TWI379192B - Buffering techniques for power management - Google Patents

Description

Fl379192 IX. Description of the Invention [Technical Field of the Invention] The present invention relates to a buffering technique for power management. [Prior Art] Power management for electronic devices such as computer systems plays an important role in preserving energy, managing heat dissipation, and improving overall system performance. Modern computer systems are designed to increase the trend to save important energy with power management when no reliable external power is available. Power management techniques allow the power supply to specific components of a computer system to be reduced or in a sleep mode, which requires less power than during operation, thereby reducing the total energy consumed by the device for a certain period of time. Energy conservation is especially important for mobile devices to conserve battery power. Even when a reliable external power source is available, precise power management within the computing system can reduce the amount of heat generated by the system, which can improve system performance. In general, computing systems perform better at lower ambient temperatures because critical components can be executed at higher speeds without damaging their circuitry. Therefore, there are many advantages to enhancing the power management of electronic devices. SUMMARY OF THE INVENTION In general, various embodiments are directed to providing a buffering technique for enhanced power management. In particular, some embodiments may be directed to power management techniques that conserve energy from an energy storage device such as a battery while the node is operating. For example, in one embodiment, a device such as a network device can include a power management module having a power management controller and a managed power system for coupling to the power management module. A managed power system can include a communication subsystem and a computing subsystem. The power management controller can be configured to switch the communication subsystem and the computing subsystem to a low power state to conserve energy. In various embodiments, the communication subsystem of the first node can be configured to process and store information received from other nodes within the communication system to increase the energy storage of the computing subsystem of the first node. In some embodiments, the communication subsystem can be extended for a period of time to maintain the low power state when the computing subsystem can be buffered and event information until ready for processing by the computing subsystem. This reduces the number of interrupts that are transferred to the computing subsystem when in a low power state, with each interrupt forcing the computing subsystem to return to a high power state to service the outage. Sometimes this technique is referred to as "interrupt combination J." For example, in one embodiment, the communication subsystem may further include a transceiver, a buffer, a watermark generator, and a buffer manager. The transceiver may be configured to use For communicating information on the network, the buffer may be coupled to the transceiver and configured to store the information packet of the transceiver during the idle period of the communication for generating a calculation idle period. For example, the communication idle period may represent The time interval during which the communication subsystem does not receive (or expect to receive) information from the network. For example, the calculation of the idle period may represent the time interval when the calculation of the secondary system does not receive (or expect to receive) information from the communication subsystem. Connected to a buffer and configured to generate a variable receive threshold. The buffer manager can be coupled to the buffer and watermark generator and configured to specifically address the variable receive threshold The stored information packet is transferred from the buffer-5-1379192 to the calculation subsystem. The variable reception threshold can be derived by calculation based on changing the communication power status information. As described in detail below, in this manner, the power management module can be implemented by implementing buffering techniques and/or logic that allows the communication subsystem and/or the computing subsystem to enter and remain in a low power state. Enhanced energy conservation of the managed power system while maintaining quality of service (QoS) and communication subsystems and/or other performance requirements of the computing subsystem. Other embodiments are illustrated and claimed. Various embodiments may include one or more components. Any structure configured to perform a particular operation may be included. Each element may be implemented as hardware, software, or any combination thereof, as required by a given set of design parameters or performance limitations, albeit by way of example in a particular topology. The finite element number is used to illustrate the embodiment, but this embodiment may include more or fewer elements in the alternative topology as needed for a given implementation. It must be noted that "one embodiment" or "an embodiment" Any reference to the features, structures, or characteristics described in connection with the embodiment is included in at least one embodiment. The appearances of the phrase "in one embodiment" and " 1 shows a block diagram of a communication system 100. In various embodiments, communication system 100 can include multiple nodes. A node may generally be included in communication system 1 任何 for any substance or logical entity used for communication information and may be implemented as hardware, software, or any combination thereof, as required by a given set of design parameters or performance limitations. Although Figure 1 may represent a finite number of nodes exemplified in a particular topology, it must be understood that more or fewer nodes may be used in different topologies for a given implementation. 1379192 In various embodiments, the communication system can include a wired communication system, a wireless communication system, or a combination of both, or form part of a wired communication system, a wireless communication system, or a combination of both. For example, communication system 100 can include one or more nodes im-m configured to communicate information over one or more types of wired communication links, such as wired communication link 140-1. Examples of wired communication links 14 0-1 may include, but are not limited to, wires, cables, bus bars, printed circuit boards (P CB ), Ethernet connections, peer-to-peer (P2P) connections, backplanes, fiber optic switches, semiconductor materials, Twisted pair, coaxial cable, fiber optic connections, and more. Communication system 100 can also include one or more nodes 11 0-1 -m configured to communicate communications over one or more types of wireless communication links, such as wireless shared medium 140-2. Examples of wireless shared medium 140-2 may include, but are not limited to, a radio channel, an infrared channel, a radio frequency (RF) channel, a wireless fidelity (WiFi) channel, a portion of an RF spectrum, and/or one or more authorized or unlicensed bands. . In the latter case, the wireless node may include one or more wireless interfaces and/or components for wireless communication, such as one or more radios, transmitters, receivers, transceivers, chipsets, amplifiers, filters , control logic, network interface cards (NICs), antennas, antenna arrays, and more. Examples of antennas may include, but are not limited to, internal antennas, omnidirectional antennas, monopole antennas, dipole antennas, 'end-fed antennas', cyclically polarized antennas, microstrip antennas, diversity antennas, dual antennas, antenna arrays, and the like. In one embodiment, a particular device may include an antenna array of multiple antennas to implement various adaptive antenna techniques and spatial diversity techniques. As shown in the embodiment of Figure 1, the communication system 1 includes a plurality of nodes 1379192 110-l-m. A node may include or be implemented as any type or mobile electronic device or resource, including a network device, a network endpoint network infrastructure device, a cellular radio network device, a computer system, a computer subsystem, a computer, a workstation , terminal -- server, personal computer (PC), laptop, super laptop: portable computer, handheld computer, personal digital assistant (PDA), bee, smart phone, router, switch, bridge , gate, #. device, microprocessor, integrated circuit, programmable logic (PLD), digital signal processor (DSP), processor, gate, register, microprocessor, Integrated circuits, semiconductor packages, transistors, etc. In some embodiments, some nodes 110. represent heterogeneous network devices. For example, in one embodiment, a 110-l_m may include a mobile computer mobile computer system (e.g., a laptop, a handheld computer, a smart phone, a cellular telephone, etc.) using, for example, one or more batteries. # Although - some nodes 1 ΙΟ-1-m may contain different network devices Each node UO-1-m may contain elements such as nodes 110-1-m. For example, nodes 1 1 0 -1 - m may each include various power components to implement a power management method operable to perform power management operations of node 1 -. For example, in the embodiment shown in FIG. 1, the 101-1 may include the managed system 120-1 coupled to the power management module 130-1. The power management module 130-1 is operable to communicate power state information with one of the second node nodes 1 10-2-m over the established communication connection via the ports 140-1, 140-2. In the fixed equipment, processing system, servo computer, nested power grid, electric circuit, road, logic, crystal -1 - some of the node source of each type of power, but the general number of management elements 1 0 -1 - m One-node power system connection (for example, the power state of the managed power system 12〇-1 in the power management module 130-1 can manage the first node no in the operation -8- 1379192. The power status information can include The past, present or future power status of one or more portions of the managed power system 120-1 of the first node 110-1. Thus, the portion of the managed power system 120-1 is interchangeable with power status information, Improving or enhancing the power state management of the first node 1 1 0 - 1. For example, the power management module 13 3 〇 1 can power the secondary systems 210-1, 230-1 of the managed power system 120-1 The management operation synchronizes, such as during operation or expected operation of the communication element of communication subsystem 210-1, to cause the computing element of computing subsystem 230-1 to be in a low power state for a given power state, and vice versa. 1 1 0 · 1 Figure 1 The only node shown to include the managed power system 120-1 and the power management module 130-1, but it must be understood that each node ΙΙΟ-1-m may include the same or similar managed power system 120-1- π and power management module 130-1-p. For example, node 11〇_2 may include managed power system 120-2 coupled to power management module 130-2; node 110-3 may include component 120-3 130-3, etc. Further, the description and examples of the structure and operation provided with reference to the managed power system 120-1 and the power management module 1 30-1 can also be applied to other nodes 1 1 O-2-m. Corresponding components. Example embodiments of managed power system 1 20-1 -η and power management module 1 3 0-1 -ρ can be described in more detail with reference to Figure 2. Figure 2 shows managed power system 12 and power A more detailed block diagram of the management module 130. In the embodiment illustrated in Figure 2, the managed power system 120 can include a communication subsystem 210 and a computing subsystem 230. Although Figures 2-9-1379192 can represent examples. a limited number of power management components in a particular configuration, but must be known to be used in different configurations More or less power management components are used for a given implementation. In various embodiments, managed power system 120 can include any electrical or power that consumes power 232 and a node 110-lm that is suitable for power management operations. Electronic components. Power management techniques allow the power supply to a particular component of an electronic device or system (eg, a computer system) to drop or be in a sleep mode that requires less power than active operation, thereby reducing the total energy consumed by the device for a certain period of time. Power management techniques can be implemented by power gating and/or clock gating of various hardware components of managed power system 120. More specifically, the managed power system 120 can include various electrical or electronic components that can operate at various power states of the node 110-lm, which pull multiple levels of power from the power source 232, such as by the power management module 130. The controller of the power management controller 234. The various power states can be defined by any number of power management methods. In some instances, the power state can be defined, for example, according to the Advanced Fabric and Power Interface (ACPI) family of specifications, including their subsequent content, versions, and modifications. In one embodiment, the power state can be defined, for example, according to ACPI Revision 3_0a ("ACPI 3.0a Specification Revision") of December 30, 2005. The ACPI Series specification defines various power states of an electronic device, such as global system state (Gx state), device power state (Dx state), sleep state (Sx state), processor power state (Cx state), device and processor performance state. (Px status) and so on. It must be understood that other power states that change the power level -10- 1379192 can be implemented in response to a given set of design parameters and performance limitations. This document is not intended to limit the embodiments. In some embodiments, various electrical or electronic components of nodes Π0-1-m suitable for power management operations may generally be grouped or organized into a communication subsystem 210 and a computing subsystem 230. However, it is understood that this is to provide a secondary system 2 1 0, 230 for illustrative purposes, but is not limited thereto, and the managed power system 120 may include a node 1 1 0-1 suitable for power management operations of the power management module 130. -m of various electrical or electronic components. For example, node 1 10-1-m can generally include a computer monitor or display, such as a digital electronic display or an analog electronic display. Examples of digital electronic displays may include electronic paper, digital tube display, vacuum fluorescent display, light emitting diode display, electric field light emitting display, plasma display panel, liquid crystal display, thin film transistor display, organic light emitting diode display, surface Conducting electron emitter displays, laser television displays, carbon nanotubes, nano crystal displays, and the like. An example of an analog electronic display can include a cathode tube display. When the operating system detects that the computer has not received input from the user for a set period of time, the computer monitor is usually set to sleep mode. Other system components may include digital cameras, touch screens, video recorders, recorders, storage devices, vibrating elements, oscillators, system clocks, controllers, and other platform or system architecture devices. When these other system components are not in use, they can also be set to sleep or power down to conserve energy. The computer system monitors the input device and the wake-up device as needed. This document is not intended to limit the embodiments. In various embodiments, managed power system 120 can include communication subsystem 210. The communication subsystem 210 can include various communication components configured to -11 - 1379192 for communicating information and performing communication operations between the nodes 110-hm. Examples of suitable communication components can include any electrical or electronic component designed to communicate information over communication links 140], 140-2, including but not limited to radios, transmitters, receivers, transceivers, chipsets, amplifiers , filters, control logic, interfaces, network interfaces, network interface cards (NICs), antennas, antenna arrays, digital signal processors, baseband processors, media access controllers, cell units, and more. In various embodiments, the communication subsystem 210-1 can include one or more transceivers 204-1-r that can operate at different communication rates. The transceivers 204-1-r can include various wired media formats (eg, copper, single mode fiber 'multimode fiber, etc.) and wireless media types (eg, RF spectrum) that can be used in the communication links 140-1, 140-2. Any communication component that transmits and receives information. Examples of transceivers 204-1_r may include various Ethernet-based PHY devices, such as fast Ethernet PHY devices (eg, 100Base-T, 100Base-TX, 100Base-T4, 1OOB ase-T2, 1 OOBase) -FX, 1 OOBase-SX, 1 OOBaseBX, etc.), Gigabit Ethernet (GbE) PHY device (l〇〇〇Base-T, lOOOOBase-SX, 1 OOOBase-LX, 1 0 0 0 B As e - BX 1 0 , 1 OOOBase-CX, 1 000Base-ZX, etc.), 10 GbE PHY devices (eg lOGBase-SR, 1 OGBase-LRM, 1 OGBase-LR, 1 OGBase-ER, lOGBase-ZR ., 1 0GBase-LX4, 1 0GBase-CX4, 1 OGBase-

Kx, 10GBase-T, etc.), 100 GbE PHY devices, etc. The transceiver 2 04-1 -r can also include various radio or wireless PHY devices, such as for mobile broadband communication systems. Examples of mobile broadband communication systems include, but are not limited to, -12-1379192 in systems compatible with the Institute of Electrical and Electronics Engineering (IEEE) standards, such as the IEEE 802.11 standard for wireless local area networks (WLAN) and its variations, The Wireless Metropolitan Area Network (WMAN) and its changing IEEE 802.16 standard, the IEEE 802.20 standard for Mobile Broadband Wireless Access (MBWA) and its changes. The transceivers 204-1-r can also be implemented in various other types of mobile broadband communication systems and standards, such as the Universal Mobile Telecommunications System (UMTS) system family of standards and their variations, the Code Division Multiple Access (CDMA) 2000 system family of standards. And its changes (such as CDMA2000 lxRTT, CDMA2000 EV-DO, CDMA EV-DV, etc.), high-performance RF metropolitan area network (HIPERMAN) developed by the European Telecommunications Standards Institute (ETSI) Broadband Wireless Access Network (BRAN) System series standards and their changes, wireless broadband (WiBro) system series standards and their changes, mobile communications global system (GSM) and general packet radio frequency service (GPRS) system (GSM/GPRS) series of standards and their changes, global development The Enhanced Data Rate (EDGE) System family of standards and their variations, the High Speed Linked Packet Access (HSDPA) system family of standards and their variations, the High Speed Orthogonal Frequency Division Multiplexing (OFDM) Packet Access (HSOPA) system family of standards and Its changes, high-speed uplink packet access (HSUPA) system series standards and their changes. This document is not intended to limit the embodiments. In various embodiments, controller 208 is operable to switch the communication rate using a single-receiver (e.g., 204-1) or a plurality of transceivers 204-1-r. The controller 208 can be implemented as any computing element or logic device capable of performing logical operations, such as a processor 'microprocessor, chipset, controller, -13-1379192 microcontroller, embedded controller, media access The controller, baseband controller, etc. "receivers 204-lr can operate individually or collectively at different communication rates or link rates. In one embodiment, for example, a single transceiver 204-1 can operate at various communication rates. In another embodiment, for example, the first transceiver 204-1 can operate at a first communication rate, the second transceiver 204-2 can operate at a second communication rate, and the like. The controller 20 8 can switch the first transceiver 204_1 from the first communication rate to the second communication rate, or switch from operating the first transceiver 204-1 to operating the second transceiver 204-2 to complete the The required communication rate. Controller 208 can switch the communication rate according to a control strategy, such as in accordance with, for example, one or more energy efficient Ethernet (EEE) control strategies. Controller 208 can also switch the communication rate based on instructions from buffer manager 216. In various embodiments, communication subsystem 210-1 can include one or more buffers 206-1-t that are managed by buffer manager 216. The buffers 206-1-t are operable to store network packets received by the transceivers 204-Ι-r or are ready for transmission by the transceivers 204-1-r. For example, when the secondary system 210-1 or the secondary system 230-1 enters a low power state and is therefore unable to communicate or process the packet, the buffers 206-1-t can be used to buffer the packet. In another example, the buffers 206-1-t can be used to buffer the packet until the communication rate of the transceiver has been completely switched or modified, since the communication rate of the switching transceiver 204-1-r is usually not instantaneous. . For example, buffers 206-1-t can be implemented as a standard first in first out (FIFO) sequence. The buffer manager 216 can implement various types of buffer logic to manage the operation of the buffers 206-1 -t. - 14 - 1379192 In various embodiments, managed power system 120 can include computing subsystem 230. The computing subsystem 230 can include various computing elements configured to process information and perform computational operations of nodes 1 1 0· 1 -m. Examples of suitable computing elements can include any electrical or electronic component designed to process information, including but not limited to processors, microprocessors, chipsets, controllers, microcontrollers, embedded controllers, clocks, oscillators , audio card, video card, multimedia card, peripheral device, memory unit, memory controller, video controller, audio controller, multimedia controller, etc. In various embodiments, the power management module 130 can include a power source 232. Power source 232 can be configured to provide power (generally) to the components of nodes 110-1-m, and in particular to managed power system 120. In one embodiment, for example, power source 2 3 2 is operable to provide varying levels of power to communication subsystem 210 and computing subsystem 230. In various embodiments, the power source 23 2 can be implemented by a rechargeable battery, such as a removable and rechargeable lithium ion battery, to provide direct current (DC) power, and an alternating current (AC) adapter to Pull power from a standard AC mains supply. In various embodiments, power management module 130 can include power management controller 234. Power management controller 234 generally controls the power consumption of managed power system 120. In one embodiment, the power management controller 234 is operable to control the power provided to the communication subsystem 210 and the varying level of the secondary system 230 based on the particular power state. For example, the power management controller 234 can modify, switch, or transform the power level provided by the power supply 23 2 to the secondary system 210, 230 to a higher or lower power level, thereby effectively modifying the secondary systems 210, 230. Power status. -15- 1379192 In various embodiments, power management module 130 can include one or more power control timers 236. The power management controller 234 can use the power control timer 236 to maintain a particular power state for a given power state period. The power state period may represent the time interval defined by a node or a portion of a node at a given power state. For example, the power management controller 234 may switch the computing subsystem 230 from a high power state to a low power state for a defined time interval 'and when the time interval elapses, the computing subsystem 230 is switched back to a high power state. .

In order to coordinate the power management operations of the nodes 11 0-1-m, the communication subsystem 2 1 0, the computing subsystem 23 0 and the power management module 130 can be connected via the communication bus 220 and the individual power management interfaces 214-1, 214. -2 and 214-3 to connect various power management messages 24 0 - 1 -q. In order to manage the power of all devices in the system, the operating system typically uses standard techniques for communicating control information over special input/output (I/O) interconnects. Examples of various input/output (I/O) interconnects suitable for implementation as communication bus 220 and associated interface 214 may include, but are not limited to, Peripheral Component Interconnect (PCI), PCI Express (PCIe), CardBus, Universal Serial Bus (USB), IEEE 23 94 FireWire, and more. Referring again to FIG. 2, the communication subsystem 2 1 0 can include a network status module 2 1 2 » The network status module 2 1 2 can be configured to monitor a particular status or feature of the communication subsystem 210, such as the communication connection 250 -1 -V traffic activity, capability information, and other operations of various communication components of communication subsystem 210. The network status module 2 1 2 can transmit the communication power management message 2 4 0 -1 - q to the power management module 1 300 using the measured characteristics. The power management module 1 3 0 -16-1379192 can generate power status information 260 of the managed power system 120 based in part on the communication power management messages 240-1-q. Similarly, the 'computing subsystem 323 can include a computing state module 23 2 ^ The computing state module 23 2 can be configured to monitor a particular state or feature of the computing subsystem 030, such as system activity level, capability information, and calculations. Other operations of various computing elements of system 230. The computing status module 23 2 can communicate the calculated power management messages 240-1-q to the power management module 130 using the measured features. The power management module 130 can generate the power status information of the managed power system 120 based on the calculated power management information 240-lq. In general operation, the power management module 1 30-1 can be connected to the node 1 1〇-1. Power management is performed by the operation of a portion of the managed power system 120-1, which is based on power status information received from other portions of the first node 110-1. In some examples, for example, the power management module 130-1 of the node 110-1 is operable to communicate from the communication network 220 to the network status module of the communication subsystem 210-1 of the managed power system 120-1. 2 1 2 Receive communication power status information. The power management module 130-1 can manage various power states of the computing subsystems Ο-ΐ of the managed power system 120-1 of the node 1 10-1 based on the communication power state information of the communication subsystem 210-1. The power management module 130-1 and the secondary systems 210-1, 23 0-1 can communicate communication power status information in the communication bus 220 according to various communication bus communication protocols. The communication power status information may represent information explicitly or implicitly regarding the power status of the communication subsystem 2 10 . The communication power status information may also represent various characteristics or attributes of the power status of the -17- 1379192 communication subsystem 210, such as status periods, idle periods, restart wait times, and the like. In the embodiment, for example, the 'communication power status information may include, but is not limited to, a power status parameter, a communication idle period parameter, a communication restart wait 'parameter, or a power status period. The communication idle period parameter represents the amount of time that the network or communication subsystem 210-1 will maintain a given power state.  Set the time interval. The communication idle period parameter allows the secondary system 2 1 (# 23 0-1 to enter and leave the low power state in a deterministic manner. The communication restart waiting time parameter represents the network connection or the communication subsystem 210-1 departs the power state and enters the high state. The amount of time or defined interval required for the power state. The communication restart start time parameter allows the secondary system 2 1C 2 3 0-1 to decide how fast it can expect the communication subsystem 210-1 to wake up ready for providing The service that is transmitted out can be restarted by the power management information 240-1 -q on the communication bus 220 to communicate the communication idle period. # In various embodiments, the network status module 212 can be configured. The communication idle period parameter and the restart wait time parameter are generated according to the capability of the communication subsystem 210-1. For example, the communication subsystem 210-1 can receive various buffers to store the received from the communication connection 250-1 -V. (such as a network packet) and forward the information for service and processing by the computing subsystem _ 1. In another embodiment, the communication subsystem 210 can implement various buffers to store the slave The communication bus 220 receives the message (such as a network packet) and forwards the information for use in the communication connection 210-1 via the communication link 140-1, 140-2 in the communication connection 250-1. Or 1-1 > new opening to time -1 ' and the number of quasi-political signals and the implementation of the information of the root communication communication 230-1 also V -18 · 1379192 to connect this information to other nodes ll 〇 -2- m. In yet another example, the communication subsystem 210-1 may include various wired or wireless transceivers that operate at different communication speeds, such as IEEE 802. 3-2005 standard 10 billion bit Ethernet (lOGbE or lOGigE), IEEE 802. 3ba recommends a standard of 100 billion bit Ethernet (lOOGbE or lOOGigE) and so on. In yet another example, communication subsystem 210-1 can include various processors operating at different speeds, such as a baseband or communication processor. In yet another example, the network status module 2 1 2 can monitor the rate at which information is received over the communication connections 250-1 -V via the communication links 14 0-1, 140-2. In this example, the network status module 2 1 2 of the communication subsystem 2 1 0 1 can monitor the communication links 140-1, 140-2 to measure the arrival time of the packets with each other. Other examples of communication capabilities may include other network traffic load measurements (eg, synchronous traffic, asynchronous traffic, burst traffic, etc.) on the communication links 140-1, 140-2, signal to noise ratio (SNR), Received Signal Strength Indication (RSSI), communication bus 220 flux, physical layer (PHY) speed, other nodes received via one or more power management packet data units (PMPDUs) 150-1-s ΙΙΟ-2- m power status information 260 and so on. The network status module 2 1 2 can evaluate the communication capabilities of these and other network or communication subsystems 210-1 and generate appropriate communication idle period parameters and communication re-establishment based on the communication capabilities of the evaluated communication subsystem 210-1. Start waiting time parameter. In various embodiments, nodes 1 1 0-1 -m can use power state information to enhance power management operations of a given node 1 1 〇-1 _ m to improve energy conservation (eg, increase battery life or reduce battery size) ) 'Cooling or -19- 1379192 overall system performance. In one embodiment, for example, the network status module 212 of the communication subsystem 210-1 can monitor the communication links 140-1, 140-2 and various communication components (eg, radio, baseband processor, chipset, memory) Unit, etc.) to determine the communication power status information of the communication subsystem 210-1. The network status module 212 can transmit power management messages 240-1-q to the power management module 130-1 via the communication power status information on the communication bus 220 and the interface 2 14-1 ' 2 14-3. The power management module 130-1 can receive the power management messages 240-1-q and retrieve the communication power status information from the power management messages 240-1 -q. The power management module 130-1 can manage the power state of the secondary system 23 0-1 based on the communication power status information of the communication subsystem 210-1. For example, the power management module 130-1 may modify the power level of the computing subsystem 123-1 of the managed power system 12〇-1 using the communication power status information of the communication subsystem 210-1, and modify the power level from the first power level. It is the second power level. In addition, the power management module 1 3 Ο-1 can modify the power level of the calculation subsystem 023 by using the communication power state information of the communication subsystem 210-1, and modify the power level from the first power level to the first power level. The second power level passes through a defined time interval and is referred to as the power state period. Whenever the communication subsystem 210-1 is in a low power state, the communication subsystem 210-1 can still continue to receive information packets from the other nodes 110-2-m on the communication links 14〇-1, 140-2, and The receiving information packet from the computing subsystem 230 is used to transmit it from the computing subsystem 230-1 to the other nodes by the communication subsystem 210-1 at the communication link 14〇-1, 14〇-2. m. If each packet received by the communication subsystem 210-1 is directly transferred to the computing subsystem 230-1 for processing, then the computing subsystem 23 0_ -20-1379192 1 will continuously have to leave the low power state and enter high. Power status to handle each packet. This will consume a considerable amount of energy from the power supply 23 2 . To address these and other issues, the communication subsystem 210-1 can implement an interrupt combining technique. For example, the communication subsystem 210-1 can be configured to use one or more buffers 206-1-t to store or buffer incoming packets, and the buffer manager 216 can decide when to release or forward the stored packets to the calculation. The secondary system 23 0-1 is used for processing. The buffer manager 216 can determine when to transfer the packet from the buffer 206-1 -t to the calculation time according to a buffer management strategy designed to increase the energy storage of the calculation subsystem 2-3 and the entire node 1 〇-1. System 2 3 0-1. In various embodiments, the buffer manager 216 can implement a buffer management policy designed to allow the communication subsystem 210-1 and/or the computing subsystem 230-1 to enter and remain in a low power state for a longer period of time. While maintaining the communication subsystem 20-1 and/or calculating the Q〇S, throughput, and other performance requirements of the subsystem 230-1. In some embodiments, the buffer management policy can include various buffer management rules to determine when to forward packets from buffers 206-1 -t to computing subsystems 23 0-1. In one embodiment, for example, the buffer manager 216 can implement buffer management rules to store packets in the buffers 206-1-t during communication idle periods and to satisfy any combination of the following four buffer management conditions. When the stored packet is transferred from the buffer 206-Ι-t to the computing subsystem 23 0-1: (1) when the number of packets stored by one or more buffers 206-1 -t exceeds the variable reception threshold 値(2) When the buffer unload is paused, it has expired; (3) when the buffer unload event signal is received; and/or (4) when the communication is idle, the parameter is less than the communication idle period - 21 - 1379192 limit. It must be understood that these four buffer management conditions are exemplified by 'not limitation. This document is not intended to limit the embodiments. Buffer manager 216 can utilize one or more buffer management conditions in various combinations to trigger the release or transfer of any packets stored in buffers 206-1-t to computing subsystem 230" for processing. In one embodiment, for example, the buffer manager 216 may forward the packets stored in the buffers 206-1 -t to the computing subsystem 230-1 using direct memory access (DMA) techniques. The packet acceleration is moved from the buffer 206-1-t to the memory unit 2 3 4 used to calculate the secondary system 2 3 0-1. The buffer manager 216 can then issue an interrupt to the computing subsystem 230-1 (e.g., a processor) to indicate that the packet is within the memory unit 234 and is ready for processing by the secondary system 230-1. In one embodiment, when the number of packets stored by one or more buffers 206-1 -t exceeds a variable receive threshold, buffer manager 216 can forward the stored packets from buffers 206-1 -t To calculate the secondary system 23 0-1. For example, the communication subsystem 2 1 0-1 can include a watermark generator 2 1 7 coupled to the buffer manager 216 and the buffers 206-1 -t. Watermark generator 2 1 7 is operable to generate a variable receive threshold. The variable reception threshold may include a threshold or watermark defined for the buffer 20 6-1 -t. The variable access threshold can be calculated, derived, or determined as a function using at least the following three inputs: (1) receive data rate parameters; (2) buffer size parameters; and (3) communication restart start time parameters. The Receive Data Rate parameter may represent the communication rate of one or more of the transceivers 204-1-r. The buffer size parameter may represent the size of one or more buffers 206-1-t. Communication Weight -22- 1379192 The New Start Waiting Time parameter represents the amount of time or defined time interval (eg, 1 ms) required for the network link or communication subsystem 2 to leave a given power state and enter a high power state. For these or other communication parameters, the watermark generator 217 can generate variable reception thresholds periodically, continuously, or as needed to ensure that the variable reception threshold accurately reflects changes in communication rate, latency, and other network traffic. The buffer manager 216 can receive the variable receive threshold from the watermark generator 217 and receive the current buffer usage parameters from the buffer 20 6 - 1 t, and the current buffer usage parameters and the variable receive threshold.値 doing a comparison, and then initiating DMA transfer based on the comparison result. For example, when the number of information packets stored in the buffer exceeds the variable reception threshold, the buffer manager 216 initiates DMA transfer. In one embodiment, when buffering When the device unload pause has expired, the buffer manager 216 can forward the stored packet from the buffer 206_l-t to the computing subsystem 230-1. For example, the communication subsystem The 2 1 0-1 may include a buffer timer 218 to be coupled to the buffer manager 216. The buffer timer 218 may be a hardware or software timer to set or measure the defined time interval. The buffer unload pause can be set to load or load the buffer timer 218. The buffer timer 218 can monitor or perform the countdown until the buffer unload pause has expired. The buffer manager 216 can be configured to unload when the buffer is unloaded. When the pause has expired, the stored information packet is forwarded from buffer 206-1-t to computing subsystem 230-1. When buffer manager 216 is used in combination with another buffer management condition (such as variable receive The buffer management -23 - 1379192 216 can forward the stored packet when the buffer unloading pause expires before the buffer 206-1 -t stores the number of packets exceeding the variable reception threshold. In one embodiment, when the buffer manager 216 receives the buffer unload event signal, the buffer manager 216 can forward the stored packet from the buffer 206-Ι-t to the computing subsystem 22-1. For example, through The signaling system 210-1 can randomly receive event signals from portions of the node 110-1 that facilitate the buffer manager 216 to forward packets from the buffers 206-1 -t to the computing subsystem 230-1. For example, assume that computing subsystem 230-1 implements a chipset computer architecture that includes a memory controller hub (MCH) and an input/output (I/O) controller hub (ICH), such as Intel Corporation of Santa Clara, California ( "North Bridge" and "South Bridge" controller hubs manufactured by Intel® Corporation. Further, it is assumed that MCH and ICH use direct media interface (DMI) and associated links to connect information. Regarding the calculation of other portions of the subsystems 230-1, the DMI connections can be in various power states, such as a high power state L0 and a low power state L1. When the DMI link leaves the low power state L1 due to the activity of other devices to the high power state L0, the calculation subsystem 2300-1 can transmit a buffer offload event signal to the buffer manager 216. The buffer unload event signal can indicate to the buffer manager 216 that the computing subsystem 23 0-1 is already in a high power state and is active, and thus the buffer manager 216 can use the computing subsystem 2 3 0 - 1 The high power state is to transfer the packet from the buffer 206-l_t to the memory 234. When the buffer manager 216 is used in combination with another buffer management condition (such as a variable receive threshold), the buffer is received before the number of packets stored in the buffer 202-1-t exceeds the variable reception threshold. When the event signal is unloaded, the Buffer Manager-24-1379192 216 can forward the stored packet. In one embodiment, when the communication idle period parameter is less than the communication idle period threshold, the buffer manager 216 can implement a buffer management rule to transfer the stored packet from the buffer 2 06-1 -t to the computing subsystem. 2 3 0- 1. The communication idle period threshold may include a defined threshold for the communication idle period parameter. In general, the higher the parameters during communication idle, the lower the communication rate of the information on the communication links 140-1, 140-2, and vice versa. The communication idle period is configurable and can be set to determine when the communication idle period parameter is low enough to indicate a high communication rate, which does not need to tolerate the buffer 206-1 -t Extra waiting time. The buffer manager 216 can be configured to disable one or more of the buffers 2 06-1 -t to avoid buffers 206-1 -t when the communication idle period parameter is less than the communication idle period threshold Information packet. In some embodiments, the buffer manager 216 can be configured to send a signal to the controller 208 to modify the communication rate of the transceivers 204-1-r. For example, buffer manager 216 can instruct controller 208 to modify the communication rate based on its FIFO size and/or the amount of energy left. In one embodiment, for example, buffer manager 216 can receive or maintain buffer size parameters, energy measurement parameters, or both. The buffer size parameter can represent the FIFO size or FIFO remaining capacity. The energy measurement parameter can represent the energy remaining for a power source such as power source 232. The buffer manager 216 can be configured to issue a request to the controller 20 8 to adjust the communication rate of the transceivers 204-1-r based on the buffer size parameters. Similarly, the buffer manager 216 can be configured to receive power management messages having energy -25 - 1379192 measurement parameters from the power management controller 234 and to communicate the requirements based on the energy measurement parameters to adjust the transceivers 204·1·γ Communication rate. FIG. 3 illustrates a logic flow 300 in accordance with one or more embodiments. Logic flow 300 may be performed by various systems and/or devices, which may be implemented as a hardware 'software' and/or any combination thereof, as required by a given set of design parameters or performance limitations. For example, logic flow 300 may be implemented by a logic device (e.g., a processor) and/or logic (e.g., instructions, data, and/or code) executed by the logic device. To understand and not limit, logic flow 300 is illustrated with reference to Figures 1 and 2. The logic flow 300 can illustrate various operations of the nodes 1 10-1-m in a general manner, and in particular, the managed power system 120 and the power management module 130. As shown in FIG. 3, at block 302, logic flow 300 can modify the communication subsystem and calculate the power state of the subsystem, from a high power state to a low power state. At block 3 04, logic flow 3 00 may store the information packet in a buffer of the communication subsystem during communication idle to generate a calculated idle period. At block 306, logic flow 300 can generate a variable receive threshold for the buffer. At block 308, logic flow 300 can forward the stored information packet from the buffer to the computing subsystem based on the variable receive threshold. This document is not intended to limit the embodiments. In one embodiment, at block 302, logic flow 300 may modify the communication subsystem and calculate the power state of the subsystem, from a high power state to a low power state, and from a high power state to a low power state. For example, power management controller 234 can receive power management messages 240-1-q having calculated power state information generated by computing state module 23 2 and generated by -26- 1379192 network state module 2 1 2 Communication power status information. Power management controller 234 can receive power management messages 24〇_l-q from communication bus 220 via interfaces 214-1, 214-2, and 214-3. The power management controller 234 can process the power management messages Μ 0-1-q and determine appropriate communication power states (eg, NL0, NL1, NL2, etc.) and appropriate computing power states (eg, SO, SOil, S0i2, S0i3, S4 and so on). The power management controller 234 can transmit the communication power state and calculate the power state to the individual subsystems 210-1 via the power management messages 240-1-q on the communication bus 220 and the interfaces 214-1, 214-2, and 214-3. 210-2, and the secondary systems 210-1, 210-2 may thus modify their individual power states. In one embodiment, at block 304, logic flow 300 may store the information packet in a buffer during communication idle for use in the communication subsystem to generate a computational idle period. For example, the communication idle period may represent a time interval when the communication subsystem 210-1 does not receive (or expect to receive) information from the network on the communication links 140-1, 140-2. For example, the calculation idle period may represent a time interval when the computing subsystem 230-1 does not receive (or expect to receive) information from the communication subsystem 210-1. For example, the communication subsystem 2 1 0 -1 can receive the communication idle period parameter from the power management message 240-1-q for the communication subsystem 210. 1. It is transmitted by the power management controller 234. In this example, the communication idle period parameters can be calculated by the power management controller 234 using the power status information received from the secondary systems 210-1, 230-1. In another example, the communication subsystem 2 1 0 -1 can receive the communication idle period parameter from the network status module 2 1 2 . In either case, the communication subsystem 2 1 0 -1 can enter a low power state represented by the communication power state for a period of time -27 - 1379192 (as defined by the communication idle period parameter). The low power state can be entered directly by reducing the power to all of the communication elements of the communication subsystem 210-1 or indirectly by modifying the communication rate of the transceivers 204-1-r. Once the communication subsystem 210-1 enters a low power state, the buffer manager 216 can store the information packets in one or more receive buffers 206-1 -t during the communication idle period (as defined by the communication idle period parameter). Used in communication subsystem 210-1 to generate a computational idle period for computing subsystem 230-1. In some examples, communication subsystem 210-1 can calculate a idle period for computing subsystem 230-1, such that computing subsystem 230-1 can operate accordingly, such as switching to a low power state for a time interval, corresponding to Expect to calculate the idle period. In one embodiment, at block 306, logic flow 300 may generate a variable receive threshold for the buffer. For example, the watermark generator 217 can receive: a receive data rate parameter, a buffer size parameter, and/or a communication restart wait time parameter, and generate a variable receive threshold of the receive buffer 206-1-t according to the parameters ( For example, a buffer watermark). The watermark generator 217 can output the variable reception threshold to the buffer manager 216. In one embodiment, at block 308, logic flow 300 may forward the stored information packets from the buffer to the computing subsystem in accordance with the variable receive threshold. For example, the buffer manager 216 may receive a variable reception threshold from the watermark generator 2 17 , set the buffers 206 - 1 -t with the variable reception threshold, and may be periodic or non-periodic The number of information packets stored in the receive buffers 206-1 -t is compared to the variable receive threshold. After the number of information packets stored in the receive buffers 206-1 -t meets or exceeds the variable -28 - 1379192 receive threshold, the buffer manager can transfer the contents of the receive buffer 2 〇 6-lt via the DMA transfer mode. The billion unit 234 of the secondary system 230 is calculated for further processing. FIG. 4 illustrates a logic flow 4〇〇 in accordance with one or more embodiments. Logic flow 400 may be performed by various systems and/or devices, which may be implemented as hardware, software, and/or any combination thereof, as desired for a given set of design parameters or performance limitations. For example, logic flow 400 can be implemented by a logic device (e.g., a processor) and/or logic (e.g., instructions, data, and/or code) executed by the logic device. To understand and not limit, logic flow 400 is illustrated with reference to Figures 1 and 2. The logic flow 400 can illustrate the various operations of the node ′ in a general manner and specifically illustrates the managed power system 120 and the power management module 130. As shown in FIG. 4, logic flow 400 begins at block 420, wherein watermark generator 217 calculates or recalculates the variable by receiving data rate parameters, buffer size parameters, and/or communication restart latency parameters. Receive a threshold (such as a buffer watermark) and then fabricate a buffer watermark trigger. This allows the buffer manager 216 to react to various link speeds and delays. The buffer manager 2 16 can utilize the communication idle time parameter to determine how long the communication subsystem 210-1 and/or the computing subsystem 230-1 can be maintained in a low power state. Since the parameters do not have the expected traffic entering during the idle period of communication, the secondary systems 210-1, 230-1 can be electrically gated (if there is not enough time to enable the separate components of the secondary system 2 1 0 - 1 , 2 3 0 - 1 Do this economically). For example, a communication component implemented as a wireless transceiver (such as a transceiver 2〇4·1-〇 typically requires at least 8 ms of communication idle period to reference -29-1379192 to become a low power state. When the communication is idle period parameter Below 8 ms, this condition indicates that the incoming traffic is at a high data rate. According to this assumption, in the diamond block 404, when the communication idle period parameter is less than 8 ms, the buffer 206-1 is enabled at block 406. t disables to eliminate the extra delay (latency) that the buffer 2 0 6-Ι-t may cause. In diamond block 4 04, when the communication idle period parameter is greater than 8 ms, at block 408, anyway, The buffer manager 216 can set the variable receive threshold of the buffers 206-1 -t and the buffer offload pause of the buffer timer 218. At block 410, the buffer manager 216 can begin from one or more. The multi-receiver 204-1-r receives the packet and buffers the received packet in block 412 in one or more buffers 206-1-t until one or more buffer management conditions become available in diamond block 414 TRUE. In one embodiment, 'diamond block 414 can Estimate at least 4 conditions, including: (1) The first condition (condition 1) is TRUE, according to the variable reception threshold; (2) the second condition (condition 2) is TRUE, suspended according to buffer unloading値(3) The third condition (Status 3) is TRUE, according to the receipt of the buffer unloading event signal; and (4) The fourth condition (Status 4) is TRUE 'According to the event counter exceeding the third condition (Status) c) The 'for example' buffer manager 216 can use the various system events (eg, transfer interrupts) from the drive or the DMI link state between the ICH and the MCH as an input to trigger unbuffered from the receive buffer 206- A packet of 1 - t. For example, when the DMI is changed from L1 to L0 because of activity from other devices, its signaling to the communication subsystem "-" indicates that the host system is already active, and thus the buffer manager 216 This machine -30- 1379192 should be used to de-buffer the buffered packets from the receive buffer 206-lt if possible. The fourth condition (condition 4) is explained in more detail with reference to Figure 5, such as Referred to by reference 4 1 4a. When one of the four conditions tested in diamond block 414 is TRUE, 'buffer manager 216 invalidates buffer timer 218 at block 416 (eg, when powered), and Block 418 triggers the DMI link to leave the low power state L1 and enter the high power state L0. At block 420, the buffer manager 216 transfers the packet from the buffer 206-lt to the computing subsystem 230-1 using the DMA transfer mode. The memory unit 234, without buffering the packet, issues an interrupt to the computing subsystem 230-1. At block 422', the buffer manager 216 can selectively modify the communication rate of the transceivers 204-1-r to increase or decrease the rate of incoming packets based on their FIFO size and/or remaining energy. At diamond block 424, buffer manager 216 can determine if more packets have arrived within the awake timer. The buffer manager 2 1 6 maintains the time stamp of the last packet. When the time stamp minus the current time is less than the timer that keeps awake, the network is assumed to be busy. In this case, the buffer manager 216 will continue to operate at block 420. When the time stamp minus the current time is greater than the timer that is awake, the network is assumed to be idle. In this case, at block 426, the buffer manager 216 triggers the DMI link to exit the high power state L0 and enter the low power state L1, and continues operation at block 502 buffer manager 216. Referring again to one of the conditions evaluated by diamond block 414' buffer manager 216 may include a fourth condition (condition 4), sometimes referred to as a "failure - -31 - 1379192 security" trigger. The "fail-safe" trigger is designed to avoid a return condition 'i.e., a small number of packets (or a single packet) remain in the buffers 206-1 -t until the buffer timer 218 fails. This condition can be detected by monitoring the number of times a single packet is inside the buffer 206-1 -t when the buffer timer 218 fails. For example, if this condition occurs more than a certain number of times (eg, 3 times), the buffer manager 216 may temporarily disable the buffer 206-1 -t until the number of restart conditions can be set, thereby indicating the buffer 206- 1 -t should be enabled or reactivated. Examples of restarting conditions may include: (1) enabling buffers 2 06-1 -t according to a timer, such as after a defined time interval (eg, 1 sec); and (2) depending on a packet count, such as After receiving the defined number of packets (eg, 4000 packets), the buffers 206-1 -t are enabled. An example logic flow for evaluating the fourth condition is described in more detail with reference to FIG. FIG. 5 shows a logic flow 500 in accordance with one or more embodiments. The logic flow 500 can be performed by various systems and/or devices, and the logic flow 500 can be implemented as hardware, software, and/or any combination thereof, as desired for a given set of design parameters or performance limitations. For example, logic flow 500 can be implemented by a logic device (e.g., a processor) and/or logic (e.g., instructions, data, and/or code) executed by the logic device. For the sake of understanding and not limiting, the logic flow 500 will be explained with reference to Figs. The logic flow 500 can describe various operations of the node 1 1 〇 -1 - m in a general manner, and specifically describes the managed power system 120 and the power management module 1 130. The logic flow 500 can provide an example of a logic flow for evaluating the fourth condition (Case 4) -32-1379192 as indicated by the block 414 as described in Figure 4, with reference to Figure 4. To implement the fourth condition (Case 4), the buffer manager 216 can implement an event counter that counts the number of occurrences (X) of a particular event. The event may include the case where only a limited number of packets (e.g., one) are within the buffers 206-1 -t when the buffer unload is paused. When the event counter calculated event (X) exceeds the event threshold (N) (eg, N = 3), the buffer manager 216 can temporarily disable the buffer 206-1-t until the match is restarted. The situation is up. In one embodiment, for example, the buffer manager 216 can implement a packet counter to calculate the number (M) of packets in the buffer 506-1-t and set it with a buffer disable pause (B). Buffer timer 218. As shown in FIG. 5, in diamond block 502, when Μ is not equal to the packet threshold Μ (eg, Μ > 1 ), then at block 516, the buffer manager 2 16 resets the event counter (eg, Χ = 〇) and transfer control to block 416 of logic flow 400. At diamond block 502, when Μ is equal to the packet threshold 例如 (e.g., M = 1), at block 504, the buffer manager 216 increments the event counter by one (e.g., Χ = Χ +1). At diamond block 506, buffer manager 216 can determine if event counter (X) is greater than or equal to event threshold (e.g., Ν = 3). When the diamond block 506 is F A L S Ε, then at block 502, the buffer manager 216 is restarted. When the diamond block 506 is TRUE, then at block 508, the buffer manager 216 disables the buffer 206-1 -t. The buffer 216 then determines if a restart condition has occurred, such as when the buffer timer 218 is greater than the buffer disable pause (e.g., the timer is greater than L seconds) or the packet received due to the buffer 2 06-1 -t is disabled. The number is greater than the packet threshold (for example, P> 4000 seals - 33 - 1379192 packages). When both of the restart conditions are FALSE, the block returns to block 5 08. When any of the resume conditions are at block 510, the buffer manager 216 enables the buffer to send control to block 4 16 of logic flow 400.値Names N, L, and P are configurable settings. Various embodiments may provide several advantages for multiple uses. In one embodiment, for example, the buffer manager 206-1 -t can be used to allow calculation of the secondary system 23 0-1 state, thereby increasing energy conservation. In one embodiment, the power section communication is about 500 milliwatts (mW) to 2 watts (W). In some examples, various embodiments may be implemented as an article that may include computer readable media or storage media, compilations and/or materials for performing one or more embodiments of brain readable media or storage media. Examples that can be included but are not known. In various embodiments, for example, a disc, a disc, a flash memory or a firmware, which contains a computer for use in a general purpose processor or special application processor, is not intended to limit the embodiments. A hardware component, a software component, or a combination of both may be used. hard. Examples of physical components may include any of the examples previously proposed, and further include microprocessors, electrical (eg, transistors, resistors, capacitors, inductors, etc., logic gates, registers, semiconductor devices, wafers, when, then When the control is TRUE, then 206-1 -t, and E means that, Μ, : use and 216 and the buffer is held in a low power state, for example, can be achieved for the initiative to create articles. To store various logic operations. The power is limited to the previously mentioned items may include magnetic programming instructions, suitable for implementation. However, the implementation of various logic devices, circuit components, etc.), integrated micro-chip, wafer-34- 1379192 group and many more. Examples of software components can include software components, programs, application software, computer programs, applications, system programs, machine programs, operating system software, intermediate software, firmware, application programming interfaces (APIs), instruction sets, calculation codes, Computer code, code segment, computer code segment, word, 値, symbol or any combination thereof. Whether the embodiment is to be implemented using hardware components and/or software components can be varied according to any number of factors, such as required calculation rate, power level, thermal tolerance, processing cycle budget, input #data rate. 'Output data rate' memory resources, data bus speed and other design or performance limitations, as required for a given implementation. The words "couple" and "connect" may be used together with their derivatives. These terms are not necessarily synonymous with each other. For example, the terms "connected" and/or "coupled" may be used to mean that two or more elements are in direct physical or electrical contact with each other. However, the term "intimate" may also mean that two or more elements are not in direct contact with each other, but still operate or interact with each other. It must be emphasized that the abstract of the present invention allows the reader to quickly determine an abstract of the nature of the disclosure. It must be understood that the abstract is not intended to explain or limit the scope or meaning of the scope of the patent application. In addition, the various features are set forth in a single embodiment for the clarity of the disclosure. This method of revealing cannot be interpreted as a counter.  The claimed embodiment requires more features than what is described in each patent application. Rather, as the scope of the invention is claimed, the invention is intended to be less than all features of a single disclosed embodiment. Therefore, the scope of the patent application is hereby incorporated by reference in its entirety in its entirety in the extent of the disclosure of the disclosure of the disclosure. In the scope of the patent application, the terms "including -35-1379192 (including)" and "in which" are used as the respective terms "comprising" and "wherein". Equal terms. In addition, the terms "first", "second" and "third" etc. are used only for labeling purposes and are not intended to include numerical provisions in their objectives. Although it has been clearly defined in terms of structural characteristics and/or methodological behavior, it must be understood that the subject matter defined by the scope of the patent application is not limited to the specific characteristics or actions described above. Rather, the specific characteristics and behaviors set forth above are disclosed as examples of the scope of the application. Attached to the scope of patent application. BRIEF DESCRIPTION OF THE DRAWINGS Fig. 1 shows an embodiment of a communication system. Figure 2 shows an embodiment of the device. Figure 3 shows an embodiment of the first logic diagram. Φ Figure 4 shows an embodiment of a second logic diagram. Figure 5 shows an embodiment of a third logic diagram. [Main component symbol description].  100: communication system 110-1: node 1 10-2: node 1 1 0 - m: node 120: managed power system - 36 - 1379192 120-1: managed power system 1 3 0 : power management module 130- 1 : Power Management Module 1 4 0 - 1 : Communication Link 140-2 : Communication Link 150-1-s: Power Management Packet Data Unit 204 - Bu r: Transmitter 206-1-t: Buffer 2 0 8 : controller

2 1 〇: Communication subsystem 2 1 2 '·Network status module 214-1: Power management interface 214-2: Power management interface 214-3: Power management interface 216: Buffer manager 2 1 7: Watermark generation 218: Buffer Timer 2 2 0: Communication Bus 2 3 0: Calculation Sub System 2 3 2: Power Supply 234: Power Management Controller 236: Power Control Timer 240-1 -q: Power Management Message 2 5 0 - 1 -v : Communication connection -37- 1379192 260 : Power status information

-38-

Claims (1)

1379192 囔r Annex 3A: No. 97135495 Patent Application Amendment Amendment of August 31, 101 of the Republic of China. Patent Application Area 1. A power management device comprising: a power management module having a power management controller, the power The management controller is coupled to the transceiver: The managed power system is coupled to the power management module, the managed power system includes a communication subsystem and a calculation subsystem, and the power management controller switches the communication subsystem and the calculation subsystem from a high power state to a low a power state to store energy, the communication subsystem having: the transceiver; a buffer coupled to the transceiver, the buffer storing the information packet of the transceiver during communication idle period, for generating calculation idle a watermark generator coupled to the buffer, the watermark generator operable to generate a variable receive threshold, and a buffer manager coupled to the buffer and the watermark generator, the buffer The manager is operative to forward the stored information packet from the buffer to the computing subsystem based on a variable receiving threshold to receive an energy power management message having energy measurement parameters from the power management controller and transmit A requirement to adjust the communication rate of the transceiver based on the energy measurement parameter. 2 as claimed in claim 1, wherein the buffer manager is operable to buffer the stored information from the buffer when the number of information packets stored in the buffer exceeds the variable reception threshold Transfer to the 1379192 households less than I曰 correction / 3⁄4 . , calculate the secondary system. 3_ The apparatus of claim 3, wherein the watermark generator is operative to generate the variable reception threshold based on the received data rate parameter, the buffer size parameter, and the communication restart waiting time parameter. 4. If the device of claim 1 is used, the buffer manager is operable to disable the buffer when the communication idle parameter is less than the communication idle period threshold to prevent the buffer from storing the information. Packet. 5. The device of claim 1, wherein the buffer timer is 'affined to the buffer manager to set the buffer timer with a buffer unloading pause '. The buffer manager is operable to The stored information packet is forwarded from the buffer to the computing subsystem when the number of packets stored by the buffer exceeds the variable reception threshold before the buffer unloads. 6. The device of claim 1, wherein the buffer manager is operative to receive a buffer unload event signal before the number of packets stored in the buffer exceeds the variable receive threshold, and The stored information packet is forwarded from the buffer to the computing subsystem to respond to the buffer unload event signal. 7. The device of claim 1, comprising a controller for coupling the transceiver, the buffer manager operable to transmit a demand for adjusting the transmission based on a buffer size parameter Communication rate of the device. 8. A power management method comprising: modifying a communication subsystem and calculating a power state of the subsystem, from a high power state to a low power state; -2- 1379192 A And storing the information packet in the buffer of the communication subsystem during the communication idle period for generating a calculation idle period; generating a variable reception threshold for the buffer; storing the storage according to the variable reception threshold Transmitting the information packet from the buffer to the computing subsystem; receiving a power management message having energy measurement parameters; and transmitting a demand to adjust the communication speed based on the energy measurement parameter 9 - The method of claim 8, wherein the stored information packet is forwarded from the buffer to the computing subsystem when the number of information packets stored in the buffer exceeds the variable reception threshold . The method of claim 8, wherein the variable reception threshold is generated based on a received data rate parameter, a buffer size parameter, and a communication restart waiting time parameter. 1 1 · The method of claim 8 includes the method of disabling the buffer when the communication idle period φ is less than the communication idle period threshold to prevent the buffer from storing the information packet. 12. The method of claim 8, comprising storing the stored information when the number of packets stored in the buffer exceeds the variable reception threshold before the buffer is unloaded and expired Transfer from the buffer to the computing subsystem. 1 3 - The method of claim 8, comprising receiving the buffered information packet when the number of packets stored in the buffer exceeds the variable reception threshold before receiving the buffer unloading event signal Buffer to -3- 1379192 The following year ^ I 曰 Correction gray I is sent to the calculation subsystem. 1 4. The method of claim 8, comprising a set of computer readable program elements operable to construct a system for performing the method of claim 8 of the patent application.