KR20180086884A - Efficient power management method, power management device and computer program - Google Patents

Abstract

The present invention relates to an efficient power management method for enhancing the frame-address matching function of a network controller. In a full-power operating mode, the controller executes normal functions. In a low-power mode, the controller includes frame-address and frame-data matching capabilities and free responses which are determined in advance with respect to the network protocols and are programmable. When a match is made with respect to frame destination addresses or frame data, a system related to the controller is put into an activated state. The controller automatically generates and transmits a simple response packet to a network upon which such match is made in frame data areas when the controller is programmed to execute a pattern matching operation in the frame data areas. Conversely, if no match is detected, the controller does not execute any operations. Frame data patterns can be dependent on protocols and be programmed to be able to execute a mask-out operation on a desired data field substantially dependent on protocols.

Description

[0001] The present invention relates to an efficient power management method, a power management apparatus,

The present invention relates to efficient power management of a computer, and more particularly, to a system and method for reducing power consumption of a computer in computer networks.

Today there is increasing interest in environmental conservation and more efficient energy consumption, and new design techniques have been explored to embed energy saving features in new products.

The computer industry has felt this emergence of a more energy-efficient computer system on its own, an example of which is to use a low-power standby state or a hibernation state whenever possible to conserve energy Computer.

Applications such as PCs or workstations that can toggle between active and low power states exist within the computer network. In theory, a PC or workstation (generally referred to as a station for later purposes) may be interconnected via a network and may be selectively switched to a normal powered-up state by a message over the network, Lt; / RTI > However, it is very difficult to maintain such a station in a low-power or hibernate state where there is no activity in the standby state while connected to a network, such as an Ethernet network, where the actual station is waiting for a message to be activated. This is due to the nature of network transmission taking the form of broadcast packets. Typically, certain forms of such broadcast packets unfortunately often wake up or activate stations unnecessarily in a network environment. A representative example of such a broadcast packet is the IPX-SAP packet that occurs very frequently, for example, once per second, and in fact, such a station does not have any meaning because it frequently generates broadcast packets with irrelevant information to unnecessarily activate the station It can not be in a standby state or a hibernation state.

BACKGROUND OF THE INVENTION 1. Field of the Invention The present invention relates to efficient power management of a computer, and in particular, to provide a system and method for reducing power consumption of a computer in computer networks.

In order to solve the above problems, the present invention eliminates all such broadcast packets through filtering by the network controller of the station. Although this approach may seem to solve the problem of unnecessary activation of the station, a serious side effect of this approach is that many existing protocols will cease to operate on this network, and likewise an important problem is that the connection may be disconnected as is. Thus, it is common in today's PC and workstation network technologies to ensure that such stations are always in full power-on state.

This is, of course, contrary to the above-mentioned modern trend of energy conservation policies in line with the Environmental Protection Agency's Energy star guidelines, which always raise the concept of energy conservation and low-power conditions when available technically. Unfortunately, this guideline strongly recommends that for a network connection, the connection is not desirably disconnected while the station is in a low power state. This presents a clearly contradictory design criterion that, while saving power by designing a low power state of the network station, should also prevent disconnections that occur due to toggling to these low power states.

A typical protocol such as the TCP / IP protocol uses an Address Resolution Protocol (ARP) broadcasting mechanism to provide a network IP address and a physical address (for example, IEEE Ethernet 48 bit address) are affected. For example, if it is desirable for the host A to attempt to disassemble the IP address Ib in operation, according to the protocol, the host may send a special packet that requires the host with the IP address Ib to respond with its corresponding physical address, Broadcast. all hosts on the network will receive this request, but only host b will recognize its IP address I-b on broadcast, so that only host b will respond by sending a response containing its physical address.

When A receives this response, A may send the desired Internet packet (e. G., Mail, etc.) directly to station B using the thus received physical address. Therefore, in most cases, an ARD broadcast packet is transmitted prior to the host sending a dedicated packet to another host for mail or other applications. Ideally, if the station B is in the low power mode for energy saving purposes as it is transmitting, it will ignore any broadcast packets while maintaining the low power state and low energy consumption mode and only if the broadcast packets over the network are actually intended for the B station It is preferred that the activated or activated.

However, most controller chips, such as those actually associated with Ethernet protocols, only perform physical destination address matching of the frame and therefore simply compare the destination address field of the incoming frame with a number of predefined addresses. In particular, such a controller does not have the capability to perform pattern matching in the frame data area. However, the goal of saving network system energy while maintaining the integrity of the network, without the ability to analyze the data in the frame data area, is to only enable and disable the broadcast address to solve the problem Is simply not achieved. In other words, whether the data frame portion of the frame will receive a particular broadcast packet, e.g., in response to a given broadcast packet, because the data associated with the broadcast packet is independent of the particular station, Because it provides information about whether it will be active, or vice versa.

Another example illustrating the problem in practice can be found in the case of a service advertisement protocol (SAP) mechanism operating on another protocol, such as the IPX protocol associated with the network, especially the IPX protocol. In an implementation of the present invention, SAP broadcast packets (e.g., server broadcasts that indicate services available over the network) are transmitted by the server once per minute. If the client host is in low power mode and has to respond to such broadcast packets, the client has no opportunity to remain in a low power mode apparently to achieve the desired energy conservation objective in any significant or meaningful way. However, to achieve this goal, if the client host is kept in a low power mode and thus does not respond to such SAP broadcast packets, the client may obviously miss several important pieces of information such as the fact that the server has been shut down. Therefore, there is a need for a mechanism such as an Ethernet controller that is capable of conditionally activating a station only by a selected subset of SAP packets.

Another problem associated with maintaining a station's low-power state in a networked environment is that some network protocols require that the client send a heart-beat type of query (e.g., a particular client continues to survive, To verify that it functions properly). Thus, in combination with the above difficult situation, the solution to the above-mentioned problem is that in the low power mode, the client's network controller (such as the Ethernet controller described above) autonomously responds to such heart- the ability to respond autonomously.

It is therefore an object of the present invention to provide improved energy saving features in a network environment.

It is another object of the present invention to provide an inexpensive hardware solution to the problem of energy management at a network computer station while preventing connection and station disconnection that may cause loss of important data.

Another aspect of the present invention is to provide energy-saving peer-to-peer station communication in a computer network environment that is transparent to and dependent on network software and servers.

Another aspect of the present invention provides a programmable power management system and method embodied in a network interface controller for a network computer station capable of easily, inexpensively changing, enabling and disabling system power-on for selected packets .

Another aspect of the invention is to provide a response in a computer network to an owner ' s heartbeat query without having to activate the station.

Another aspect of the present invention is to provide a programmable network interface that allows filtering in response to network packets providing interactions with a power-saving workstation or a computer over a network.

Another aspect of the present invention is to provide a programmable network interface for use in a computer network in response to a power-on / off sequence without the need to modify the application or control software.

The present invention relates to efficient power management of a computer, and in particular, provides a technical effect of providing a system and method for reducing power consumption of a computer in computer networks.

1 is a block diagram of a computer network in which an alias associated with power management of a networking station may be implemented.
2 is a functional block diagram illustrating the toggling of a station operating in the network of FIG. 1 and including a network interface controller between a full-power state and a power-down state.
Figure 3 is a functional block diagram of the exemplary network interface controller of Figure 2;
4 is a flow diagram of software illustrating the MAC controller portion of the NIC of FIG. 3 implementing the present invention;
5 is a schematic diagram illustrating the format of a typical representative frame or packet on a network including a portion of frame data used for pattern matching according to the present invention;
FIG. 6 is a schematic diagram illustrating frame data and predetermined responses in accordance with the present invention. ≪ RTI ID = 0.0 > 1 < / RTI > is an example illustrating a schematic structure of a partial-

In a preferred embodiment of the present invention, a frame-address matching function of the network controller is enhanced. In the full power-on mode, the controller performs its original function. However, in accordance with the present invention, in the low power mode, the controller includes the capability of frame-address and frame-data matching and includes a predetermined and programmable autonomous response to the network protocol.

In particular, in the preferred embodiment, a controller such as an Ethernet controller provides up to 16 frame-addresses and data matching. When a match is made to the frame destination address or frame data, the system associated with the controller is put into an activated state. When the controller is programmed to perform the above-described pattern matching in the frame data area, and a match occurs in the frame data area, the controller will automatically generate and transmit a simple response packet over the network. Conversely, no match is detected and no action is taken. The frame-address may be any valid address, such as broadcast, multicast, or a fixed IEEE address. The frame-data pattern is a programmable pattern with the ability to mask out any desired data fields and is a protocol-dependent pattern.

In an alternative embodiment, a microcontroller is included in conjunction with a network controller, such as an Ethernet controller, to perform the low-power mode function, where the network controller is a low cost controller (in the embodiment, (Which can take the form of a simple interface). According to this embodiment, the Ethernet or similar controller is subordinate to the microcontroller in the low power mode. In the full-on mode, the microcontroller is disabled. The performance requirements for the microcontroller may be minimized or unnecessary because most broadcast packets are less than 64 bytes and therefore easily applicable to the Ethernet or similar controller's internal FIFO buffer.

According to the foregoing capabilities, the workstation or PC may be programmed to enter the start mode via a particular broadcast packet, or alternatively, be able to autonomously respond to broadcast packets with a simple predetermined packet. In the hibernation mode, in a similar manner, the station is programmed to be invoked only by a unique address and data frame pattern that may be subject to future standards to be implemented as desired. Thus, the station will only be activated when appropriate, and the Ethernet controller handles the simple response of the broadcast packet without unnecessarily putting the station into the start-up mode, but preferably ensures that the network connection is maintained for the entire time.

A representative and more detailed description of the general principles associated with network technology, such as the Ethernet technology discussed herein, is provided in " INTERNET WORKING WITH TCP / IP, VPL.1, PRINCIPLES, PROTOCOLS AND ARCHITECTURE, DE Comer, Prentice-Hall Publishing Co., 1991 , In particular from section 2.4 of these references, which are incorporated herein by reference.

Referring firstly to FIG. 1, a high-level description of a network environment in which the present invention may be preferably implemented will be provided first. Referring to FIG. 1, there is shown a diagrammatic representation of a data processing system 8 that may be utilized to implement the system and method of the present invention. As can be appreciated, data processing system 8 may include multiple networks, such as local area networks (LANs) 10 and 32, each of which preferably includes a plurality of individual computers 12, 12a-12c, 30, 31, 33 and 35). (The discussion will then be referred to randomly, even if the discussion relates to all of the computers in the network 32.) The computers 12 and 30 may be referred to as " Such as an IBM Personal System / 2 (also referred to as PS / 2) computer, or an IBM RISC System / 6000 computer workstation, which is a product of IBM Corporation, located in Monmouth. RISC System / 6000 is a trademark of IBM Corporation; Personal System / 2 and PS / 2 are registered trademarks of IBM Corporation. Of course, I understand that a number of intelligent work stations (IWS) coupled to a host processor may be used for each such network.

As is common in such data processing systems, each individual computer may be coupled to a storage device 14 and / or a printer / output device 16. In accordance with the method of the present invention, one or more storage devices 14 may be implemented as objects, such as documents, resource objects, or executable code that all users of the data processing system 8 may periodically access. Can be stored. Each of these objects stored in the storage device 14, in a manner well-known in the art, can be freely interchanged via the data processing system 8, for example, by transferring objects from the individual computers 12 or 30 to the user. It may be exchanged.

1, the data processing system 8 may also include a number of mainframe computers, such as a mainframe computer 18, which is preferably connected to the communication link 22 To the LAN 10 via a network. The mainframe computer 18 may be implemented using Enterprise Systems Architecture / 370 (also referred to as ESA / 370) or Enterprise Systems Architecture / 390 (also referred to as ESA / 390) . Depending on the application, a medium-sized computer such as Application System / 400 (also referred to as AS / 400) may be utilized. Enterprise Systems Architecture / 370, ESA / 370, Enterprise Systems Architecture / 390, and ESA / 390 are registered trademarks and Application System / 400 and AS / 400 are registered trademarks of IBM Corporation. The mainframe computer 18 may also be coupled to a storage device 20, which may be provided as a remote storage device for the LAN 10. Similarly, the LAN 10 is coupled to a subsystem control / communication controller 26 via a communication link 24 and communicated via a communication link 34 to a gateway server < RTI ID = 0.0 > ) ≪ / RTI > The gateway server 28 is preferably an individual computer or an IWS that connects the LAN 32 and the LAN 10.

As described above with respect to LAN 32 and LAN 10, an object may be stored in storage device 20 and stored in a mainframe, such as a Resource Manager or File System Manager, May be controlled by the computer 18. Of course, it will be appreciated that the mainframe computer 18 may be located at a geographical distance from the LAN 10, and similarly, the LAN 10 may be located at a substantial distance from the LAN 32 . For example, LAN 32 may be located in Seoul, while LAN 10 may be located in battle, and mainframe computer 18 may be located in Busan.

The preferred embodiment of the present invention may be embodied in various computers shown in the data processing system 8. [

Referring to FIG. 2, a second diagram illustrates a mechanism by which a station can be toggled between a power-down (sleep) mode and a full power state as a result of a network interconnected. At the top of the figure, the network interface controller 112, which takes the form of an adapter card or is mounted within the motherboard of the station, may be a personal computer, a workstation, May be interconnected or integrated with the station (110). In general, the function of the NIC 112 is to provide interface logic between the network and the network access device 126 and the real station 110 so that the station can access the same network, such as Token Ring, Ethernet, ATM, FDDI, So that it can easily communicate with other apparatuses. In the normal operation, when the NIC 112 receives the appropriate signal over the network medium 126, and the NIC 112 receives the appropriate signal over the network medium 126, By causing the interruption of the processes running on the station, such software running on the station processes the packets. If the station is in a full-on mode, as shown at the top of FIG. 2, the process is repeated cyclically as shown by loop 118, and an active event, such as a mouse, Such as I / O from a suitable pointing device. This loop essentially implements a predetermined power management policy such that, in the absence of a predetermined activity event sensed during the iterative cycling of the loop 118, the station toggles to the sleep or power-down state shown below the second figure do. This transition is illustrated by the power management station arrow 114, which policy is defined by the operating system in which the station is operating.

After the absence of an active event is detected, the station 110 enters a power-down state and a feedback loop or process 120. However, in this case, the loops are detected during the iterative cycling of the loop 118, and after a wake event such as a keyboard stroke or any other activity event in which the station would have been held in the full- up event. As soon as such a start event is detected in the loop 120, the station 110 will exit the power-down state as shown by the start-up arrow 116 while at the same time the station 110 is once again in the full power state Will enter.

Referring to FIG. 3, the NIC 112 of FIG. 2 is described in more detail. These conventional INCs typically include three basic functional components: a media access controller (MAC) 132, a physical interface or transceiver 134, and generally several self- Lt; RTI ID = 0.0 > 136 < / RTI >

First, with respect to the MAC 132 controller, the conventional type of MAC 132 is essentially a physical medium or represents the lowest level of the pre-layer network protocol, encapsulates it, and is dependent on the particular network technology utilized.

That is, in the embodiment discussed below where an Ethernet network is provided, the specific MAC protocol associated with the Ethernet will be predefined and implemented by the MAC controller 132. [ It will be appreciated from the figure that the MAC controller 132 is interfaced to the host bus 130 of the particular station installed.

A physical interface or transceiver 134 is then coupled between the digital domain in which the host bus 130 and the MAC controller 132 are operating and the analog domain in which the magnetic medium 136 and the network medium 126 are operating. Lt; RTI ID = 0.0 > level. ≪ / RTI >

Finally, with respect to functionality, the magnetic block 136 provides suitable impedance matching between the transceiver 134 and the network medium 126 and provides an interface between the transceiver 134 and the network medium 126, such as Ethernet, FDDI, Displays the function to be provided. Thereafter, a more detailed description will follow with respect to the modified function of the MAC 132 according to the present invention.

Referring to FIG. 4, there is shown a flow of a MAC controller 132 to implement the present invention. A given station implementing the present invention in a network, in the initial condition or state shown by reference numeral 140, is in a low power or power saving state, as described by the power management policy described above, State. It should be noted that despite the power-down state of the station itself, the NIC 112 associated with the station continues to be powered down, even though the well-known transmission circuit associated with the NIC may be properly powered off, By being able to remain in the power-up mode, an event is expected to be detected and power-up of the transmitter circuitry will be required.

Referring now to FIG. 4, the MAC controller in the NIC 112, as shown by block 144, monitors the media 126 of the network to determine the NIC Lt; RTI ID = 0.0 > a < / RTI > packet is being received. As long as the NIC determines that the packet is not being received, the NIC will remain in the loop, continuing to monitor for the packet, as shown by the arrows 152 and 142. It is important to note that at this point the station itself remains in a power-down state, thereby achieving the energy efficiency of the present invention despite the fact that the station remains an essentially active component in the network. This is because the NIC 112 performs the function of monitoring the network activity for information contained within the packet of interest of a particular station.

Once the NIC determines that a packet has been received 146, the MAC controller code will determine at block 148 whether the received packet is a link beat pulse. Link bit pulses are associated with a particular medium in which the network is operating, such as Ethernet, Token Ring, etc., with their own definition of how they will operate, and each of these pulses is a link bit mechanism. Essentially, the purpose of the link bits is to enable the network to periodically poll the stations on the network, for example, to detect heart-beak signals, such as whether or not various stations are on the network and remain operational . In response to this detection that a period link beat pulse has actually been received in the packet, the process exits 150 and accordingly the NIC performs an appropriate link bit pulse process 154 on the particular medium . After this processing operation, the process exits at step 158 and returns to the monitoring state via path 156, 142, where the NIC monitors the reception of the packet and remains in the loop.

It is also noted that in providing self-developed information in accordance with proprietary protocols to perform such heart-beating mechanism checks, stations may desire not to rely on a particular physical medium protocol I will. In this situation, the solution provided by the present invention can be provided to automatically respond to heartbeat queries without requiring the activation of an entire workstation or PC.

Continuing with reference to FIG. 4, a packet is received, and it is determined that this is not a simple link bit pulse and, as shown by the escape path 160 from block 140, (frame address match) 162 is present.

Referring briefly to FIG. 5, FIG. 5 illustrates a typical format of a frame or packet in a network moving across an Ethernet network. A detailed description of this exemplary format can be found on pages 26-27 of section 2.4 above, cited above. It should be noted that one field in this frame, which is typical of various network protocols, may be referred to as the destination address in the present case. It is common for each set of hardware, or more suitably each station in the network, to have such a destination address uniquely defined. As shown in block 162 of FIG. 4, the MAC controller flow at this point will perform a comparison to determine if such a frame address, as indicated by the destination address in the packet, matches the unique address of the station. If so, the process escapes along path 164, so that such a frame address match has been detected, so it is clear that it is clearly desirable for the packet being received to interact with this particular station.

At this point the system will respond to this signal on path 164 to determine if the corresponding station is in a hibernation state, as shown by block 166. [ If not in the hibernation state, it is shown as escaping along path 168, and the system should simply be actuated, as shown by block 170, by appropriate interrupts, And therefore the system will be in full-on state as shown by arrow 172. [ Alternatively, at block 166 it is determined that the station is actually in a complete hibernation state and the process will escape along path 176, so that system power is turned on and at step 178 a full system boot Will be executed. The difference between the two states 170 and 178 is that in state 170 the CPU may have a low power applied to the CPU where the system can only be started by an interrupt while the block 178, In the case of the pole system boot shown in FIG. It should also be noted that in this hibernation state, substantial energy savings can be realized because typically only about two watts of power may be consumed in this state and the attributes for NIC 112 remain intact Because. In the prior art, it should be noted that various types of maneuvers can be provided, such as when a ring start abnormality is provided in a dedicated PC. An important difference, however, is that in this case there is no modem in the network where the installed system is needed to handle the network start-up command. Thus, while the station needs to be up all the time to check all states, the station according to the present invention allows the system to be shut down or hibernated completely in the network and to check the status of the station To be invoked in spite of the ongoing provision of the ability to enable the server. This is because the MIC with the modified MAC controller flow described herein responds to the packet and only the powered up component can always provide the function of monitoring them while consuming less power in association with the NIC itself.

Referring to FIGS. 5 and 6 in more detail, they correspond to the novel frame data matching function 182 and the predetermined response function 188 shown in FIG. 4 Will be described in more detail. As shown in FIG. 4, if the MAC controller flow determines that there is no frame address match 162 as shown in path 180, then the program implemented in the MAC of the NIC includes frame data It is determined whether there is a match 182 or not. Referring to FIG. 5, it should be noted that a frame data field is typically provided in a network packet of variable size, followed by an implementation of a different frame data mask, such as denoted by 200, 201 in FIG. 6 To provide a facility that allows for Such a facility as just described may include a frame data field and a check 182 shown as a block 182 examining the first 64 bytes to match the predetermined data stored in the MAC that uniquely identifies the station associated with the MAC, ).

It is important to note that in today's network technology, a given station can support multiple protocols simultaneously, and thus this frame data field allows different frame data masks associated with each support protocol to be stored. In particular, for the tcp / ip protocol, for example, there may be no predetermined or required predefined response. However, in the case of the Net bios protocol, a specific type of predetermined response may be required. It will be appreciated that this predetermined response is essentially a unique message from a particular station in a particular protocol to notify a particular server that the station is operational and functional.

Following the embodiment, another protocol, such as the Netware protocol, may require additional and different predetermined responses to packets transmitted in accordance with the NetWare protocol. In particular, SAP broadcast packets can be transmitted from the server once per minute for the IPX protocol of NetWare, as described above, so that the server polls the stations over the network to determine their operability. Thus, it will be readily appreciated that the provided stations provide a predetermined response to such polling that is uniquely designed to match this particular protocol. Also, as described above, if the client host is a low power mode and does not respond to SAP broadcast packets, the client station may miss some important information. As shown by block 188, the predetermined response may be known from the frame data matching function shown in block 182 (indicated to flow along path 184 and block 188). After the look-up table function determines that this pre-defined response is actually stored as frame data pre-determined response information in FIG. 6, the system will exit the path 192 to allow the appropriate protocol-dependent pre-determined response message to be transmitted, The process continues to monitor for additional packets 144 by cycling again along paths 196, 156 and 142. However, an important feature is the fact that the NIC controller will conditionally start the station from some packets, such as those SAP packets for which a predetermined response has not been detected. This is shown as escaping the existing block 188 along path 190, and thus the system can be started as desired.

Continuing referring to FIG. 4, if it is determined that a frame data match did not occur at block 182, this essentially means that the particular station where this flow is being executed should not be activated as a result of this packet, The process exits at step 186 and multiple returns to the loop at block 144, which monitors the reception of the additional packet 144. [ However, by escaping along path 184, indicating that the frame data has received an agreement, it indicates that it will be necessary to activate the station. However, according to the present invention, a final check to determine if a pre-stored protocol dependent response is available prior to this is performed by block 188, and the process exits at step 192, And will send a dependent response message.

Referring now to FIG. 6, FIG. 6 shows a data structure that may be stored in the NIC for this detection of the pre-stored response according to block 188 and other operations of the appropriate response shown in block 194, For example. The data stored in the NIC to perform this function includes a frame data mask 200, a predetermined response bit 202, a length and pointer field 204, 206 Quot; field "). The important thing to note is that each field is programmable as desired. Thus, for example, each frame data mask 200 or 201 may selectively mask out a predetermined area, such as a protocol message, to which the desired station responds or does not want to be activated, Allows selective filtering of messages to be activated.

Each of these frame data masks will be associated with a different network protocol, such as Netbios, NetWare, and the like. The purpose of the pre-canned response bit field 202 is to indicate which mask of the frame data mask is to be activated. For example, response bit 202 is provided for the frame data mask associated with NetWare 201, and particularly when the appropriate portion of the frame data field of Figure 5 is being compared to this frame data mast 201, The request requires a match indicating that a predetermined response is needed, as shown by the path in FIG. Next, it is a problem to determine whether there should be a properly expected response message only by the NIC, without activating the station. This is why length and pointer fields 204 and 206 exist. The length field 204 indicates the length of a suitably predetermined response message corresponding to a particular frame data mask, such as the mask 201. Similarly, the corresponding pointer 206 points to a location in the response buffer 208 within the NIC that contains a particular appropriate response corresponding to the particular frame data mask selected by the provision of the predetermined response bits 202. [ For example, in performing the frame data matching and predetermined response functions 182 and 188, it is determined whether the data of the input packet in the frame data of FIG. 5 is a NetWare PIX message for which a predetermined response is required, (After making such a determination using the lockup table of FIG. 6), a pointer to the appropriate response and a pointer to the length of this response in the response buffer 208, and then, as shown in block 196 After obtaining a pointer to this appropriate length, a packet is generated that transfers this appropriate response message to the network as shown by block 196.

To clarify the present invention, it is assumed that the station is generally in a low-power or power-saving state with reference to FIG. 4, but the invention is not limited to the case where the station is only in this state. Rather, the present invention provides various alternative forms of such low-power or power-saving states (e.g., sleep, hibernation, standby, or power saving but not limited to, power saving states.

5 and 6, the teachings of the present disclosure relate to the frame data or portions of the mask shown in FIG. 5 of the frame data mask, such as the mask of reference numeral 200 of FIG. 6, It does not mean that it should be limited only to the case where it should be completely matched. Thus, the present invention allows for monitoring for coincidence between all subsets of the frame data mask and corresponding portions of the desired frame data.

Although the present invention has been shown and described with reference to specific embodiments thereof, it will be apparent to those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention.

8: Data processing device
10: Local Area Network
14: Storage device
18: Mainframe computer
132: Media access controller
134: Transceiver

Claims (21)

An improved power management method in a network of each computer station comprising a network interface controller, the method comprising: generating at least one programmed pattern in frame data, , Storing a corresponding predetermined response message in one of the controllers corresponding to one of the stations; Monitoring by the one of the controllers the frame data on the network; Detecting from the monitored frame data that the stored pattern matches a pattern in the monitored frame data in a controller of one of the controllers; Determining, in response to the detecting step of the match, whether the corresponding predetermined response message is stored; And sending the predetermined response message over the network to determine whether the corresponding predetermined response message has been stored.
2. The method of claim 1, further comprising: determining, by the controller, whether the station is in a hibernation state, in response to determining that the predetermined response message is not stored.
3. The method of claim 2, wherein the at least one programmed pattern and a corresponding response message comprise a plurality of pairs of the pattern and the corresponding response message.
4. The method of claim 3, wherein the response message in each pair corresponds to a different network protocol.
5. The method of claim 4, further comprising: placing one of the stations in a low-power state, wherein the storing, monitoring, detecting, determining, Is performed by a controller of one of the controllers corresponding to the one of the stations during the low-power state.
6. The apparatus of claim 5, further comprising: a controller, responsive to different ones of the stations, that is different from the controller of the one of the controllers, ≪ / RTI > further comprising the step of transmitting said frame data to said first terminal.
7. The method of claim 6, wherein the step of storing to the one of the controllers further comprises: identifying a different one of a plurality of the predetermined response message, a plurality of message lengths, and the predetermined response message, And storing a response buffer of a pair of message pointers corresponding to the mask.
An improved power management apparatus in a network of each computer station, comprising a network interface controller, the apparatus comprising: at least one programmed pattern in frame data; Means for storing a corresponding predetermined response message in one of the controllers corresponding to one of the stations; Means for monitoring the frame data on the network by the one of the controllers; Means for detecting in the one of the controllers that the stored pattern from the monitored frame data matches a pattern in the monitored frame data; Means for determining whether said corresponding predetermined response message is stored in response to detection of said match by said detection means; And means for transmitting the predetermined response message onto the network in response to the determination by the determination means to determine whether the corresponding predetermined response message is stored.
9. The apparatus of claim 8, further comprising means for determining, by the controller, whether the station is in a hibernation state, in response to determining that the predetermined response message is not stored.
10. The apparatus of claim 9, wherein the at least one programmed pattern and a corresponding response message comprise a plurality of pairs of the pattern and the corresponding response message.
11. The apparatus of claim 10, wherein the response message in each pair corresponds to a different network protocol.
12. The apparatus of claim 11, further comprising means for placing one of the stations in a low-power state, wherein the storing, monitoring, detecting, determining, Wherein the controller is one of the controllers corresponding to the one of the stations.
13. The apparatus of claim 12, further comprising: a controller, responsive to different ones of the stations, from a controller of one of the controllers different from the one of the controllers, Further comprising: means for transmitting the frame data to the control unit.
14. The apparatus of claim 13, wherein the means for storing to the one of the controllers comprises means for identifying a different one of a plurality of the predetermined response message, a plurality of message lengths, and the predetermined response message, And means for storing a response buffer of a message pointer pair corresponding to the mask.
A computer program product for use in improving the power management in a network of each computer station comprising a network interface controller, the computer program product comprising: at least one Computer readable program code means for storing the programmed pattern and a corresponding predetermined response message in one of the controllers corresponding to one of the stations; Computer readable program code means for monitoring by said controller one of said controllers for said frame data on said network; Computer readable program code means for detecting from the monitored frame data that the stored pattern matches the monitored frame data in the one of the controllers of the controller; Computer readable program code means for determining whether said corresponding predetermined response message has been stored and in response to said determining step of determining whether said corresponding predetermined response message is stored, Computer readable program code means for transmitting the predetermined response message to the network.
16. The computer program product of claim 15, further comprising computer readable program code means for determining, by the controller, whether the station is in a hibernation state, in response to determining that the predetermined response message is not stored.
17. The computer program product of claim 16, wherein the at least one programmed pattern and the corresponding response message comprise a plurality of pairs of the pattern and the corresponding response message.
18. The computer program product of claim 17, wherein the response message in each pair corresponds to a different network protocol.
19. The apparatus of claim 18, further comprising computer readable program code means for placing one of the stations in a low-power state, wherein the computer readable program code means for storing, monitoring, detecting, Wherein the one of the stations is executed by a controller of one of the stations corresponding to the one of the stations in the low-power state.
20. The apparatus of claim 19, wherein the controller of one of the stations corresponds to the one of the stations and is different from the controller of the one of the controllers to the one of the controllers corresponding to the one of the stations And computer readable program code means for transmitting the frame data.
21. The computer readable medium of claim 20, wherein the computer readable program code means for storing the one of the controllers comprises: a plurality of the predetermined response message, a plurality of message lengths, and the predetermined response message, And computer readable program code means for identifying a different one of the predetermined response messages and for storing a response buffer of a message pointer pair corresponding to a different frame data mask.

Images