carried by the packet. Thus, the IP address is not an actual physical address, but instead merely represents a physical address.
When the router
15
receives a packet, it determines whether it knows the physical address represented by the IP address. The router
15
temporarily stores physical addresses to which it has recently transmitted, and checks this store whenever it receives a packet. If the physical address of the received packet can be determined from the stored information, the router
15
transmits the packet right away. If not, then the router
15
must “resolve” the physical address. The router
15
typically does this using what is known as an “address resolution protocol” (“ARP”).
The ARP is part of the TCP/IP protocol suite and is used to convert an IP address into the physical address that is the destination of the packet. The ARP requires the router
15
to broadcast to all LAN devices on the network
10
what is known as an ARP request. The ARP request instructs the LAN device corresponding to the IP address to respond by transmitting its physical address back to the router
15
. The intended destination then replies to the request with its physical hardware address. The router
15
receives the physical address and stores it temporarily.
During the ARP, the router
15
either retains or discards the packet for which the ARP is issued. If the packet is retained, the router
15
simply transmits it to whichever LAN device responded to the ARP request. However, some routers do not retain the packet during ARP and the packet is lost. The LAN devices are typically programmed for this eventuality such that they will send the packet to the router
15
several times. Thus, the router
15
might receive the packet two or three times while transmitting the packet to the destination only on the second or third try.
For example, assume that a workstation
18
transmits a data packet to a workstation
22
over the network
10
and that the router
15
does not retain the data packet. The workstation
18
transmits the data packet over the LAN
12
to the router
15
. The router
15
then broadcasts an ARP request and the workstation
22
replies to the request by sending its physical address to the router
15
. In the meantime, the router
15
has dropped the data packet. The router
15
then receives the physical address of the workstation
22
over the LAN
14
and temporarily stores it. The workstation
18
once again sends the data packet to the router
15
. The router
15
then checks its memory, finds the previously resolved physical address of the workstation
22
, and forwards the data packet to the workstation
22
. Although such a network obviously lacks something in efficiency, this protocol simplifies the design and reduces the cost of implementing the network
10
overall.
This procedure works reasonably well unless one or more of the LAN devices on the network
10
includes a power management feature that inactivates the LAN device when not in use. If, for instance, a workstation
22
incorporates a power management system, the operating system of the workstation
22
will switch to a “sleep” state to reduce power consumption after a predetermined period of inactivity. When the router
15
broadcasts an ARP request to which a sleeping workstation
22
should respond, the workstation
22
has to “wake up” before responding.
The presence of power-managed devices is important because of the way they typically implement their retention capabilities. LAN devices such as the servers
16
and
20
and the workstations
18
and
22
typically have a retention capability. However, the retention capability for some devices deactivates when the device goes to sleep. Thus, in the context of ARP, the LAN device when it is asleep may retain the ARP request o
Blakely , Sokoloff, Taylor & Zafman LLP
Hyun Soon-Dong
Intel Corporation
Marcelo Melvin
LandOfFree
Method and apparatus for routing a packet in a network does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with Method and apparatus for routing a packet in a network, we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and Method and apparatus for routing a packet in a network will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-2869518