HK1166571A - System and method for managing energy efficiency and control mechanism in network - Google Patents

Description

System and method for managing energy efficiency and control mechanisms in a network

Technical Field

The present invention relates generally to managing energy consumption in a network.

Background

In recent years, energy costs continue to increase with a trend of escalation. In view of this, industries are becoming increasingly sensitive to these rising costs. IT infrastructure is an area that attracts increasing attention. Many companies today take their IT system energy usage into account to decide whether the energy costs can be reduced. Accordingly, industries directed to energy efficient networks have emerged in an effort to address the increasing cost of overall IT equipment usage (e.g., PCs, displays, printers, servers, network components, etc.).

Modern network components are increasingly implementing Energy Consumption and Efficiency (ECE) control mechanisms. Traditional ECE mechanisms such as power scheduling are also beginning to work with networks. Some modern ECE control mechanisms allow physical layer components to enter and exit low power consumption states. The ECE control strategy controls when and under what circumstances physical layer components with ECE control functionality enter and exit low power consumption states. The device control strategy plays a crucial role in producing the least performance impact on the network with the greatest power savings.

Although ECE mechanisms and control strategies are being widely used, their use is not coordinated in traditional use. Multiple uncoordinated, asynchronous power saving mechanisms may cause power saving failures at the individual component and network level.

Disclosure of Invention

According to an aspect of the invention, there is provided a system for managing energy efficiency and control mechanisms in a communication network, the communication network comprising a plurality of network components, the system comprising:

a network energy manager (NPM) coupled to at least one of the plurality of network components, wherein the NPM is configured to:

receiving energy information from at least one of the plurality of network components;

analyzing the energy information;

generating configuration instructions based on the analysis of the energy information; and

sending a configuration instruction to at least one of the network components.

Preferably, the NPM is further configured to:

receiving configuration information from at least one of the network components; and

sending the configuration information to at least one of the network components.

Preferably, the energy information comprises an operational characteristic of one of the plurality of network components.

Preferably, the operating characteristics are:

a supported link speed available to the network component, or an operational mode available to the network component.

Preferably, the configuration instructions include at least one of routing information and switching information for traffic on the network.

Preferably, the configuration instructions comprise a control strategy for controlling the energy efficiency and control mechanism.

Preferably, the Energy efficiency and control mechanism comprises an Energy efficiency ethernet (Energy efficiency ethernet) control strategy.

Preferably, the NPM is further configured to coordinate configuration instructions sent to at least two of the plurality of network components.

Preferably, the energy information received by the NPM comprises a link utilization level of at least one of the plurality of network components.

Preferably, the energy information received by the NPM comprises characteristics of a link between two of the plurality of network components.

Preferably, the characteristics of the link between two of the plurality of network components comprise at least one of:

the size of the burst on the link is,

the interval time of the burst on the link, an

Idle time on the link.

Preferably, the energy information received by the NPM comprises a control policy applied to one of the plurality of network components.

Preferably, the control policy applied to one of the plurality of network components comprises a link utilization threshold.

Preferably, at least one of the plurality of network components is a network switch.

Preferably, at least one of the plurality of network components is a port of a host.

Preferably, at least one of the plurality of network components is an optical network component.

Preferably, the NPM coordinates configuration instructions for optical and non-optical network components.

Preferably, the NPM receives energy information from a network component, wherein the energy information is encoded using at least one of a Link Layer Discovery Protocol (Link Layer Discovery Protocol) and a simple network management Protocol (simple management Protocol).

Preferably, the NPM sends configuration instructions to the network component, the configuration instructions being encoded using a simple network management protocol.

Preferably, the NPM coordinates low power consumption states in two different network components.

Preferably, the first component is an optical network component and the second component is a non-optical network component.

According to another aspect of the present invention, there is provided a method for managing energy efficiency and control mechanisms in a network comprising a network energy manager (NPM) and a plurality of network components, the method comprising:

receiving energy information from at least one of the plurality of network components;

analyzing the energy information;

generating configuration instructions based on the analysis of the energy information; and

sending the configuration instructions to at least one of the network components.

Preferably, the energy information comprises an operational characteristic of one of the plurality of network components.

Preferably, the configuration instruction comprises at least one of a routing instruction and a switching instruction.

Preferably, the routing instructions comprise specific traffic paths taken for at least two of the plurality of network components.

Preferably, the configuration instructions comprise a control strategy for controlling an energy efficiency and control mechanism of at least one of the plurality of network components.

Preferably, the energy efficiency and control mechanism comprises an Energy Efficient Ethernet (EEE) control strategy.

Preferably, configuration instructions are generated for at least two network components, the configuration instructions comprising configuration instructions for coordinating sleep cycles of the at least two network components.

Preferably, configuration instructions are generated for at least two network components, the configuration instructions comprising configuration instructions for coordinating wake-up periods of the at least two network components.

Preferably, the first and second electrodes are formed of a metal,

the energy information comprises information describing a network traffic event, the occurrence of the traffic event being related to a first network component, the traffic event being predictive of future traffic of a second network component; and

the configuration instructions include instructions for adjusting a hibernation setting of the second component.

Preferably, the first and second electrodes are formed of a metal,

the energy information includes information corresponding to link speeds of at least two links between network components;

the configuration instructions include instructions for coordinating the link speed so as to reduce buffering requirements associated with network components; and

the sending includes sending the configuration instructions to a network component associated with the at least two links.

Preferably:

the energy information includes information describing buffering times of traffic in three network components, the three network components including:

the originating network component or components of the network,

a first network component connected to the originating network component, and

a second network component connected to the first network component; and

the configuration instructions include instructions to:

increasing the buffering time of the originating network component based on the buffering times of the first and second network components, and

reducing a buffering time of the first and second network components.

Drawings

The drawings illustrate the present invention and, together with the detailed description, further serve to explain the principles of the invention and to enable a person skilled in the pertinent art to make and use the invention.

FIG. 1 is a block diagram of an exemplary network topology of the present invention;

FIG. 2 is a block diagram of an example network topology having a network energy manager in accordance with an embodiment of the present invention;

FIG. 3 is a block diagram of an example network topology showing component configuration and owning a network energy manager in accordance with an embodiment of the present invention;

FIG. 4 is a block diagram of an example network topology illustrating different physical and logical locations of network energy managers in accordance with an embodiment of the invention;

FIG. 5 is a block diagram of an example network topology showing component buffers and owning network energy managers in accordance with embodiments of the invention;

FIG. 6A depicts an example time plot featuring energy consumption and efficiency of network components;

FIG. 6B depicts an example time graph featuring energy consumption and efficiency of network components in accordance with an embodiment of the present invention;

figures 7A-B depict a network having an Optical Network Unit (ONU) in accordance with an embodiment of the present invention;

fig. 8-11 provide flow diagrams of example methods for managing energy efficiency and control mechanisms in a communication network having a network energy manager (NPM) and a plurality of network components.

The present invention is described with reference to the accompanying drawings. The drawing in which an element first appears is generally indicated by the leftmost digit(s) in the corresponding reference number.

Detailed Description

The following detailed description of the invention refers to the accompanying drawings that illustrate exemplary embodiments consistent with this invention. Other embodiments are possible, and modifications may be made to the invention within the spirit and scope of the invention. The detailed description is, therefore, not to be taken in a limiting sense. Rather, the scope of the invention is defined by the appended claims.

The features and advantages of the invention will be set forth in the detailed description which follows, and in part will be obvious from the description, or may be learned by practice of the invention. The advantages of the invention will be realized and attained by the structure and particularly pointed out in the written description and claims hereof as well as the appended drawings. The following detailed description is exemplary and explanatory and is intended to provide further explanation of the invention.

The embodiments and "one embodiment", "an example embodiment", etc., described herein indicate that the embodiment described may include a particular feature, structure, or characteristic. However, not every embodiment may need to include the particular features, structures, or characteristics. Moreover, such terms are not necessarily referring to the same embodiment. When a particular feature, structure, or characteristic is described in connection with an embodiment, it is submitted that it is within the knowledge of one skilled in the art to effect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described.

SUMMARY

In general, some embodiments of the invention provide an improved method for universal coordinated network management. In contrast to the above-mentioned incoherent implementation of policies and configurations, the present invention uses a network energy manager to efficiently coordinate policies, operational parameters, and other configuration information at each layer of the network.

Modern networking teams have a variety of useful mechanisms designed to facilitate a variety of beneficial results. Modern networks also employ a variety of connectivity components, such as fiber optic components and wired components. Network components that do not employ optical network technology (e.g., passive optical network components), as typically employed in the present invention, may be referred to as "non-optical" components. Embodiments of the present invention describe different methods of coordinating these different parts of the network using a common network management approach.

Energy Consumption and Efficiency (ECE)

As used herein, an energy consumption and energy saving (ECE) control mechanism refers to various techniques for controlling the energy consumption and efficiency of a device. Generally, these ECE mechanisms are designed to reduce energy consumption and improve energy efficiency while maintaining an acceptable performance level.

One example of a modern ECE control mechanism is the IEEE Std 802.3az (tm) -2010 standard, also known as the power save ethernet standard, which is incorporated herein by reference. EEE is an IEEE standard designed to store power in Ethernet over a selected set of physical layer devices (PHYs). Example PHYs involved in the EEE standard include 100BASE-TX and 1000BASE-T PHYs and the emerging 10GBASE-T technology and backplane (backplane) interface such as 10 GBASE-KR.

Traditionally, in networks with multiple different components linked, different mechanisms for improving performance, energy savings, and efficiency may be implemented at various steps. Three different types of mechanisms are discussed in connection with embodiments of the present invention: optical power savings, EEE mechanisms, and traditional methods such as device "power sharing". From the description, it will be appreciated by those skilled in the relevant art that embodiments of the invention are applicable to other types of energy savings.

For convenience, the term "EEE energy saving" is used herein to describe the method of saving electrical connection (electrical connection) network components (e.g., copper). The methods discussed herein may also be used for "non-electrical" components, such as fiber optic connections and components. It should be mentioned here that the method described in the invention can also be used for new network standards, objects and aspects of the implementation method. Network methods developed by specific product vendors may also benefit from such methods, such as sub-rate (mixing).

By adding an additional control layer, EEE functional devices may have their ECE characteristics managed by a class of configuration instructions called control policies. As discussed herein, the network energy manager may generate control policies by considering different types of energy information such as traffic patterns, traffic, performance characteristics, traffic categories and profiles over time, and related information to help decide when to take advantage of EEE features. The control policy generation may also be decided by considering the activity of the hardware subsystem as a proxy for the actual traffic analysis. Broadly, the energy information collected by an embodiment may include network configuration, resource and energy usage information for all network hardware, as well as software and traffic related or possibly related to ECE optimization.

For example, the control policy of the switch may describe when and under what conditions the switch enters and exits the power saving low power consumption state. In a system, a control strategy may be used to control one or more physical or virtual devices. Control strategy (also called physical control strategy or device control strategy), e.g. adding an extra control layer to an EEE functional device

It should be noted that the principles of the present invention may be used in a wide variety of contexts, such as in all PHYs that implement ECE (e.g., backplane, twisted pair, fiber optics, etc.). Furthermore, the principles of the present invention may be applied to standard or non-standard (e.g., 2.5G, 5G, 100M, 1G, and 10G fiber interfaces, passive fiber networks, etc.) link rates as well as future link rates (e.g., 40G, 100G, 400G, megabits, etc.). Later extensions of these standards, such as IEEE 802.3 and IEEE P1904.1, may also benefit from the methods discussed herein. It should also be noted that the principles of the present invention may be used asymmetrically or symmetrically for a given link. The teachings of the present invention are not limited to a particular media type. In addition to the media types mentioned herein, other existing and non-existing may utilize the methods of the present invention, such as structured cabling, fiber optic cables, and the like.

In an embodiment, one common approach to energy consumption and energy efficiency addressed by some embodiments is to reduce the energy consumed by as many network components/links as possible for as long a time as possible. As mentioned above, this goal may result in unacceptable performance loss in the network if not managed efficiently. For example, each device that is powered down (either into a sleep mode or in a low power state) must be restored within a reasonable amount of time to perform the required functions.

In fig. 1, the topology 100 depicts a user-network interface (UNI)104 coupled to an access network 105, the access network 105 being coupled to a core network 107. The UNI104 includes gateways 115A-B and user devices 110A-D. The access network 105 includes a Passive Optical Network (PON)109, and the PON109 has a cable termination equipment (OLT)130 coupled to Optical Nodes (ONUs) 120A-B. The aggregation switch 140 links the core network 107 with the access network 105, the core network 107 being linked with the internet 101. As noted, topology 100 has both fiber optic network components and wired network components.

With respect to topology 100, those skilled in the relevant art will appreciate the non-limiting A1-A10 items listed below, described below:

A1. the access network 105, as described in some embodiments herein, is a part of a communication network that utilizes user devices 110A-D to connect users with their service providers that run the OLT130, ONUs120A-B, and core 150. Those skilled in the relevant art will appreciate that: the methods detailed by embodiments of the present invention may be applied to a variety of different network topologies and configurations. Although topology 100 is described as having fiber optic components, the teachings of the present invention are not limited to a certain type of network. Other types of networks, both existing and yet to be invented, may employ the method of the present invention, such as digital subscriber line and DOCSIS cable modems.

In a broad sense, the method of the invention can be used for any type of PON. Such as Gigabit PON (GPON), Ethernet PON (EPON) and wavelength division multiplexing PON (WDM PON). Methods using the PON protocol in non-fiber networks may also benefit from the methods described herein. Such as an ethernet passive optical network (EPOC) over coaxial cable. Those skilled in the relevant art can apply the energy information and configuration instructions described herein to different types of networks by understanding the concepts of the present invention, and not just the specific description of the invention.

A2. The PON109, which has the fiber network components OLT130 and ONUs120A-B, is an exemplary network type of one embodiment, in conjunction with other components and examples described herein. One skilled in the relevant art will appreciate that non-fiber optic networks may also benefit from some of the concepts of the present invention.

A3.onus120a-B is typically installed in a user's home and provides an interface in the user's home and between wired/WiFi ethernet transmission of data from the user's home and optical data transmission of the PON 109. The direct coupling of the ONUs120A-B to the gateways 115A-B is illustrated in fig. 1-2. In a typical implementation, the ONU is linked to the gateway using a media converter and then using the CPE (customer premises equipment). In various implementations, all of these functions may be implemented in one box. Fig. 1 and 2 simplify the schematic by omitting these components. To provide further details, the discussion of fig. 7B below describes an embodiment of a CPE component coupled with a user device.

A4. It should be appreciated that although aggregation switch 140 is discussed and described as a single aggregation switch 140, aggregation switch 140 may also be a collection of switches designed to optimize the links between downstream components (e.g., OLT130) and upstream components (e.g., core 150).

A5. The user devices 110A-D herein refer to end user devices that are coupled to the access network 105 as endpoints. Examples include personal computers and network-enabled devices.

A6. The gateways 115A-B provide an interface for the end-user devices 110A-D. Examples include cable modems, set-top boxes, and coaxial cable (MOCA) interfaces.

A7. Core 150, also referred to as a network core, is a term primarily related to telecommunications networks. This non-limiting term generally refers to the network infrastructure used to link the service provider and the internet 101. The core 150 may become a website primarily used for switching, routing, and data processing functions of the network.

A8. The network components described with respect to fig. 1, as included in other figures of the present invention, are intended to provide an exemplary, non-limiting illustration of one linked network component and are not intended to depict the required topology.

A9. The links between the user devices 110A-D, the gateways 115A-B, and the ONUs120A-B are typically electronic (e.g., copper wire based, WiFi). These links are generally capable of implementing EEE energy saving methods. These components and links from the ONUs120A-B may form a user-network interface (UNI 104).

The link between onu120a and OLT130 is an optical fiber-based link and may employ optical power-saving methods. These optical network components may be implemented in a passive optical network (PON 109) architecture.

Network energy management

Fig. 2 adds an integrated network energy manager 210 to the topology 100 of fig. 1. Embodiments of the NPM210 may provide a deeper view of how the links of the topology 100 and the network components are related and may also allow for management of the included components.

As described above, conventional approaches to ECE in a network do not provide for end-to-end management of network components. The lack of ECE management is particularly important for enabling ECE improvements. For example, in the topology 100, there is no central control over different ECE performance, control policies, and other energy consumption characteristics of different network components. Embodiments of the NPM210 will be used to address many of these types of problems by collecting energy information, analyzing the energy information, and generating configuration instructions, as described in the present invention. Stated another way, in an embodiment, it is a feature of the NPM210 to collect the physical characteristics and ECE logic of the associated network components and then perform changes in order to improve ECE performance.

Fig. 3 depicts a network topology 300 having a core 350, an aggregation switch 340, an OLT330, ONUs 320A, and a user device 310A, each of which has a respective configuration 380A-E. Topology 300 is also depicted as having a network energy manager (NPM) 210.

Generally speaking, the NPM210 is communicatively coupled to one or more network components and receives and collects energy information from the network components. This energy information will be further described below. After collecting the energy information, the NPM210 analyzes the energy information and generates configuration instructions based on the analysis. These configuration instructions, which will be further explained below, are then forwarded to the respective network components.

The analysis and generation features described above may balance energy information with other network considerations such as performance, security, and the like. In an embodiment, the improvement of ECE performance by the NPM210 may be balanced, coordinated, or influenced by other performance characteristics and network set goals.

Any features or similar components available to the NPM210 may be analyzed and used to generate configuration instructions. In an embodiment, the NPM210 may become a uniform resource to advance ECEs associated with the topology 300 and coordinate the functionality of different network components with different ECE targets throughout the network.

The configuration instructions include all potential parameters, settings, configurations, and other similar characteristics for the network components. As mentioned above, there is no end-to-end management unified configuration 380A-E in legacy networks.

Upon generating the configuration instructions, the NPM210 may receive various types of energy/power related information (energy information) regarding the network components. Examples of the energy information include physical layer (PHY) information, link information, ECE control policy information, and usage information. After reading the teachings of the present invention, one skilled in the relevant art will appreciate that a wide variety of information, characteristics, strategies, and the like may be quantified as (qualify as) the energy information used in the present invention.

Physical layer (PHY) information may relate to the operational characteristics and capabilities of the network components themselves, including characteristics such as supported link rates available to the network components, different operational modes (e.g., subset modes) available to the components, and so forth.

The link information may relate to the use of links between network components. An example of linking information is traffic fullness (fullness). In another example, the link information may include burstiness parameters (e.g., size of bursts on the link, time interval between bursts, idle time on the link, etc.) that enable determination of actual link utilization. Another example is the percentage of link capacity usage over time, e.g. if the average usage of a 10G link is always below 1G over a period of time, it may become a test value for the amount of link usage that is commonly used.

The ECE policy parameters may relate to those parameters that are capable of managing the analysis and/or operation of control policies for network components. Upon configuring the network components, policy parameters, including link usage thresholds, IT policies, user parameters, etc., may be set to manage ECE operation of the device, for example. Finally, the application information may relate to system application features that manage the operation of the network components. Examples of useful application information include flow information through network components that have been analyzed, such as: in an L2 switch without virtualization, knowing the AVB flow information through the component can help decide whether a low power state is useful.

It is noted that the particular set of energy information received, the analysis performed on the energy information, and the process of generating configuration instructions based on the energy information are performed in association. In addition to the analysis mechanisms that collect data and use, it is important that the NPM210 integrates, analyzes, and utilizes energy information in network components to direct the configuration of specific components, and generally covers the configuration and routing/switching of the entire network.

The energy information of the component may be determined in a number of ways. In the methods used in the embodiments, a representative sample of network components is monitored and the collected metrics may be extrapolated to other components in the network. An example of this method of inferring energy information may be found in U.S. patent application Ser. No.12/947,537 (case #2875.483000), entitled "Measuring and managing Power use and coating in a Network," filed on 16.11.2010, which is incorporated herein by reference in its entirety.

In another example, NPM210 may collect energy information from ONUs 320A-B, OLT330 and aggregation switch 340. Such information may include non-limiting example types T1-T3:

t1. run characteristics such as wake-up time, link speed, buffer size, manufacturer, generation of device, location of device in network and configuration options.

T2. policy information that has been executed such as sleep trigger (sleep trigger) and buffering requirements.

T3. control strategy settings such as how aggressive energy saving strategies are set, timers, etc.

Given this description, one skilled in the relevant art will appreciate that the added physical and logical characteristics of network components may provide useful information for generating configuration instructions.

It should also be noted that the term "energy" in network energy manager (NPM)210 is not intended to limit the management functions of the embodiments. Although embodiments of the present invention have discussed energy consumption and energy efficiency (ECE) mechanisms and strategies, other types of strategies, mechanisms, targets, methods, etc., may also be implemented using the general overview of the concepts of embodiments of the present invention.

Network energy manager placement

FIG. 4 illustrates a system 400 of an alternative physical and logical configuration of various embodiments of the NPM210 of FIG. 2. Each described arrangement of NPMs 410A-D is non-limiting and presents an arrangement that can work independently or in concert with other NPM410A-D components. For example, the system 400 may have a single NPM410A, two NPMs 410A-B, all four NPMs 410A-D components, or configurations where network components are not shown.

In an embodiment, unlike the external arrangement shown in FIG. 2, the NPM410A depicted in FIG. 4 is arranged as a component of the core 350. As described above, the NPMs 410A-D may be implemented in various network devices such as aggregation switch 340, OLT330, ONUs 320A (not shown), user devices 310A-D, and other components of topologies 300 and 400. Embodiments of the NPMs 410A-D may be implemented as one of a software component and a hardware component.

The NPM 410B depicted in fig. 4 is disposed as a component of the aggregation switch 340. The integration in the switch/router may be implemented as a software component or "plug-in" or as a hardware device. One skilled in the relevant art will appreciate that other software and hardware means are possible.

The NPM 410C depicted in fig. 4 is disposed as a component of OLT 330. In other embodiments, the NPM410 is described independently of the exposed legacy network components.

In an embodiment, the NPM 210/410 need not be directly coupled with the network components in order to gather energy information and send configuration instructions to the components. Those skilled in the relevant art will appreciate that different network protocols may be used to perform these collection and command functions. In examples discussed further below, Link Layer Discovery Protocol (LLDP) may be used to collect configuration/policy information and features from network components, and Simple Network Management Protocol (SNMP) may be used to collect information and issue configuration instructions.

The placement illustrated in fig. 4 is not limiting. One skilled in the relevant art will appreciate that the functionality of the NPMs 410A-D as described may be located in a variety of places in the system as described herein, may be implemented by either software or hardware, or a combination of both. Very noteworthy is: the logic and functionality of the NPMs 210, 410A-D need not be centralized in a single component, rather the logic and components of the described embodiments of the invention may be distributed throughout network components.

Examples of the collection of energy information, the analysis of energy information, and the generation and distribution of configuration information are further described below.

NPM collection and control mechanism

One skilled in the relevant art has appreciated that the NPM210 may collect energy information from network components in a variety of ways. Embodiments may collect energy information in real time, or at specific points such as deployment of network components or changes in configuration of network components.

Embodiments may use conventional data collection protocols such as Link Layer Discovery Protocol (LLDP). In contrast to the traditional use of these common protocols, some embodiments use LLDP to gather information from the entire network. One way to implement this extended functionality is to pass and aggregate energy information between components using LLDP until the information reaches the NPM 210.

Conventional protocols may also be used to collect energy information and distribute configuration instructions. The Simple Network Management Protocol (SNMP) allows energy information and configuration instructions to be sent to and received from the NPM210 and network components over the network.

In an embodiment, SNMP provides end-to-end network element management using a configuration file and Management Information Block (MIB). For each network component, the NPM210 creates and maintains a configuration file that can be embedded in the MIB and transported by the SNMP. The NPM210 may create and reference configuration files when implementing individual and general configuration instructions.

In an example, a service provider may install ONU320A at a client site, which requires an initial ECE configuration. Managing the configuration file by the NPM210 using SNMP may allow the service provider to know the characteristics of the device and certain parameters of the scheme for the device. An exemplary parameter is that the EEE policy in ONU320A buffers each received packet for 1 millisecond before sending the packet to the network access port. Because the NPM210 has profiles of other components in the system, exemplary policies may be integrated with these other components. For example, OLT330 may have a policy that accounts for 1 millisecond 320A requirements.

Examples of the present invention

Some embodiments described herein allow for the generation of configuration instructions to coordinate the operation parameters, policy-based parameters, and the maintenance and management of network components' related parameters by collecting energy information from the network components.

The coordinated configuration may be based on a central power saving strategy, wherein configuration parameters such as time interval of sleep periods, time interval of active periods, service related configuration are centrally controlled by NMP210 for multiple power saving mechanisms. This coordination may improve the likelihood that configuration parameters on different devices work together.

Embodiments may collect, analyze, and coordinate configuration information for trigger events (triggers) across different network devices. The analysis considers the rules and criteria of each power saving mechanism on each device to enter and wake from sleep periods. Embodiments may also use the sleep/active state of a power saving mechanism as a trigger for other related power saving mechanisms.

Because of the above-mentioned conventional approaches, many different conventional types of energy consumption, energy efficiency, and performance characteristics may be improved by the coordination provided by the embodiments. The following non-limiting list of P1-P6 will be used to summarize some of these features. One skilled in the relevant art will recognize, upon reading the teachings of the present invention, that additional sub-optimal features may also be addressed by the embodiments. The problem features may include:

p1. delay-a time delay that occurs in the system during the transmission of information. Delays have a tremendous impact on subscriber Service Level Agreements (SLAs), especially on time critical services such as voice and video.

P2. jitter (delay variation) — variation in delay amount over time. For example: there is no longer a constant delay, which varies in systems with jitter. As known to those skilled in the relevant arts, this variation can cause significant problems in certain sensitive applications, such as voice over internet protocol (VoIP).

P3. redundant component resource requirements-different types of network components have different resource characteristics such as their buffer sizes. Different components may have different cost characteristics associated with changing their resource capacity. In an example, adding resources to the aggregation switch 140 for buffer size is more expensive than adding resources to the user device 110A for buffer size. A potentially undesirable characteristic of a network is misallocation between the performance requirements of the components and the higher resource costs. For example, with this policy, the aggregation switch 140 is given the burdensome task of buffering requirements compared to the user equipment 110A.

P4. different optimization-in the topology 100, for example, each exposed network component may or may not own any ECE mechanism. If mechanisms exist, they may be incompatible and uncoordinated. Each link in the network may have its own ECE policy.

P5. uncoordinated ECE mechanism-in an ECE mechanism uncoordinated network, each network component can only see real-time traffic-waking only when, for example, new traffic arrives, rather than when traffic is transmitted from an upstream device. Because traffic waits for a network component to wake from ECE dormancy, traffic must be buffered and delay added at various delayed steps. In the worst case, different types of network components may add different unpredictable delays. This unpredictability of buffering and delay may lead to jitter as described in P2 above.

The NPM210 controls all sleep features of the links between components. By synchronizing the sleep cycles, embodiments can help ensure that when a primary power saving mechanism, such as optical sleep, enters sleep mode, all other power saving mechanisms will be advantageously adjusted. In the example of coordinated sleep mode, other network components are configured to also enter sleep mode immediately; and when the primary power saving mechanism wakes up, the other mechanisms will wake up immediately as well.

P6. speed mismatch in the network caused by overuse. By centrally controlling the configuration instructions of the network components, embodiments may handle speed mismatches before they can produce negative effects such as jitter.

In an embodiment, the NPM210 may control network components, and the NPM210 may cause a wake event when traffic approaches from a node on the network that is only a few steps away. Further, in another embodiment, the NPM210 possesses data corresponding to link speed. In an embodiment, if traffic is licensed (warrange), the NPM210 may selectively adapt the link subrate, slow down the link and save power.

The described embodiments of the present invention both improve existing ECE mechanisms for network components and enable the implementation of new ECE mechanisms.

Fig. 5 depicts an example of the benefits of the network energy manager 210 management. As depicted in fig. 5, fig. 5 shows the core 350, aggregation switches 340, OLTs 330, ONUs 320A-B, and user devices 310A-D associated with 510A-I, respectively.

As will be appreciated by those skilled in the relevant art, when multiple power saving mechanisms are uncoordinated and asynchronous as described above, a packet may be buffered two or more times due to uncoordinated sleep periods under the different power saving mechanisms. As a result, the delay will increase the combined amount of sleep cycles.

In an embodiment, with the above described central control management, the NPM210 enables pooling of buffer time (pool) into fewer network components and enables generation of a unified buffer system for the network. Fewer network components performing buffering functions may result in less delay/jitter because most of the real-time processing described above is not required. This pooling (Pooling) can have beneficial consequences for performance, ECE and network implementation costs. In an embodiment, when the power saving mechanisms are coordinated and synchronized, the delay is only from the longest sleep period, rather than the superposition of multiple sleep periods as in the conventional approach.

In an embodiment, other advantages may be realized by pooling buffering as close as possible to the user devices 310A-D. As discussed below, pooling buffering time at the user device 310A-D (e.g., user) level may have an impact on user satisfaction, performance, and cost savings. As is known to those skilled in the relevant arts, the present invention gives its description that it is much easier to stop traffic at the source than to buffer it with a more distant path.

In an embodiment, user devices 310A-D buffer pooling is accomplished by using the following steps D1-D5:

D1. when the user device 310A generates network traffic, the network energy manager 210 receives energy information from all connected components of the upstream path, such as the core 350, aggregation switch 340, OLT330, and ONUs 320A. These updates received from the components (transport information) include information such as: the current state of the component (e.g., sleep and active) and the estimated buffer time required for active transmission by the device.

In the example, aggregation switch 340 is currently in a sleep state and has an ECE policy, whereby all traffic is buffered in buffer 510B for 500 milliseconds before being transmitted to core 350, and the sleep state takes 200 milliseconds to wake up. Thus, the example of aggregation switch 340 traditionally requires 700 milliseconds to wake up and transmit from its sleep state. As known to those skilled in the relevant arts, a variety of different states, strategies, delays may affect the estimated buffering time of a component by understanding the teachings of the present invention.

D2. This information is collected by the network energy manager 210 along with other similar information from other network components. In one embodiment the component receives a determined wake-up time and in another embodiment the network energy manager 210 receives status, control ECE policy and configuration information and calculates the wake-up time to wake up from. Embodiments may use a link layer discovery protocol to collect information from different network components.

D3. The network energy manager 210 aggregates updates received from connected/operational network components. If the network components on the traffic path from the user device 310A to the core 350 are unable to generate the required information, in an embodiment, the network energy manager is able to evaluate the transmission information based on the type of component and other characteristics.

D4. In the example, once the transmission information is aggregated, all of the estimated upstream buffer times are communicated to the user equipment 310A. According to an embodiment, this buffering time may be implemented as a buffering policy on the user equipment 310A by using the buffer 510F. By using ECE coordination, the buffering time can be combined (in parallel) to 500 milliseconds instead of buffering in series (500+200) to 700 milliseconds.

D5. Continuing with the example, once the buffering time is enforced on the user device 310A, the NPM210 may manage the network components to change their enforcement policies. Changes are made so that the policy of the network component does not conflict with the buffering policy implemented on the user device 310A. At least 3 benefits can be derived from the ECE coordination described above: 1) less overall delay (500 is needed instead of 500+200), 2) higher and more continuous overall sleep in the system (more opportunities are available for power saving), and 3) overall buffering requirements are reduced because there is no need to repeat buffering.

Some embodiments may also yield significant network resource cost savings by moving the buffered downstream items to the user devices 310A-D. As is known to those skilled in the relevant arts, plus knowledge of the present teachings, the closer a component is to the core 350, the more expensive the implementation of network component buffering on the access network becomes. For example, implementing a cache memory in the aggregation switch 340 to adequately handle the caching of all downstream components (OLT330, ONU320A, and user devices 310A-D) is much more expensive than implementing a cache at each user device 310A-D.

In an embodiment, by implementing the downstream cache pooling described above, the upstream buffers 510A-E may be eliminated or reduced from their respective components.

Also in most access networks, the cost of implementing cache memory (e.g., system RAM, hard disk space) is borne by the user, not the service provider. When storage is implemented by a single consumer, each consumer can select the size of storage space they want to use — increasing cost and performance. In another embodiment, the service provider may have different pricing tiers for the consumer based on storage policies and other energy and efficiency savings considerations.

Finally, the storage implemented at the user devices 310A-D is essentially unlimited — allowing the use of local hard drives as cache overflows.

Among the cost-saving benefits of another embodiment, better coordination of the ECE mechanisms with different types of network components may allow network designers to select cheaper components for tasks that require high performance. Further, as known to those skilled in the relevant arts, certain components may be cheaper with the same performance for a given wake-up time required to wake up from a sleep or low power consumption state.

For example, fiber optic components may be more expensive than electrical (wired) link components in a given wake performance. As discussed in some embodiments of the invention, having centralized control over network components may allow a network designer to choose, for example, to place a wired component in a critical path and require it to have a quick wake-up. In this manner, some embodiments facilitate beneficial integration of different types of network components, such as optical fibers and electrical/electromagnetic components. By using techniques such as early wake notification, embodiments may reduce the burden on all system components.

Fig. 6A-B depict time graphs illustrating an example EEE ECE660 and fiber ECE640 power saving mechanism. In this example, the EEE ECE 600 is implemented in the UNI 310A in the user device 310A. Fiber ECE640 is implemented in ONU320A in this example and controls the fiber connection between OLT130 and ONU 120A.

As known to those skilled in the relevant arts, the EEE energy consumption and efficiency (EEE ECE660) mechanism as described improves ECE with low power idle states 625A-B, active states 635A-D, and reserved states 628A-B. Likewise, fiber ECE640 uses a dormant 620A-B state (optical dormant) and an active 630A-B state.

FIG. 6A depicts a non-limiting example of a conventional uncoordinated ECE mechanism-misaligned 650A-E. In the example shown, beginning at point 605, ONU320A goes directly into the sleep 620A state. At 635A, the user device 310A enters an active state and processes the data packet for transmission upstream. At point 606, user device 310A is active 635A and needs to transmit to ONU 320A. At the transmission point, however, ONU320A is in the dormant 620A state-thus resulting in misaligned 650A. In this example, the ONU320A receives network traffic and buffers it for transmission.

Other exemplary uncoordinated mechanisms M1-M4 are listed below and are depicted in FIG. 6A.

M1. misaligned 650B occurs when the active 630A state in ONU320A occurs simultaneously with the low power consumption state 625A in user device 310A.

M2. unaligned 650C occurs when the een 320A is in the dormant state 620B, when the EEE ECE660 on the user equipment appears as the active state 635B-C.

M3. misaligned 650D occurs when the een ECE660 on the user device 310A appears as a low power consumption state 625B when the OUN 320A is in the active state 620B.

M4. finally, misaligned 650E occurs when OUN 320A is in sleep state 620C and user device 310A is shown as active state 635D.

In FIG. 6B, an embodiment of the NPM210 aligns the sleep periods of the fiber ECE640 and the EEE ECE660, thus describing aligned examples 655A-E. In the example shown, beginning at point 607, ONU320A directly enters the active state 670A. At 635A, the user device 310A enters an active state and processes the data packet for transmission upstream. When user device 310A is in active state 625A and needs to transmit to ONU320A, ONU320A is also in active state (670A) -resulting in aligned 655A state. In this example, ONU320A receives network traffic without buffering the traffic prior to transmission.

Other example coordination mechanisms C1-C4 are listed below and described in FIG. 6B.

C1. The aligned 655B occurs when the sleep state 680A state in ONU320A occurs simultaneously with the low power consumption state 625A in user device 310A.

C2. Aligned 655C occurs when ONU320A is in active state 670B and EEE ECE660 on user device 310A appears to be in active state 635B-C.

C3. The alignment 655D occurs when the EEE ECE660 on the user device 310A appears in the low power state 625B when the ONU320A is in the sleep state 670B. By having coordinated data regarding low power state 625B and optical sleep 680B, increased power control options may be implemented. Devices with limited energy capabilities, for example, may perform a higher degree of energy limitation, resulting in better energy savings.

C4. Finally, aligning 655E occurs when ONU320A is in active state 670C and user device 310A appears to be active 635D.

Fig. 7A depicts ONU arrangement 730 having optical fiber 735 and 4 ports with electrical-cable/wired-connections. Fiber 735 has active modes 750A-B and optical power savings implemented during sleep modes 755A-B. Similarly, each port 720A-D has an active mode 765A-C and a sleep mode 760. In an embodiment, port 720A includes a sleep 770 that is implemented on a different port 720A-D based on communications from the ECE control mechanism.

Fig. 7B depicts a network topology 701 of an access network 705 containing UNIs 704 and annotations. Network topology 701 includes customer premises equipment 710A-D connected to Customer Premise Equipment (CPE) 790. CPE790 is a single physical component and has PHYs 722A-D, MACs 721A-E, switch 792, MAC721E, buffer 725A, and ONU 720. Network topology 701 further includes OLT740 and aggregation switch 745.

In the UNI 740 configuration example, MACs 721A-D may operate at gigabit speeds and may interface with PHYs 722A-D in CPE 790. Using EEE, PHYs 722A-D operate at three times the speed at which the PHY operates in 1000 BASE-T. The user devices 710A-D are also able to use EEE and may own their own PHYs and buffer in the system.

In the example of an access network configuration, MACs 721A-D in CPE790 are connected to ONU 720 through switch 792 and MAC 721E. ONU 720 is an EPON ONU capable of sleeping/power saving and buffering using buffer 725B. OLT740 is connected to aggregation switch 745. The functions of OLT740 and aggregation switch 745 may also be combined into a single physical component.

It is an object of an embodiment to coordinate the power saving protocol of the access network 705 with the components in the UNI 704. ONU 720, switch 792 and PHYs 722A-D are in one CPE790 device. As mentioned in the above example, despite the physical integration of these components in the CPE790, inefficiencies can result when separate protocols and uncoordinated connections are used to transport traffic. These inefficiencies arise each time network traffic travels from one domain to another (e.g., from the access network 705 to the UNI 704).

6A-B and 7A above, using the ECE control mechanism to align the active/sleep states of fiber and non-fiber components in the network topology 701 may result in a higher level of performance and lower power consumption. For example, by using the ECE mechanism, the active/sleep cycles of ONU 720 can be aligned with PHYs 722A-D and user devices 710A-D. In addition, ONU 720 may coordinate the period with OLT 740.

Method of producing a composite material

This section and FIGS. 8-11 summarize the techniques described herein by presenting a flow diagram of an example method of managing energy efficiency and control mechanisms in a network having a network energy manager (NPM) and a plurality of network components.

Fig. 8 illustrates a method 800 of managing energy efficiency and control mechanisms in a network having a network energy manager (NPM) and a plurality of network components, which is described with respect to the NPM receiving and processing energy information for at least one network component and is not intended to be limiting.

As shown in fig. 8, an embodiment of a method 800 begins at step 810 where energy information is received from at least one of a plurality of network components. In an embodiment, the NPM210 receives energy information, such as ECE information discussed above, from the aggregation switches 130 and ONUs120A-B in fig. 2. Examples of the energy information include physical layer (PHY) information, link information, ECE control policy information, and application information. Once step 810 is complete, method 800 proceeds to step 820.

At step 820, the received energy information is analyzed by the NPM 210. In an embodiment, the energy information includes ECE information from aggregation switches 130 and ONUs 120A-B. Once step 820 is complete, method 800 proceeds to step 830.

At step 830, configuration instructions are generated based on the analysis of the energy information. In an embodiment, the NPM210 generates configuration instructions for at least one network component (e.g., aggregation switch 130 and ONUs120A-B) used to gather energy information. Once step 830 is complete, method 800 proceeds to step 840.

At step 840, the configuration information is sent to at least one of the network components. In an embodiment, the NPM210 sends configuration instructions generated for the aggregation switches 130 and ONUs120A-B to each respective network component. Once step 840 is complete, method 800 ends.

Fig. 9 illustrates a method 900 of managing energy efficiency and control mechanisms in a network having a network energy manager (NPM) and a plurality of network components. The method is described with respect to receiving and processing buffered information from various network components and is not intended to be limiting.

As shown in fig. 9, an embodiment of a method 900 begins at step 910 where a buffer time for traffic of an originating network component, a first network component connected to the originating component, and a second network component connected to the first network component is received. In an embodiment, the NPM210 receives a buffer time of: an originating component (e.g., user equipment 310A), a first network component (e.g., ONU320A), a second network component (e.g., OLT 330). Once step 910 is complete, method 900 proceeds to step 920.

At step 920, the received buffer time is analyzed. As mentioned above, user satisfaction, performance, and cost savings all benefit from pooling (pool) buffering time in the user equipment 310A (e.g., customer) layer, since it is much easier to stop traffic at the source (user equipment 310A) than to buffer it farther away in the traffic path (e.g., ONU320A or OLT 330). The description provided is for exemplary purposes only, and the received buffer time may include: user equipment 310(100 ms), ONU320A (200 ms), and OLT330(500 ms), all buffering times are analyzed by NPM 210. Once step 920 is complete, method 900 proceeds to step 930.

At step 930, to increase the buffering time of the originating network component, configuration instructions are generated based on the buffering times of the first and second network components. Using the buffering time of the upstream components (ONU 320A and OLT330), the NPM210 generates configuration instructions to increase the buffering time of the originating network component, such as user equipment 310A. In this example, NPM210 generates configuration instructions to increase the buffering time of user equipment 310A by a value related to the sum of the buffering time of ONU320A (200 msec) and the buffering time of OLT330(500 msec), or 700 msec at this time. Once step 930 is complete, method 900 proceeds to step 940.

At step 940, configuration instructions are generated to reduce the buffering time of the first and second network components. Based on the present example, the buffering time of upstream components, such as ONU320A (200 msec) and OLT330(500 msec), is reduced to a minimum. Once step 940 is complete, method 900 proceeds to step 950.

At step 950, the configuration information is sent to the originating, first, and second network components. In an embodiment, the generated configuration instructions are sent to the user equipment 310A, ONU320A and the OLT 330. Once step 950 is complete, method 900 ends at step 960.

Fig. 10 illustrates a method 1000 of managing energy efficiency and control mechanisms in a network having a network energy manager (NPM) and a plurality of network components, which is described with respect to receiving and processing future traffic information and is not intended to be limiting.

As shown in fig. 10, an embodiment of a method 1000 begins at step 1010 where information describing a network traffic event associated with at least a first network component is received, where the network traffic event implies future traffic of a second network component. For example, the NPM210 may receive information: ONU320A will send upstream traffic to aggregation switch 340 via OLT330 at a specific time in the future (e.g., 800 milliseconds). Once step 1010 is complete, method 1000 proceeds to step 1020.

At step 1020, the received traffic events are analyzed. In an embodiment, as in the current example, the NPM210 has access to the following information: the aggregation switch 340 is now in a sleep state and requires a predefined wake-up period (e.g., 700 milliseconds); and OLT330 requires 400 milliseconds for traffic transmission from ONU 320A. Once step 1020 is complete, method 1000 proceeds to step 1030.

At step 1030, configuration instructions are generated for adjusting the dormancy setting of the second component based on analyzing the future traffic events. As in the present example, aggregation switch 340 will remain in the sleep state for as long as possible and wake up just before the arrival of traffic from ONU 320A. The NPM210 generates configuration instructions based on the transmission time of the traffic event from ONU320A (800 msec), the transmission time through OLT330 (400 msec), and the wakeup period of the aggregation switch (700 msec). Based on the above, the configuration instructions will set the aggregation switch 340 to wake up after about 500 milliseconds ((400+800) -700 milliseconds). Once step 1030 is complete, method 1000 proceeds to step 1040.

At step 1040, the configuration instruction is sent to the second component. In the present example, the generated configuration instructions are sent to aggregation switch 340 to begin waking after approximately 500 milliseconds. Once step 1040 is complete, method 1000 ends at step 1050.

Fig. 11 illustrates a method 1100 of managing energy efficiency and control mechanisms in a network having a network energy manager (NPM) and a plurality of network components, which is described with respect to receiving and processing various link speed and status information and is not intended to be limiting.

As shown in fig. 11, an embodiment of a method 1100 begins at step 1110 where information corresponding to a link speed and a component buffer for a link between two network components is received. For example, the NPM210 receives information corresponding to the link speed of the link between the ONU320A and the OLT330 and receives the buffer time of the OLT 330. Once step 1110 is complete, method 1100 proceeds to step 1120.

At step 1120, the received link speed information and buffer time are analyzed. For example, the NPM210 has access to the buffering time (500 milliseconds) required by the OLT330 and determines that adjusting the link speed reduces the buffering time of the OLT 330. Once step 1120 is complete, method 1100 proceeds to step 1130.

At step 1130, configuration information is generated to modify the link speed to reduce buffering requirements associated with the network component. In the present example, based on the buffering time required by the OLT330(500 milliseconds), the NPM210 generates configuration instructions to slow down the link between the ONU320A and the OLT 330. By slowing down the link speed so that traffic requires an additional 500 milliseconds to reach OLT330, the buffering requirements of OLT330 can be reduced or eliminated. . Once step 1130 is complete, method 1100 proceeds to step 1140.

At step 1140, the generated configuration instructions are sent to the network component associated with the link. In an embodiment, because ONU320 can control the link speed, configuration instructions are sent to this. Once step 1140 is complete, method 1100 ends at step 1160.

NPM function implementation

The manager functions of the present invention, such as a network energy manager (NPN), may be implemented in hardware, software, or some related combination. For example, based on the description of the invention given, one skilled in the relevant art will appreciate that the NPM functionality may be implemented using a computer processor, computer logic, an Application Specific Integrated Circuit (ASIC), or the like. Thus, any processor that performs the data collection, policy management, coordination, analysis functions described herein is within the scope and spirit of the present invention. For example, embodiments of the NPMs 210, 410A-D use processors to perform functions such as data collection and management functions.

Further, the NPM functionality described herein may be embodied by computer program instructions that may be executed by a computer processor or any of the hardware devices listed above. The computer program instructions direct the processor to perform the NPM functions described herein. The computer program instructions (e.g., software) may be stored on a computer usable medium, a computer program medium, or any computer readable storage medium accessible by a computer or processor. These media include memory devices such as RAM or ROM, or other types of computer storage media such as a computer hard disk or CD ROM, or equivalents thereof. Thus, any computer storage medium containing computer program code capable of causing a processor to perform the data collection, policy management, coordination, analysis functions described herein, as well as other related functions, is within the scope and spirit of the present invention.

Conclusion

While various embodiments of the present invention have been described above, it should be understood that they have been presented by way of example only, and not limitation. It will be apparent to persons skilled in the relevant art that various changes in form and detail can be made therein without departing from the spirit and scope of the invention. Accordingly, the invention is not to be restricted except in light of the attached claims and their equivalents.

Claims (10)

1. A system for managing energy efficiency and control mechanisms in a communication network, the communication network including a plurality of network components, the system comprising:

a network energy manager (NPM) coupled to at least one of the plurality of network components, wherein the NPM is configured to:

receiving energy information from at least one of the plurality of network components;

analyzing the energy information;

generating configuration instructions based on the analysis of the energy information; and

sending a configuration instruction to at least one of the network components.

2. The system of claim 1, wherein the NPM is further configured to:

receiving configuration information from at least one of the network components; and

sending the configuration information to at least one of the network components.

3. The system of claim 1, wherein the energy information comprises an operational characteristic of one of the plurality of network components.

4. The system of claim 3, wherein the operational characteristic is:

a supported link speed available to the network component, or an operational mode available to the network component.

5. The system of claim 1, wherein the configuration instructions comprise at least one of routing information and switching information for traffic on the network.

6. The system of claim 1, wherein the configuration instructions comprise a control strategy for controlling energy efficiency and control mechanisms.

7. The system of claim 1, wherein the energy information received by the NPM comprises a link utilization level of at least one of the plurality of network components.

8. The system of claim 1, wherein the energy information received by the NPM comprises characteristics of a link between two of the plurality of network components.

9. A method for managing energy efficiency and control mechanisms in a network, the network including a network energy manager (NPM) and a plurality of network components, the method comprising:

receiving energy information from at least one of the plurality of network components;

analyzing the energy information;

generating configuration instructions based on the analysis of the energy information; and

sending the configuration instructions to at least one of the network components.

10. The method of claim 9, wherein the energy information comprises an operational characteristic of one of the plurality of network components.