Electrical computers and digital processing systems: multicomput – Distributed data processing – Client/server
Reexamination Certificate
2000-06-09
2004-01-20
Eng, David Y. (Department: 2155)
Electrical computers and digital processing systems: multicomput
Distributed data processing
Client/server
Reexamination Certificate
active
06681244
ABSTRACT:
CROSS REFERENCE TO RELATED APPLICATIONS
N/A
STATEMENT REGARDING FEDERALLY SPONSORED RESEARCH OR DEVELOPMENT
N/A
BACKGROUND OF THE INVENTION
The present invention relates generally to network computing systems, and more specifically to systems and techniques for operating a network adapter when an associated network computing system is in a low-power state.
In recent years, power management applications have been increasingly used in computers and other computerized devices to conserve or otherwise manage the power that they consume. Although power management applications have traditionally been used in battery-powered, portable computers, the proliferation of AC-powered, desktop computers throughout the world has provided strong incentives to use such power management applications in these computers as well.
In a typical power management scheme, a computer or at least a certain number of components incorporated therein is powered-down when the computer is not in use for an extended time. A computer or sub-system in this condition may be referred to as operating in a “low-power state” or a “sleep state.” Full power is then restored to the computer when it is required for use. A computer in this alternate condition may be referred to as operating in a “full-power state.” For example, a computer in the low-power state may be in a “suspended power state,” in which all power is removed from the computer except for that required to maintain data stored in memory and to refresh DRAMs. Alternatively, a computer in the low-power state may be in a “hibernated power state,” in which data stored in memory is written to disk before power is removed from the computer. In both the suspended power state and the hibernated power state, a computing system typically saves its current operational state. In contrast, a computing system typically loses its current operational state in a general power-down condition.
Although such power management schemes have greatly enhanced the energy efficiency of computers and other computerized devices, current power management techniques have drawbacks in that they are not easily adapted for use in network computing systems. For example, a computer connected to a network typically monitors the network for data packets. If the computer detects a data packet, then it generally either rejects the data packet or accepts and processes the data packet. A computer normally rejects data packets that are not addressed to that computer, and accepts and processes data packets that are addressed to the computer. However, if the computer implements a power management application that causes it to enter the low-power state after a period of non-use, then that computer in the low-power state may be incapable of monitoring the network for data packets. As a result, data packets that are addressed to that computer may go undetected and be inadvertently lost.
Not only may data packets be inadvertently lost on the network, but the computer in the low-power state may also lose its communications link to the network. For example, a computer connected to a network that uses the Internet Protocol (“IP network”) is typically identified by its Internet Protocol address (IP address). Further, IP addresses within IP networks are frequently assigned to computers using the Dynamic Host Configuration Protocol (DHCP). A characteristic of DHCP is that an IP address is assigned or “leased” to a computer for only a fixed time interval. Accordingly, that computer is responsible for periodically renewing the lease of its IP address. However, if the computer implements a power management application and enters the low-power state after a period of non-use, then that computer may be incapable of renewing its lease within the lease period. As a result, the computer in the low-power state may inadvertently lose both its current IP address and its communications link to the network.
It would therefore be desirable to have improved systems and techniques for operating network computing systems that implement power management applications. Such systems and techniques would enable a networked computer to receive and process data packets even when the computer is in the low-power state. It would also be desirable to have systems and techniques for renewing leases of IP addresses when a networked computer is in the low-power state.
BRIEF SUMMARY OF THE INVENTION
In accordance with the present invention, a system and method are disclosed for operating a network adapter when an associated network computing system is in a low-power state. Specifically, the disclosed system and method enable a network computing system in the low-power state to receive messages over the network communications link and to transmit responses thereto. Such receipt and transmission of messages and responses over the network is accomplished by maintaining power to a network interface adapter of a host network computing system when the host system is in the low-power state. The network interface adapter includes a processor that is capable of determining whether or not the host system is in the low-power state, and receiving messages over the network and transmitting responses thereto without requiring the host system to transition from the low-power state to a full-power state.
In one embodiment, a system is provided for responding to at least one predetermined message received over a communications link at a network interface of a network interface adapter. The network interface adapter is communicably coupled to a host computer, which includes a host processor executing an operating system out of a memory. The host processor has a host operating system present state in which the operating system is operational, and a host operating system absent state in which the host operating system is not operational. The system further includes a network adapter processor that detects the receipt of the at least one predetermined message at the network interface. In response to the detection of the at least one predetermined message, the network adapter processor determines whether the host computer is in the host operating system present state or the host operating system absent state. In response to a determination that the host computer is in the host operating system absent state, the network adapter processor generates a response to the at least one predetermined message and transmits the response via the network interface over the communications link.
REFERENCES:
patent: 5117500 (1992-05-01), Davis et al.
patent: 5742833 (1998-04-01), Dea et al.
patent: 6493824 (2002-12-01), Novoa et al.
Connery Glenn W.
Cross Patricia C.
Duguay Christopher J.
3Com Corporation
Eng David Y.
Weingarten Schurgin, Gagnebin & Lebovici LLP
LandOfFree
System and method for operating a network adapter when an... does not yet have a rating. At this time, there are no reviews or comments for this patent.
If you have personal experience with System and method for operating a network adapter when an..., we encourage you to share that experience with our LandOfFree.com community. Your opinion is very important and System and method for operating a network adapter when an... will most certainly appreciate the feedback.
Profile ID: LFUS-PAI-O-3247483