WO2019012546A1

WO2019012546A1 - Efficient load balancing mechanism for switches in a software defined network

Info

Publication number: WO2019012546A1
Application number: PCT/IN2017/050287
Authority: WO
Inventors: Faseela K; Ashutosh Bisht
Original assignee: Telefonaktiebolaget Lm Ericsson [Publ]
Priority date: 2017-07-11
Filing date: 2017-07-11
Publication date: 2019-01-17

Abstract

A method is implemented by a network device where the method is for autonomous dynamic load balancing at the network device functioning as a data plane node (DPN) in a software defined networking (SDN) network. The DPN forwards data traffic to a set of load-balanced virtual network function (VNF) instances proportionate to a weighting of load balance for the set of load-balanced VNF instances. The method analyzes collected timestamps for data traffic that traversed the set of load-balanced VNF instances to determine processing delay for each VNF instance in the set of load-balanced VNF instances and updates the weighting of load balance to reflect determined delay times for each VNF instance.

Description

EFFICIENT LOAD BALANCING MECHANISM FOR SWITCHES IN A

SOFTWARE DEFINED NETWORK

TECHNICAL FIELD

[0001] Embodiments of the invention relate to the field of load balancing for traffic handled by software defined networking (SDN) networks; and more specifically, to the monitoring of traffic using timestamps and using the monitored timestamp information to determine the load on associated services and balance the load between these associated services.

BACKGROUND

[0002] Software Defined Networking (SDN) is a networking paradigm that has gained increased interest and usage in network architecture design and research. Under SDN, the data plane and the control plane are separated. The data path nodes are simple forwarding engines that are programmed by a controller using a set of flow control protocol rules. One example flow control protocol is OpenFlow (OF) developed by the Open Networking Foundation (ONF). Such forwarding engines are called switches, e.g., OF switches where OF is utilized. The central controller is responsible for exposing the programming interface to user applications via standard north bound interfaces. The benefits of Software Defined Networking include remote administration, reduced costs for switches, enhanced configurability, enhanced upgradability and similar advantages.

[0003] Administrators of networks including SDN networks seek to manage network performance, which includes monitoring metrics of performance. Such metrics include network delay. Network delay is defined as the time taken by a set of related packets, referred to as a flow, as it traverses a set of network elements all managed by a single administrative entity. One can estimate it by measuring the time when the packets ingress the network and measuring the time again when the packets egress the network. The difference between the two times would give the estimate of the network delay for that packet. To minimize measurement errors, the average of a number of such measurements is provided as a network delay estimate.

[0004] Network delay can be defined for individual flows, aggregate of flows or on a per application basis. Network delay is an important metric of network performance. Often delay and jitter (delay variance) measurement values are tied to service level agreements (SLAs), which are contracts between network operators and customers for specific performance levels. Therefore, it is necessary for network operators and administrators to develop tools that can measure network delay accurately.

[0005] In many networks, there are a series of services that are applied to most traffic that traverses the network. The services can be referred to as a service chain, since they form a sequence of services that may be applied in a sequence or 'chain.' Metrics for these service chains are important for network administrators and operators. Network delay on service chains is the total delay experienced by a packet on the service chain. The network delay for the service chains is the time spent by a packet between the service plane entry point, which is the point at which the packet ingresses the service chain, and the service plane exit point, which is the point at which the packet egresses the service chain. In a topology where network service header (NSH) is used with a service function chain the network delay may be between a first service function (si) and a second service function (s2) and NSH may be used with encapsulated traffic. The measurement of the network delay for service level agreement (SLA) verification and network trouble shooting is essential as in the case of traditional networks.

[0006] Additionally, in case of service chaining networks residing in the cloud where network elements are controlled by a centralized SDN controller, the measured information on a live network can be used to measure information that can be fed into the SDN controller which can use it for service chain path load balancing. This would not be possible in traditional networks where packet path selection is decided by distributed routing protocols. In a cloud environment, the measured information can be used for determining the virtual machine load and behavior analysis and this analysis can help in virtual network function (VNF) placements elasticity (e.g., spawning new virtual machines (VMs), shutting down VMs, VM migration and similar functions). Thus, generating accurate service chain delay measurements is an important feature in service chaining solutions.

SUMMARY

[0007] In one embodiment, a method is implemented by a network device where the method is for autonomous dynamic load balancing at the network device functioning as a data plane node (DPN) in a software defined networking (SDN) network. The DPN forwards data traffic to a set of load-balanced virtual network function (VNF) instances proportionate to a weighting of load balance for the set of load -balanced VNF instances. The method analyzes collected timestamps for data traffic that traversed the set of load-balanced VNF instances to determine processing delay for each VNF instance in the set of load- balanced VNF instances and updates the weighting of load balance to reflect determined delay times for each VNF instance.

[0008] In another embodiment, a network device is configured to implement the method for autonomous dynamic load balancing at the network device functioning as the DPN in the SDN network where the DPN forwards data traffic to the set of load-balanced VNF instances proportionate to a weighting of load balance for the set of load-balanced VNF instances. The network device includes a non-transitory computer readable medium having stored therein a delay analyzer, a flow control pipeline to forward the data traffic according to a load balance, and a processor configured to execute the delay analyzer. The delay analyzer analyzes collected timestamps for data traffic that traversed the set of load- balanced VNF instances to determine processing delay for each VNF instance in the set of load-balanced VNF instances, and updates the weighting of load balance to reflect determined delay times for each VNF instance.

[0009] In one embodiment, a computing device is in communication with the network device. The computing device executes a plurality of virtual machines for implementing network function virtualization (NFV). The computing device is configured to execute the method for autonomous dynamic load balancing at the network device functioning as the DPN in the SDN network, the DPN forwarding data traffic to the set of load-balanced VNF instances proportionate to a weighting of load balance for the set of load-balanced VNF instances. The computing device includes a non-transitory computer readable medium having stored therein a delay analyzer, and a processor configured to execute a virtual machine from the plurality of virtual machines. The virtual machine executes the delay analyzer. The delay analyzer analyzes collected timestamps for data traffic that traversed the set of load-balanced VNF instances to determine processing delay for each VNF instance in the set of load-balanced VNF instances, and updates the weighting of load balance to reflect determined delay times for each VNF instance.

[0010] In a further embodiment, a control plane device is in communication with the network device, where the control plane device is configured to execute the method for autonomous dynamic load balancing at the network device functioning as the DPN in a SDN network. The DPN forwards data traffic to the set of load-balanced VNF instances proportionate to a weighting of load balance for the set of load -balanced VNF instances. The control plane device includes a non-transitory computer readable medium having stored therein a delay analyzer, and a processor configured to execute the delay analyzer. The delay analyzer analyzes collected timestamps for data traffic that traversed the set of load- balanced VNF instances to determine processing delay for each VNF instance in the set of load-balanced VNF instances, and updates the weighting of load balance to reflect determined delay times for each VNF instance.

BRIEF DESCRIPTION OF THE DRAWINGS [0011] The invention may best be understood by referring to the following description and accompanying drawings that are used to illustrate embodiments of the invention. In the drawings:

[0012] Figure 1 is a diagram of one embodiment of service chaining in a software defined networking (SDN) network.

[0013] Figure 2 is a diagram of one embodiment of connectivity of a data plane node (DPN) in the SDN network.

[0014] Figure 3A is a diagram of one embodiment of the process for measuring VNF delays.

[0015] Figure 3B is a flowchart of one embodiment of the process of the DPN to implement dynamic load balancing.

[0016] Figure 4A is a diagram of the components of the DPN that implement the dynamic load balancing process. The diagram is provided by way of example to provide details about each component and their function that is relevant for the dynamic load balancing process.

[0017] Figure 4B is a diagram of an example configuration of the group table for marking selected data packets.

[0018] Figure 5 is a diagram of the components of the switch for load balancing and delay measurement.

[0019] Figure 6 is a diagram of one embodiment of the components of the switch and their interaction with the northbound interface.

[0020] Figure 7A illustrates connectivity between network devices (NDs) within an exemplary network, as well as three exemplary implementations of the NDs, according to some embodiments of the invention.

[0021] Figure 7B illustrates an exemplary way to implement a special-purpose network device according to some embodiments of the invention.

[0022] Figure 7C illustrates various exemplary ways in which virtual network elements (VNEs) may be coupled according to some embodiments of the invention.

[0023] Figure 7D illustrates a network with a single network element (NE) on each of the NDs, and within this straight forward approach contrasts a traditional distributed approach (commonly used by traditional routers) with a centralized approach for maintaining reachability and forwarding information (also called network control), according to some embodiments of the invention.

[0024] Figure 7E illustrates the simple case of where each of the NDs implements a single NE, but a centralized control plane has abstracted multiple of the NEs in different NDs into (to represent) a single NE in one of the virtual network(s), according to some embodiments of the invention.

[0025] Figure 7F illustrates a case where multiple VNEs are implemented on different NDs and are coupled to each other, and where a centralized control plane has abstracted these multiple VNEs such that they appear as a single VNE within one of the virtual networks, according to some embodiments of the invention.

[0026] Figure 8 illustrates a general-purpose control plane device with centralized control plane (CCP) software 850), according to some embodiments of the invention.

DETAILED DESCRIPTION [0027] The following description describes methods and apparatus for improving load balancing using network performance metrics for a software defined networking (SDN) network. Specifically, the embodiments provide a method and apparatus for improving the accuracy and efficiency of network metrics using timestamping packets at the switches. The embodiments improve load balancing by taking into consideration the current processing load on associated services and the operational state of these services. The embodiments track timestamped traffic traversing the services and determine traffic delays for each service. The calculated delays are utilized to update load balancing processes to steer traffic away from services with higher delays.

[0028] In the following description, numerous specific details such as logic implementations, opcodes, means to specify operands, resource partitioning/sharing/duplication implementations, types and interrelationships of system components, and logic partitioning/integration choices are set forth in order to provide a more thorough understanding of the present invention. It will be appreciated, however, by one skilled in the art that the invention may be practiced without such specific details. In other instances, control structures, gate level circuits and full software instruction sequences have not been shown in detail in order not to obscure the invention. Those of ordinary skill in the art, with the included descriptions, will be able to implement appropriate functionality without undue experimentation. [0029] References in the specification to "one embodiment," "an embodiment," "an example embodiment," etc., indicate that the embodiment described may include a particular feature, structure, or characteristic, but every embodiment may not necessarily include the particular feature, structure, or characteristic. Moreover, such phrases are not necessarily referring to the same embodiment. Further, 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 affect such feature, structure, or characteristic in connection with other embodiments whether or not explicitly described.

[0030] Bracketed text and blocks with dashed borders (e.g., large dashes, small dashes, dot-dash, and dots) may be used herein to illustrate optional operations that add additional features to embodiments of the invention. However, such notation should not be taken to mean that these are the only options or optional operations, and/or that blocks with solid borders are not optional in certain embodiments of the invention.

[0031] In the following description and claims, the terms "coupled" and "connected," along with their derivatives, may be used. It should be understood that these terms are not intended as synonyms for each other. "Coupled" is used to indicate that two or more elements, which may or may not be in direct physical or electrical contact with each other, co-operate or interact with each other. "Connected" is used to indicate the establishment of communication between two or more elements that are coupled with each other.

[0032] An electronic device stores and transmits (internally and/or with other electronic devices over a network) code (which is composed of software instructions and which is sometimes referred to as computer program code or a computer program) and/or data using machine-readable media (also called computer-readable media), such as machine -readable storage media (e.g., magnetic disks, optical disks, read only memory (ROM), flash memory devices, phase change memory) and machine-readable transmission media (also called a carrier) (e.g., electrical, optical, radio, acoustical or other form of propagated signals - such as carrier waves, infrared signals). Thus, an electronic device (e.g., a computer) includes hardware and software, such as a set of one or more processors coupled to one or more machine-readable storage media to store code for execution on the set of processors and/or to store data. A 'set,' as used herein, refers to any positive whole number of items including one item. For instance, an electronic device may include non-volatile memory containing the code since the non-volatile memory can persist code/data even when the electronic device is turned off (when power is removed), and while the electronic device is turned on that part of the code that is to be executed by the processor(s) of that electronic device is typically copied from the slower non-volatile memory into volatile memory (e.g., dynamic random access memory (DRAM), static random access memory (SRAM)) of that electronic device. Typical electronic devices also include a set or one or more physical network interface(s) to establish network connections (to transmit and/or receive code and/or data using propagating signals) with other electronic devices. One or more parts of an embodiment of the invention may be implemented using different combinations of software, firmware, and/or hardware.

[0033] A network device (ND) is an electronic device that communicatively interconnects other electronic devices on the network (e.g., other network devices, end-user devices). Some network devices are "multiple services network devices" that provide support for multiple networking functions (e.g., routing, bridging, switching, Layer 2 aggregation, session border control, Quality of Service, and/or subscriber management), and/or provide support for multiple application services (e.g., data, voice, and video).

[0034] Network delay measurements such as two-way active measurement protocol (TWAMP) are the most commonly used path delay measurement protocols in traditional routers and switches. TWAMP can be used for both one-way and two-way delay determination. A test stream of UDP based request-response messages are generated and timestamped at the two ends of the path whose delay is to be measured. TWAMP relies on the two endpoints being time-synced using network time protocol (NTP) or precision time protocol (PTP), where PTP provides greater (microsecond) accuracy. The timestamp differences convey the path delay. TWAMP does not measure the delay experienced by the real traffic on the path, it measures the delay experienced by a test stream, which is a set of test packets sent between the two endpoints. The assumption is that this measurement is close to the delay experienced by the real traffic. This might be a valid assumption in the case of traditional network paths which typically consist of packet forwarding nodes that are application unaware. However, in other scenarios real traffic may be forwarded significantly differently from the test packet stream.

[0035] Service chain delay measurements using network service header (NSH) is another network metric collection mechanism. The NSH is a protocol developed specifically for service chains. A NSH is added to traffic that is traversing a service chain. The NSH is added in the packet header to provide a separate service plane independent of the transport protocol. The NSH defines a sequence of service functions that the corresponding packet must traverse prior to reaching its ultimate destination. The NSH is inserted by a switch at the ingress into the network or service chain. The NSH may be inserted between the original packet and any encapsulation (e.g., tunneling or label swap protocol encapsulation). The embodiments work with network devices and service functions that understand the NSH.

[0036] In the embodiments, the packet entry-point adds the timestamps to the traffic at the time of entry and this information is carried in the NSH (e.g., in the service platform context field of NSH). Similarly, the egress-point of the service chain also adds the timestamps at the time of exit in the NSH. The switch at the exit point of the service chain or network transmits the information from the NSH including the timestamp at the entry point (which is obtained from the NSH) and the timestamp at the exit point, to the controller. The controller or other monitoring application then calculates the difference to identify the network delay. Since a spare field in NSH is used for carrying the timestamp, the original packet is not modified.

[0037] The embodiments utilize entry and exit points to the service chain or network that are time synced using any process or protocol like NTP or PTP to ensure the synchronization. The embodiments also use the timestamping capability that is available in the data plane of the switches. In some embodiments, special measurement packets may be sent along the service plane.

[0038] SDN and Service Chaining

[0039] SDN networks may be implemented using flow control protocols to enable a controller to configure the forwarding processes of the data plane nodes of the SDN network. As discussed above, an SDN network may be based on a flow control protocol (e.g., the OpenFlow protocol) or similar protocol for programmatic access of the data plane of the SDN network. The nodes or switches in data plane are called DPNs (data plane nodes). These DPNs are controlled and configured by a distinct node called the controller. The controller and DPN are usually physically distinct nodes. Some SDN architectures are based on open networking standards (e.g., a white box DPN). This means that the embodiments are compatible with standards based operations and interoperable with other SDN architectures. The OpenFlow protocol is an example of such an open standard to enable control of a DPN by the controller. In other embodiments, proprietary flow control and SDN architectures may be used or may be used in combination with open standard based SDN architectural components.

[0040] SDN networks support service chaining. Service chaining is a process where data traffic is processed by a set of functions or services often across a number of network devices. These services are sometimes applied by Internet service providers and similar network operators to perform administrative and accounting services like authentication, firewalls, security, uniform resource locator (URL) filtering, deep packet inspection (DPI) and similar services. These services or functions can be implemented via virtualization via virtual network function (VNFs). For example, service chaining can be used to handle subscriber traffic such that the traffic is steered through a pre-defined sequence of services each implemented as VNFs. The VNFs provide different functionality such as content- caching, content filtering, security and similar functions. In some embodiments, there may be multiple instances of a given VNF such that load can be distributed across these VNF instances. For example, a VNF for security in a service chain may be implemented by a set of VNF instances where the traffic to be processed for security can be distributed across these VNF instances.

[0041] Figure 1 is a diagram of one example of a service chain in an SDN network. In this example embodiment, subscriber traffic is subjected to DPI, security processing and URL filtering. These services can be implemented to enforce parental control processes or similar processes. In the example, each of these services is implemented by a VNF or set of VNF instances. The services can be implemented or distributed over any number of VNF instances. In the example, subscriber traffic 1 traverses a DPN that has been configured by an SDN controller. The DPN sends subscriber traffic 1 to each of the VNF instances for processing. In this example, subscriber traffic 2 may not be subject to the same service chain, thus, the SDN controller has not configured the DPN to send this traffic to the service chain.

[0042] Use of an SDN network to implement service chaining allows the steering of traffic based on highly granular subscriber profile and application profile information. It also allows an operator to quickly and easily introduce new services / VNFs for subscribers. An SDN based service chaining implementation may be considered to include a transport domain and a service domain. The transport domain may include at least one DPN that is responsible for forwarding the subscriber traffic to each service / VNF. The transport domain can also be distributed over multiple DPNs. The service domain includes the services / VNFs that implement the functionality of the service chain.

[0043] Figure 2 is a diagram of one embodiment of a DPN connectivity in an SDN network implementing a service chain. The service nodes (i.e., the VNF instances/nodes), are connected to the SDN transport domain typically via a virtual local area network (VLAN), using a virtual extensible LAN (VxLAN), or through a similar network configuration. Unlike DPN (which are part of transport domain and can be considered to be robust Layer 2 switches), the VNF instances are more akin to servers. As illustrated in Figure 2, a DPN may be connected to a set of VNF instances/nodes as well as the controller and a subscriber node. In the example, the DPN is connected to a set of VNF instances via a VLAN, a subscriber node via a VxLAN and a controller via an SDN protocol (e.g., OpenFlow). A 'set,' as used herein, refers to any positive whole number of items including one item.

[0044] To detect a link failure between DPNs, an SDN controller employs techniques that are similar to the ones used to detect link failure between Layer 2/ Ethernet switches (since DPNs are similar to Layer 2 / Ethernet switches). These methods, for example, include use of link layer discovery protocol (LLDP) or bidirectional forwarding detection (BFD) protocol. Such techniques are, however, not sufficient for monitoring VNF instances. This is because VNF instances are similar to servers rather than Layer 2 / Ethernet switches. The VNF instances are instead monitored at the application layer. For application layer monitoring requires different techniques, processes and protocols to check the connectivity and the availability of these VNF instances.

[0045] In one example embodiment, an open standard application layer manager may be employed, e.g., Tacker is generic VNF Manager (VNFM) and a network function virtualization (NFV) Orchestrator (NFVO) implemented in conjunction with OpenFlow to deploy and operate network services and VNFs on an NFV infrastructure platform like OpenStack. One of the functions of Tacker is the health monitoring of deployed VNF. Tacker monitoring framework provides the NFV operators and VNF vendors the ability to write a pluggable driver that monitors the various status conditions of the VNF entities it deploys and manages. However, the Tacker monitoring framework implements only simple monitoring and integrates with an external monitoring system for advanced monitoring. In some embodiments, the processes presented herein can be used as external monitoring system that integrates with Tacker. The simple monitoring defined in Tacker involves pinging the management IP-address of a given service.

[0046] The NFV European Telecommunications Standards Institute (ETSI) Industry Specification Group (ISG) has produced and approved a report on active monitoring that proposes an active monitoring framework for NFV. The key elements or components of this framework are (1) a test controller, which maintains a catalogue of virtual test agents, and (2) virtual test agents (VTAs), which are similar to a traditional physical test agent, this agent provides network visibility and fault tolerance in an NFV based network. Unlike physical active test agents, the test agent is a virtual entity, so that it can be effective in NFV environment. Another component is the test result analysis module (TRAM), where this module gets the test results from virtual test agents and subsequently provides the processed results to presentation module in OSS/BSS.

[0047] Load Balancing in SDN networks where network function virtualization (NFV) is implemented (e.g., with the use of the OpenDaylight project) may utilize a process that selects packets from a set of buckets in a group for SFC (Service Function Chaining). An example of such a process may be the use of OpenFlow SELECT groups, which execute one bucket-of-action from multiple buckets-of-actions in the group. Packets are processed by a single bucket in the group, based on a switch-computed selection algorithm (e.g., hash on some user-configured tuple or simple round robin).

[0048] The configuration and state for the selection algorithm is external to the flow control protocol of the SDN (e.g., OpenFlow). The selection algorithm should implement equal load sharing amongst the VNF instances and can optionally be biased based on bucket weights. When an egress port specified in a bucket in a select group goes down, the switch may restrict bucket selection to the set of remaining buckets (those associated with forwarding actions to live ports) instead of dropping packets destined to that port. This behavior may reduce the disruption of a downed link or switch. However, this load balancing processes has limitations and inefficiencies.

[0049] The embodiments are designed to overcome the problems of the load balancing scheme using select groups. The specific problems and limitations of the load balancing scheme using select groups include the load balancing scheme does not take into consideration the current processing load on the VNF instances in the service chain. Although the load balancing may be based on the configured static bucket weights, it does not adjust based on dynamically changing processing load on VNF instances. The load balancing scheme also does not specify any mechanism to incorporate the operational state of the VNF instances in computing the bucket selection of select groups. As an example, even if the VNF instance goes down, the load balancing scheme provides no mechanism to detect such condition and redirect traffic. In addition, the load balancing scheme does not take into account the network congestion towards VNF instances.

[0050] The embodiments overcome these limitations of the load balancing scheme with a dynamic load balancing process. The dynamic load balancing process utilizes information from packet timestamping for that DPNs (e.g., OpenFlow switches) that is based on a realtime clock. The DPNs timestamp a selection of packets sent through the service chain to determine delay conditions with each of the VNFs in the service chain. The time taken by a VNF instance to respond to the traffic sent by the DPN is measured by the DPN to dynamically influence the select group bucket selection algorithm.

[0051] Figure 3A is a diagram of one embodiment of the process for measuring VNF delays. The controller configures the DPN that is managing the load balancing using a flow control protocol (e.g., OpenFlow) (1). An initial NSH template (a 'dummy' NSH packet) is provided to the DPN to be utilized as a template for insertion into a selection of traffic that is being sent to each of the services (e.g., implemented by VNF instances) in the service chain (2). Using the NSH template a timestamp is inserted into some traffic being sent to services in the service chain (3). This traffic (i.e., data packets) is processed at each step of the service chain by each service and returned to the DPN (3, 4, 5). In some embodiments, the NSH template or similar packet may be looped back to generate packets that are used specifically for delay testing. Each time that the NSH packets traverse the DPN, their timestamp information is collected and/or updated. This results in a collection of timing information that can be used to determine delays for packet processing across each service dynamically and in real-time.

[0052] Figure 3B is a flowchart of one embodiment of the process of the DPN to implement dynamic load balancing. The DPN receives configuration information from the controller that indicates an initial load balance for the set of VNF instances that are load balanced by the DPN (Block 301). The set of services can be implemented as instances of a VNF. The load balancing can be implemented as a select group algorithm (e.g., the OpenFlow Select Groups algorithm). The controller may configure timestamping and delay analysis as discussed herein above before, after or in parallel with the load balancing configuration. A subset of the data traffic being sent to each VNF instance is timestamped by the DPN as part of the delay analysis process and sent to each of the VNF instances with the other data traffic to be serviced (Block 303). The timestamped packets are marked using a unique mechanism so that they can be identified upon return to the DPN. The uniquely marked incoming data traffic is timestamped or the timestamp is recorded upon arrival and the timestamp is updated at the time the data traffic is sent to the next VNF instance in the set of load-balanced VNF instances (Block 305). A check may be made as to whether all the VNF instances have been traversed for each data packet (Block 307). Where the VNF instances/services have not all been traversed by a given packet then the packet is sent to the next VNF instance. Those packets with timestamps are updated at the time of departure and arrival or these times are recorded by the DPN. [0053] Once all of the services/VNF instances have been traversed or as each VNF instance traversal is completed, the process can determine the delay for each traversed VNF instance based on the collected timestamp information from those subset of timestamped packets sent to the respective VNF instance (Block 309). The delay analysis can be implemented by a delay analyzer module within the DPN or at the controller. The delay analyzer determines the total time taken to process the timestamped messages as well as to send and receive them for each of the VNF instances. The weighting for service selection (e.g., between multiple instances of or similar services, between order of services to be performed or similar options for changing the data traffic processing within a set of services or VNF instances of a service) is updated to reflect the computed delays (Block 311). In some embodiments, the delay information is used to influence the SELECT group algorithm for bucket selection. The load balancing may be adjusted to shift data traffic away from services (i.e., VNF instances of a given service) with higher delays toward VNF instance with lower delays. Where traffic is not returned, then the VNF instance can be removed from set of available VNF instances of that service such that the data traffic can be steered to available VNF instances (Block 313).

[0054] The embodiments provide advantages over the prior load balancing scheme, by providing a robust load balancing mechanism for services (i.e., VNF instances) by taking into consideration the live state of the VNF instances such as an overloaded VNF instance providing a given service. The embodiments of the dynamic load balancing mechanism take into effect the network congestion between DPN and the services (i.e., each VNF instance) as the total delay is measured which incorporates the congestion between the DPN and VNF instance. The embodiments of the dynamic load balancing mechanism are locally implemented by the DPN and configured by controller. The embodiments of the dynamic load balancing mechanism are agnostic to service type or VNF type, i.e., it can be used for any type of service or VNF instance. Further, the embodiments of the dynamic load balancing mechanism does not put any new requirements on the services and/or VNF instances and is contained with the controller and DPN.

[0055] Figure 4A is a diagram of the components of the DPN that implement the dynamic load balancing process. The diagram is provided by way of example to provide details about each component and their function that is relevant for the dynamic load balancing process. However, one skilled in the art would understand that other components of a DPN have been omitted for sake of clarity and conciseness. The configuration of the functions as illustrated and described is provided by way of example and not limitation. Any of the functions may be combined or sub -divided as would be understood by one skilled in the art consistent with the principles, processes and structures of the embodiments.

[0056] The described embodiments are described with reference to timestamping being implemented by use of NSH (Network Service Header) for storing timestamp values. Although the example embodiments are described in terms of NSH, the embodiments are not limited to NSH based deployment. Other fields of data traffic processed as part of the service chain can also be used for storing timestamp values.

[0057] The components of the DPN 401 can include a time stamper 403, selective time stamper and marker 405, ingress traffic handler 407, load balancer 409, flow control pipeline 411, data analyzer 413 and response traffic handler 415 amongst other components. The DPN 401 can be a switch or network device in any type of network and connected to any number of services over any number of communication links and ports.

[0058] The time stamper 403 is a function that timestamps a fraction of data packets traversing the DPN (e.g., NSH based data traffic being forwarded towards a service implemented as a set of VNF instances) before it leaves the DPN. NSH supports holding timestamp information and similar protocols and packet types can be utilized for holding timestamp information. This time stamper 403 is responsible for timestamping the response data traffic from services as well as the data traffic destined for the services.

[0059] The selective time stamper and marker 405 is responsible for determining the subset of received data traffic to be timestamped by the time stamper 403. Adding a timestamp to all data packets that traverse the DPN would cause an unacceptably high overhead. As a result, the embodiments timestamp only a fraction of data packets for delay calculation. The selective time stamper and marker 405 identifies the specific data packets to timestamp, which will be a fraction of actual data traffic going to services, so that it can be used for delay calculation. When a fraction of the egress data packets are timestamped, exactly the same data packets need to be identified for timestamping when they arrive back at the DPN from the services (e.g., from VNF instances for the service). For this, the timestamped fraction of the data packets needs to be marked using some unique mechanism. In one embodiment, the time stamper and marker 405 is configured to split the whole NSH service path id space into two buckets. One bucket will be used for global service chaining (the usual use). The other bucket will be used for local decisions between the DPN and destination service. The reason for doing this split is because in NSH based architecture all forwarding decision are based on the NSH header. Having a pool of service IDs for local decisions between the DPN and service allows for using NSH header but with local adaptations.

[0060] In one embodiment, the time stamper and marker further is configured to use even numbered NSH Service Paths for the usual service chaining, and odd numbered Service Paths to indicate timestamp marking. Whenever actual data traffic is received by a DPN, it will have even numbered service path (i.e., the least significant bit (LSB) will be zero). If this particular packet is chosen to be timestamped, the LSB of NSH Service Path Id is set to 1, to indicate that this is timestamped. In other embodiments, the inverse arrangement or similar arrangement is used. Whenever the data traffic returns from the service, the LSB of Service Path is again matched to see if the LSB is non-zero, i.e., if it is a timestamped packet. If it is a timestamped packet, a copy of the timestamped packet will be sent to delay analyzer, after which the Service Path LSB will be reset to 0, and then the data traffic will be forwarded as usual.

[0061] In one embodiment, in order to mark only a fraction of data traffic, the controller may use the SELECT_GROUP construct of the control flow protocol (e.g., the OpenFlow protocol) with multiple action buckets. One action bucket for this SELECT group will have the following additional actions apart from the normal forwarding actions: (1) set LSB in NSH header (to make it next ODD number); and (2) add current timestamp to NSH.

[0062] Figure 4B is a diagram of an example configuration of the group table for marking selected data packets. As an example, the diagram identifies on configuration to mark one- fourth of traffic with a timestamp and a unique identifier where there are four buckets for a given service (i.e., a set of VNF instances of the service) and one of the four adds the timestamping and marking. This group table is part of the flow control pipeline 411.

[0063] Returning to the discussion of Figure 4A, the response traffic handler 415 manages data traffic received from the services (e.g., from VNF instances). This handler manages the flow entries to identify the time stamped packets coming from services (VNF instances) to the DPN. In one embodiment, each flow entry for identifying returned marked data traffic contains a match pattern based on NSH headers that uniquely identifies that the data packet is the timestamped traffic; an action to time stamp the packet in the NSH header. This timestamp is added to accurately identify the time packet reached back to DPN; an action to PUNT (i.e., forward) a copy of the packet to the controller and/or the data analyzer; and a reset NSH service path id to original even value. While the response traffic handler 415 may be shown as a separate component, it may also be conceived as the flow entries themselves and a part of the flow control pipeline 411. [0064] The following table entry shows a match entry for the dummy packet coming from a VNF instance of a service, where the entry has an inactivity timeout value of 5 sec.

Table I

[0065] The delay analyzer 413 in the DPN 401 analyzes the delay incurred by various services (VNF instances) for processing their respective traffic. In some embodiments, the response traffic handler 415 includes an action to punt (forward) a copy of all timestamped traffic to the controller via a PACKET-IN message or similar message. All such messages going from the DPN to the controller pass through or are intercepted by the delay analyzer module 413. When a PACKET-IN message indicates that it includes a locally timestamped NSH message, the data analyzer 413 can locally consume the packet, instead of sending the packet to the controller. The data analyzer 413 may be implemented as a logical table outside of the control flow pipeline 411. The table implementing the data analyzer has match criteria (to match timestamped traffic from services (VNF instances)). The match criteria may include of a Table identifier, cookie field and ingress port so that the component can identify that the data packets are timestamped and to be handled by the delay analyzer 413. Each of these fields is present in and can be utilized in the PACKET-IN message. Figure 5 is a diagram of one embodiment of handling of data packet processing by the delay analyzer.

[0066] Once a data packet is received by the data analyzer 413 and identified for processing by the data analyzer, the data analyzer 413 may compare the egress and ingress timestamps of the data packet and calculate the time taken by the service (VNF instance) to process the data packet. The data analyzer 413 can maintain a map of processing delays per output port (which corresponds to specific VNF instances for each service) as a cumulative average of the delay calculated every time a data packet is received. In other embodiments, the map of processing delays per output port can be based on a weighted moving average, exponential moving average, or similar weighting. The type of averaging employed can be selected to give more or less weight to recent data points.

[0067] Once the data analyzer 413 updates its map of the processing delays it can provide feedback or configuration information to the load balancer 409, which in turn may modify the operation of the Select Group algorithm for data packet forwarding within the flow control pipeline 411.

[0068] The load balancer 409 can be any algorithm for managing and configuring a flow control pipeline 411 to implement a distribution of data packets to a set of services and/or service instances (e.g., VNF instances). The load balancer 409 can receive an initial configuration from a controller and subsequently update the load balancing dynamically and in real-time based on delay information provided by the delay analyzer 413.

[0069] Figure 6 is a diagram of one embodiment of the interaction between the DPN components and the northbound interface in the SDN architecture. These components interact with the establishment of the delay analysis. The northbound interface receives instructions from a user or application to set up the test delay analysis. In response to these instructions, the controller sets up flow control group entries in the selective time stamper and marker for known services (e.g., for VNF instances of a service), so that a fraction of traffic towards the service will be timestamped and marked. Depending on the number of entries in a SELECT group, the controller is able to configure the fraction of traffic that needs to be timestamped (e.g., if the controller is seeking to select twenty percent of the data traffic for time stamping, then one entry out of five is modified to insert time stamps in the data traffic for a VNF). The controller sets up flow table entries in the response traffic handler to punt (forward) identified marked data traffic coming from known service instances (e.g., VNF instances) to the controller. As discussed above, the table entries of the select time stamper and marker as well as the response handler can be discrete configuration information for discrete functions or can be implemented by configuring tables within the flow control pipeline of the DPN.

[0070] Once configured, delay analysis can begin with the return of the timestamped response traffic from traversed services (e.g., VNF instances). Upon receipt of response data traffic from a service, the DPN checks whether the data traffic message is matched against the response data traffic handler entries (i.e., it matches a marker identifier for data traffic sent to the service and timestamped). Where a match is found the response traffic handler timestamps the inbound data packet and sends the data packets to the delay analyzer in a PACKET_IN message. The delay analyzer then extracts the input port from the PACKET_IN message to determine the service. The delay analyzer extracts the ingress and egress timestamps from the NSH or similar header. Using the ingress and egress timestamp, the data analyzer calculates the delay incurred by the service (e.g., a VNF instance) for processing the associated data packet and/or dummy NSH message. The delay analyzer builds a map of per port (per service instance/VNF) delay.

[0071] The map can be provided to the load balancing algorithm of the DPN to rebalance the weighting of the load balancing. In one embodiment, the Select Group Algorithm is utilized. Whenever actual (non-testing) data traffic reaches the SELECT group algorithm in the flow control pipeline to load balance the data traffic across service instances/VNFs, the DPN can query the per port delay map generated by the delay analyzer to influence the bucket selection. The delay map can be used to re-weight service instance/VNF selection inverse to the current delay (i.e., the current load). Where the delay exceeds a defined threshold or where the delay indicates no data traffic responses were received, then the service instance/VNF can be categorized as unavailable and removed from the selection process until delay information improves to show availability.

[0072] Thus, the embodiments provide a process to improve dynamic load balancing within an SDN that is implemented at the DPN such that it is autonomous from the controller and responsive to current load and network congestion between the DPN and the service instance/VNF in particular for use with load balancing for service chain functions. The embodiments provide for the dynamic load balancing at the DPN without creating excessive overhead by selecting a fraction of the data traffic being forwarded towards service instances/VNF to be timestamped by the DPN (egress timestamp), before sending it out towards the service instance/VNF. These timestamped packets are marked using a unique mechanism so that they can be identified upon return to the DPN. The marked incoming messages are again timestamped (ingress timestamp) at the DPN, before being sent for delay analysis. The delay analyzer module is within the DPN and checks the total time taken to process the NSH messages, by the various load balanced service instances/VNFs being utilized by the DPN. The delay information generated by the delay analyzer is used to influence the select group algorithm for bucket selection (i.e., load distribution between the available service instances/VNFs.

[0073] Figure 7A illustrates connectivity between network devices (NDs) within an exemplary network, as well as three exemplary implementations of the NDs, according to some embodiments of the invention. Figure 7A shows NDs 700A-H, and their connectivity by way of lines between 700A-700B, 700B-700C, 700C-700D, 700D-700E, 700E-700F, 700F-700G, and 700A-700G, as well as between 700H and each of 700A, 700C, 700D, and 700G. These NDs are physical devices, and the connectivity between these NDs can be wireless or wired (often referred to as a link). An additional line extending from NDs 700A, 700E, and 700F illustrates that these NDs act as ingress and egress points for the network (and thus, these NDs are sometimes referred to as edge NDs; while the other NDs may be called core NDs).

[0074] Two of the exemplary ND implementations in Figure 7 A are: 1) a special-purpose network device 702 that uses custom application-specific integrated-circuits (ASICs) and a special-purpose operating system (OS); and 2) a general-purpose network device 704 that uses common off-the-shelf (COTS) processors and a standard OS.

[0075] The special-purpose network device 702 includes networking hardware 710 comprising a set of one or more processor(s) 712, forwarding resource(s) 714 (which typically include one or more ASICs and/or network processors), and physical network interfaces (NIs) 716 (through which network connections are made, such as those shown by the connectivity between NDs 700A-H), as well as non-transitory machine readable storage media 718 having stored therein networking software 720. During operation, the networking software 720 may be executed by the networking hardware 710 to instantiate a set of one or more networking software instance(s) 722. Each of the networking software instance(s) 722, and that part of the networking hardware 710 that executes that network software instance (be it hardware dedicated to that networking software instance and/or time slices of hardware temporally shared by that networking software instance with others of the networking software instance(s) 722), form a separate virtual network element 730A-R. Each of the virtual network element(s) (VNEs) 730A-R includes a control communication and configuration module 732A-R (sometimes referred to as a local control module or control communication module) and forwarding table(s) 734A-R, such that a given virtual network element (e.g., 730A) includes the control communication and configuration module (e.g., 732A), a set of one or more forwarding table(s) (e.g., 734A), and that portion of the networking hardware 710 that executes the virtual network element (e.g., 730A).

[0076] The special-purpose network device 702 is often physically and/or logically considered to include: 1) a ND control plane 724 (sometimes referred to as a control plane) comprising the processor(s) 712 that execute the control communication and configuration module(s) 732A-R; and 2) a ND forwarding plane 726 (sometimes referred to as a forwarding plane, a data plane, or a media plane) comprising the forwarding resource(s) 714 that utilize the forwarding table(s) 734A-R and the physical NIs 716. By way of example, where the ND is a router (or is implementing routing functionality), the ND control plane 724 (the processor(s) 712 executing the control communication and configuration module(s) 732A-R) is typically responsible for participating in controlling how data (e.g., packets) is to be routed (e.g., the next hop for the data and the outgoing physical NI for that data) and storing that routing information in the forwarding table(s) 734A-R, and the ND forwarding plane 726 is responsible for receiving that data on the physical NIs 716 and forwarding that data out the appropriate ones of the physical NIs 716 based on the forwarding table(s) 734A- R.

[0077] Figure 7B illustrates an exemplary way to implement the special-purpose network device 702 according to some embodiments of the invention. Figure 7B shows a special- purpose network device including cards 738 (typically hot pluggable). While in some embodiments the cards 738 are of two types (one or more that operate as the ND forwarding plane 726 (sometimes called line cards), and one or more that operate to implement the ND control plane 724 (sometimes called control cards)), alternative embodiments may combine functionality onto a single card and/or include additional card types (e.g., one additional type of card is called a service card, resource card, or multi-application card). A service card can provide specialized processing (e.g., Layer 4 to Layer 7 services (e.g., firewall, Internet Protocol Security (IPsec), Secure Sockets Layer (SSL) / Transport Layer Security (TLS), Intrusion Detection System (IDS), peer-to-peer (P2P), Voice over IP (VoIP) Session Border Controller, Mobile Wireless Gateways (Gateway General Packet Radio Service (GPRS) Support Node (GGSN), Evolved Packet Core (EPC) Gateway)). By way of example, a service card may be used to terminate IPsec tunnels and execute the attendant authentication and encryption algorithms. These cards are coupled together through one or more interconnect mechanisms illustrated as backplane 736 (e.g., a first full mesh coupling the line cards and a second full mesh coupling all of the cards).

[0078] Returning to Figure 7A, the general purpose network device 704 includes hardware 740 comprising a set of one or more processor(s) 742 (which are often COTS processors) and physical NIs 746, as well as non-transitory machine readable storage media 748 having stored therein software 750. During operation, the processor(s) 742 execute the software 750 to instantiate one or more sets of one or more applications 764A-R. While one embodiment does not implement virtualization, alternative embodiments may use different forms of virtualization. For example, in one such alternative embodiment the virtualization layer 754 represents the kernel of an operating system (or a shim executing on a base operating system) that allows for the creation of multiple instances 762A-R called software containers that may each be used to execute one (or more) of the sets of applications 764A-R; where the multiple software containers (also called virtualization engines, virtual private servers, or jails) are user spaces (typically a virtual memory space) that are separate from each other and separate from the kernel space in which the operating system is run; and where the set of applications running in a given user space, unless explicitly allowed, cannot access the memory of the other processes. In another such alternative embodiment the virtualization layer 754 represents a hypervisor (sometimes referred to as a virtual machine monitor (VMM)) or a hypervisor executing on top of a host operating system, and each of the sets of applications 764A-R is run on top of a guest operating system within an instance 762A-R called a virtual machine (which may in some cases be considered a tightly isolated form of software container) that is run on top of the hypervisor - the guest operating system and application may not know they are running on a virtual machine as opposed to running on a "bare metal" host electronic device, or through para- virtualization the operating system and/or application may be aware of the presence of virtualization for optimization purposes. In yet other alternative embodiments, one, some or all of the applications are implemented as unikernel(s), which can be generated by compiling directly with an application only a limited set of libraries (e.g., from a library operating system (LibOS) including drivers/libraries of OS services) that provide the particular OS services needed by the application. As a unikernel can be implemented to run directly on hardware 740, directly on a hypervisor (in which case the unikernel is sometimes described as running within a LibOS virtual machine), or in a software container, embodiments can be implemented fully with unikernels running directly on a hypervisor represented by virtualization layer 754, unikernels running within software containers represented by instances 762A-R, or as a combination of unikernels and the above-described techniques (e.g., unikernels and virtual machines both run directly on a hypervisor, unikernels and sets of applications that are run in different software containers).

[0079] The instantiation of the one or more sets of one or more applications 764A-R, as well as virtualization if implemented, are collectively referred to as software instance(s) 752. Each set of applications 764 A-R, corresponding virtualization construct (e.g., instance 762A-R) if implemented, and that part of the hardware 740 that executes them (be it hardware dedicated to that execution and/or time slices of hardware temporally shared), forms a separate virtual network element(s) 760A-R. The Applications 764A-R can include any of the components of the dynamic load balancing process including the data analyzer 764A-R. These components can be implemented as applications local to the special purpose network device 702 or as functions in VNFs in the COTS network device 704 or in similar permutations thereof.

[0080] The virtual network element(s) 760A-R perform similar functionality to the virtual network element(s) 730A-R - e.g., similar to the control communication and configuration module(s) 732A and forwarding table(s) 734A (this virtualization of the hardware 740 is sometimes referred to as network function virtualization (NFV)). Thus, NFV may be used to consolidate many network equipment types onto industry standard high volume server hardware, physical switches, and physical storage, which could be located in Data centers, NDs, and customer premise equipment (CPE). While embodiments of the invention are illustrated with each instance 762A-R corresponding to one VNE 760A-R, alternative embodiments may implement this correspondence at a finer level granularity (e.g., line card virtual machines virtualize line cards, control card virtual machine virtualize control cards, etc.); it should be understood that the techniques described herein with reference to a correspondence of instances 762A-R to VNEs also apply to embodiments where such a finer level of granularity and/or unikernels are used.

[0081] In certain embodiments, the virtualization layer 754 includes a virtual switch that provides similar forwarding services as a physical Ethernet switch. Specifically, this virtual switch forwards traffic between instances 762A-R and the physical NI(s) 746, as well as optionally between the instances 762A-R; in addition, this virtual switch may enforce network isolation between the VNEs 760A-R that by policy are not permitted to communicate with each other (e.g., by honoring virtual local area networks (VLANs)).

[0082] The third exemplary ND implementation in Figure 7A is a hybrid network device 706, which includes both custom ASICs/special-purpose OS and COTS processors/standard OS in a single ND or a single card within an ND. In certain embodiments of such a hybrid network device, a platform VM (i.e., a VM that that implements the functionality of the special-purpose network device 702) could provide for para- virtualization to the networking hardware present in the hybrid network device 706.

[0083] Regardless of the above exemplary implementations of an ND, when a single one of multiple VNEs implemented by an ND is being considered (e.g., only one of the VNEs is part of a given virtual network) or where only a single VNE is currently being implemented by an ND, the shortened term network element (NE) is sometimes used to refer to that VNE. Also in all of the above exemplary implementations, each of the VNEs (e.g., VNE(s) 730A- R, VNEs 760A-R, and those in the hybrid network device 706) receives data on the physical NIs (e.g., 716, 746) and forwards that data out the appropriate ones of the physical NIs (e.g., 716, 746). For example, a VNE implementing IP router functionality forwards IP packets on the basis of some of the IP header information in the IP packet; where IP header information includes source IP address, destination IP address, source port, destination port (where "source port" and "destination port" refer herein to protocol ports, as opposed to physical ports of a ND), transport protocol (e.g., user datagram protocol (UDP), Transmission Control Protocol (TCP), and differentiated services code point (DSCP) values.

[0084] Figure 7C illustrates various exemplary ways in which VNEs may be coupled according to some embodiments of the invention. Figure 7C shows VNEs 770A.1-770A.P (and optionally VNEs 770A.Q-770A.R) implemented in ND 700A and VNE 770H.1 in ND 700H. In Figure 7C, VNEs 770A.1-P are separate from each other in the sense that they can receive packets from outside ND 700A and forward packets outside of ND 700A; VNE 770A.1 is coupled with VNE 770H.1, and thus they communicate packets between their respective NDs; VNE 770A.2-770A.3 may optionally forward packets between themselves without forwarding them outside of the ND 700A; and VNE 770A.P may optionally be the first in a chain of VNEs that includes VNE 770A.Q followed by VNE 770A.R (this is sometimes referred to as dynamic service chaining, where each of the VNEs in the series of VNEs provides a different service - e.g., one or more layer 4-7 network services). While Figure 7C illustrates various exemplary relationships between the VNEs, alternative embodiments may support other relationships (e.g., more/fewer VNEs, more/fewer dynamic service chains, multiple different dynamic service chains with some common VNEs and some different VNEs).

[0085] The NDs of Figure 7A, for example, may form part of the Internet or a private network; and other electronic devices (not shown; such as end user devices including workstations, laptops, netbooks, tablets, palm tops, mobile phones, smartphones, phablets, multimedia phones, Voice Over Internet Protocol (VOIP) phones, terminals, portable media players, GPS units, wearable devices, gaming systems, set-top boxes, Internet enabled household appliances) may be coupled to the network (directly or through other networks such as access networks) to communicate over the network (e.g., the Internet or virtual private networks (VPNs) overlaid on (e.g., tunneled through) the Internet) with each other (directly or through servers) and/or access content and/or services. Such content and/or services are typically provided by one or more servers (not shown) belonging to a service/content provider or one or more end user devices (not shown) participating in a peer- to-peer (P2P) service, and may include, for example, public webpages (e.g., free content, store fronts, search services), private webpages (e.g., username/password accessed webpages providing email services), and/or corporate networks over VPNs. For instance, end user devices may be coupled (e.g., through customer premise equipment coupled to an access network (wired or wirelessly)) to edge NDs, which are coupled (e.g., through one or more core NDs) to other edge NDs, which are coupled to electronic devices acting as servers. However, through compute and storage virtualization, one or more of the electronic devices operating as the NDs in Figure 7A may also host one or more such servers (e.g., in the case of the general purpose network device 704, one or more of the software instances 762A-R may operate as servers; the same would be true for the hybrid network device 706; in the case of the special-purpose network device 702, one or more such servers could also be run on a virtualization layer executed by the processor(s) 712); in which case the servers are said to be co-located with the VNEs of that ND.

[0086] A virtual network is a logical abstraction of a physical network (such as that in Figure 7A) that provides network services (e.g., L2 and/or L3 services). A virtual network can be implemented as an overlay network (sometimes referred to as a network virtualization overlay) that provides network services (e.g., layer 2 (L2, data link layer) and/or layer 3 (L3, network layer) services) over an underlay network (e.g., an L3 network, such as an Internet Protocol (IP) network that uses tunnels (e.g., generic routing encapsulation (GRE), layer 2 tunneling protocol (L2TP), IPSec) to create the overlay network).

[0087] A network virtualization edge (NVE) sits at the edge of the underlay network and participates in implementing the network virtualization; the network-facing side of the NVE uses the underlay network to tunnel frames to and from other NVEs; the outward-facing side of the NVE sends and receives data to and from systems outside the network. A virtual network instance (VNI) is a specific instance of a virtual network on a NVE (e.g., a NE/VNE on an ND, a part of a NE/VNE on a ND where that NE/VNE is divided into multiple VNEs through emulation); one or more VNIs can be instantiated on an NVE (e.g., as different VNEs on an ND). A virtual access point (VAP) is a logical connection point on the NVE for connecting external systems to a virtual network; a VAP can be physical or virtual ports identified through logical interface identifiers (e.g., a VLAN ID).

[0088] Examples of network services include: 1) an Ethernet LAN emulation service (an Ethernet-based multipoint service similar to an Internet Engineering Task Force (IETF) Multiprotocol Label Switching (MPLS) or Ethernet VPN (EVPN) service) in which external systems are interconnected across the network by a LAN environment over the underlay network (e.g., an NVE provides separate L2 VNIs (virtual switching instances) for different such virtual networks, and L3 (e.g., IP/MPLS) tunneling encapsulation across the underlay network); and 2) a virtualized IP forwarding service (similar to IETF IP VPN (e.g., Border Gateway Protocol (BGP)/MPLS IP VPN) from a service definition perspective) in which external systems are interconnected across the network by an L3 environment over the underlay network (e.g., an NVE provides separate L3 VNIs (forwarding and routing instances) for different such virtual networks, and L3 (e.g., IP/MPLS) tunneling encapsulation across the underlay network)). Network services may also include quality of service capabilities (e.g., traffic classification marking, traffic conditioning and scheduling), security capabilities (e.g., filters to protect customer premises from network - originated attacks, to avoid malformed route announcements), and management capabilities (e.g., full detection and processing).

[0089] Fig. 7D illustrates a network with a single network element on each of the NDs of Figure 7 A, and within this straight forward approach contrasts a traditional distributed approach (commonly used by traditional routers) with a centralized approach for maintaining reachability and forwarding information (also called network control), according to some embodiments of the invention. Specifically, Figure 7D illustrates network elements (NEs) 770A-H with the same connectivity as the NDs 700A-H of Figure 7A.

[0090] Figure 7D illustrates that the distributed approach 772 distributes responsibility for generating the reachability and forwarding information across the NEs 770A-H; in other words, the process of neighbor discovery and topology discovery is distributed.

[0091] For example, where the special-purpose network device 702 is used, the control communication and configuration module(s) 732A-R of the ND control plane 724 typically include a reachability and forwarding information module to implement one or more routing protocols (e.g., an exterior gateway protocol such as Border Gateway Protocol (BGP), Interior Gateway Protocol(s) (IGP) (e.g., Open Shortest Path First (OSPF), Intermediate System to Intermediate System (IS-IS), Routing Information Protocol (RIP), Label Distribution Protocol (LDP), Resource Reservation Protocol (RSVP) (including RSVP- Traffic Engineering (TE): Extensions to RSVP for LSP Tunnels and Generalized Multi- Protocol Label Switching (GMPLS) Signaling RSVP-TE)) that communicate with other NEs to exchange routes, and then selects those routes based on one or more routing metrics. Thus, the NEs 770A-H (e.g., the processor(s) 712 executing the control communication and configuration module(s) 732A-R) perform their responsibility for participating in controlling how data (e.g., packets) is to be routed (e.g., the next hop for the data and the outgoing physical NI for that data) by distributively determining the reachability within the network and calculating their respective forwarding information. Routes and adjacencies are stored in one or more routing structures (e.g., Routing Information Base (RIB), Label Information Base (LIB), one or more adjacency structures) on the ND control plane 724. The ND control plane 724 programs the ND forwarding plane 726 with information (e.g., adjacency and route information) based on the routing structure(s). For example, the ND control plane 724 programs the adjacency and route information into one or more forwarding table(s) 734A-R (e.g., Forwarding Information Base (FIB), Label Forwarding Information Base (LFIB), and one or more adjacency structures) on the ND forwarding plane 726. For layer 2 forwarding, the ND can store one or more bridging tables that are used to forward data based on the layer 2 information in that data. While the above example uses the special-purpose network device 702, the same distributed approach 772 can be implemented on the general-purpose network device 704 and the hybrid network device 706.

[0092] Figure 7D illustrates that a centralized approach 774 (also known as software defined networking (SDN)) that decouples the system that makes decisions about where traffic is sent from the underlying systems that forwards traffic to the selected destination. The illustrated centralized approach 774 has the responsibility for the generation of reachability and forwarding information in a centralized control plane 776 (sometimes referred to as a SDN control module, controller, network controller, OpenFlow controller, SDN controller, control plane node, network virtualization authority, or management control entity), and thus the process of neighbor discovery and topology discovery is centralized. The centralized control plane 776 has a south bound interface 782 with a data plane 780 (sometime referred to the infrastructure layer, network forwarding plane, or forwarding plane (which should not be confused with a ND forwarding plane)) that includes the NEs 770A-H (sometimes referred to as switches, forwarding elements, data plane elements, or nodes). The centralized control plane 776 includes a network controller 778, which includes a centralized reachability and forwarding information module 779 that determines the reachability within the network and distributes the forwarding information to the NEs 770A- H of the data plane 780 over the south bound interface 782 (which may use the OpenFlow protocol). Thus, the network intelligence is centralized in the centralized control plane 776 executing on electronic devices that are typically separate from the NDs.

[0093] For example, where the special-purpose network device 702 is used in the data plane 780, each of the control communication and configuration module(s) 732A-R of the ND control plane 724 typically include a control agent that provides the VNE side of the south bound interface 782. In this case, the ND control plane 724 (the processor(s) 712 executing the control communication and configuration module(s) 732A-R) performs its responsibility for participating in controlling how data (e.g., packets) is to be routed (e.g., the next hop for the data and the outgoing physical NI for that data) through the control agent communicating with the centralized control plane 776 to receive the forwarding information (and in some cases, the reachability information) from the centralized reachability and forwarding information module 779 (it should be understood that in some embodiments of the invention, the control communication and configuration module(s) 732A-R, in addition to communicating with the centralized control plane 776, may also play some role in determining reachability and/or calculating forwarding information - albeit less so than in the case of a distributed approach; such embodiments are generally considered to fall under the centralized approach 774, but may also be considered a hybrid approach).

[0094] While the above example uses the special-purpose network device 702, the same centralized approach 774 can be implemented with the general purpose network device 704 (e.g., each of the VNE 760A-R performs its responsibility for controlling how data (e.g., packets) is to be routed (e.g., the next hop for the data and the outgoing physical NI for that data) by communicating with the centralized control plane 776 to receive the forwarding information (and in some cases, the reachability information) from the centralized reachability and forwarding information module 779; it should be understood that in some embodiments of the invention, the VNEs 760A-R, in addition to communicating with the centralized control plane 776, may also play some role in determining reachability and/or calculating forwarding information - albeit less so than in the case of a distributed approach) and the hybrid network device 706. In fact, the use of SDN techniques can enhance the NFV techniques typically used in the general-purpose network device 704 or hybrid network device 706 implementations as NFV is able to support SDN by providing an infrastructure upon which the SDN software can be run, and NFV and SDN both aim to make use of commodity server hardware and physical switches.

[0095] Figure 7D also shows that the centralized control plane 776 has a north bound interface 784 to an application layer 786, in which resides application(s) 788. The centralized control plane 776 has the ability to form virtual networks 792 (sometimes referred to as a logical forwarding plane, network services, or overlay networks (with the NEs 770A-H of the data plane 780 being the underlay network)) for the application(s) 788. Thus, the centralized control plane 776 maintains a global view of all NDs and configured NEs/VNEs, and it maps the virtual networks to the underlying NDs efficiently (including maintaining these mappings as the physical network changes either through hardware (ND, link, or ND component) failure, addition, or removal). The applications 788 can include any of the components of the dynamic load balancing process including the data analyzer 781. These components can be implemented at the application layer 786 at the network device or any combination or sub-combination thereof.

[0096] While Figure 7D shows the distributed approach 772 separate from the centralized approach 774, the effort of network control may be distributed differently or the two combined in certain embodiments of the invention. For example: 1) embodiments may generally use the centralized approach (SDN) 774, but have certain functions delegated to the NEs (e.g., the distributed approach may be used to implement one or more of fault monitoring, performance monitoring, protection switching, and primitives for neighbor and/or topology discovery); or 2) embodiments of the invention may perform neighbor discovery and topology discovery via both the centralized control plane and the distributed protocols, and the results compared to raise exceptions where they do not agree. Such embodiments are generally considered to fall under the centralized approach 774, but may also be considered a hybrid approach.

[0097] While Figure 7D illustrates the simple case where each of the NDs 700A-H implements a single NE 770A-H, it should be understood that the network control approaches described with reference to Figure 7D also work for networks where one or more of the NDs 700A-H implement multiple VNEs (e.g., VNEs 730A-R, VNEs 760A-R, those in the hybrid network device 706). Alternatively or in addition, the network controller 778 may also emulate the implementation of multiple VNEs in a single ND. Specifically, instead of (or in addition to) implementing multiple VNEs in a single ND, the network controller 778 may present the implementation of a VNE/NE in a single ND as multiple VNEs in the virtual networks 792 (all in the same one of the virtual network(s) 792, each in different ones of the virtual network(s) 792, or some combination). For example, the network controller 778 may cause an ND to implement a single VNE (a NE) in the underlay network, and then logically divide up the resources of that NE within the centralized control plane 776 to present different VNEs in the virtual network(s) 792 (where these different VNEs in the overlay networks are sharing the resources of the single VNE/NE implementation on the ND in the underlay network).

[0098] On the other hand, Figures 7E and 7F respectively illustrate exemplary abstractions of NEs and VNEs that the network controller 778 may present as part of different ones of the virtual networks 792. Figure 7E illustrates the simple case of where each of the NDs 700A-H implements a single NE 770A-H (see Figure 7D), but the centralized control plane 776 has abstracted multiple of the NEs in different NDs (the NEs 770A-C and G-H) into (to represent) a single NE 7701 in one of the virtual network(s) 792 of Figure 7D, according to some embodiments of the invention. Figure 7E shows that in this virtual network, the NE 7701 is coupled to NE 770D and 770F, which are both still coupled to NE 770E.

[0099] Figure 7F illustrates a case where multiple VNEs (VNE 770A.1 and VNE 770H.1) are implemented on different NDs (ND 700A and ND 700H) and are coupled to each other, and where the centralized control plane 776 has abstracted these multiple VNEs such that they appear as a single VNE 770T within one of the virtual networks 792 of Figure 7D, according to some embodiments of the invention. Thus, the abstraction of a NE or VNE can span multiple NDs.

[00100] While some embodiments of the invention implement the centralized control plane 776 as a single entity (e.g., a single instance of software running on a single electronic device), alternative embodiments may spread the functionality across multiple entities for redundancy and/or scalability purposes (e.g., multiple instances of software running on different electronic devices).

[00101] Similar to the network device implementations, the electronic device(s) running the centralized control plane 776, and thus the network controller 778 including the centralized reachability and forwarding information module 779, may be implemented a variety of ways (e.g., a special purpose device, a general -purpose (e.g., COTS) device, or hybrid device). These electronic device(s) would similarly include processor(s), a set or one or more physical NIs, and a non-transitory machine -readable storage medium having stored thereon the centralized control plane software. For instance, Figure 8 illustrates, a general- purpose control plane device 804 including hardware 840 comprising a set of one or more processor(s) 842 (which are often COTS processors) and physical NIs 846, as well as non- transitory machine readable storage media 848 having stored therein centralized control plane (CCP) software 850.

[00102] In embodiments that use compute virtualization, the processor(s) 842 typically execute software to instantiate a virtualization layer 854 (e.g., in one embodiment the virtualization layer 854 represents the kernel of an operating system (or a shim executing on a base operating system) that allows for the creation of multiple instances 862A-R called software containers (representing separate user spaces and also called virtualization engines, virtual private servers, or jails) that may each be used to execute a set of one or more applications; in another embodiment the virtualization layer 854 represents a hypervisor (sometimes referred to as a virtual machine monitor (VMM)) or a hypervisor executing on top of a host operating system, and an application is run on top of a guest operating system within an instance 862A-R called a virtual machine (which in some cases may be considered a tightly isolated form of software container) that is run by the hypervisor ; in another embodiment, an application is implemented as a unikernel, which can be generated by compiling directly with an application only a limited set of libraries (e.g., from a library operating system (LibOS) including drivers/libraries of OS services) that provide the particular OS services needed by the application, and the unikernel can run directly on hardware 840, directly on a hypervisor represented by virtualization layer 854 (in which case the unikernel is sometimes described as running within a LibOS virtual machine), or in a software container represented by one of instances 862A-R). Again, in embodiments where compute virtualization is used, during operation an instance of the CCP software 850 (illustrated as CCP instance 876A) is executed (e.g., within the instance 862A) on the virtualization layer 854. In embodiments where compute virtualization is not used, the CCP instance 876A is executed, as a unikernel or on top of a host operating system, on the "bare metal" general purpose control plane device 804. The instantiation of the CCP instance 876A, as well as the virtualization layer 854 and instances 862A-R if implemented, are collectively referred to as software instance(s) 852.

[00103] In some embodiments, the CCP instance 876A includes a network controller instance 878. The network controller instance 878 includes a centralized reachability and forwarding information module instance 879 (which is a middleware layer providing the context of the network controller 778 to the operating system and communicating with the various NEs), and an CCP application layer 880 (sometimes referred to as an application layer) over the middleware layer (providing the intelligence required for various network operations such as protocols, network situational awareness, and user - interfaces). At a more abstract level, this CCP application layer 880 within the centralized control plane 776 works with virtual network view(s) (logical view(s) of the network) and the middleware layer provides the conversion from the virtual networks to the physical view. The applications 880 can include any of the components of the dynamic load balancing process including the data analyzer 881. These components can be implemented at the CCP application layer 880 at the network device or any combination or sub -combination thereof.

[00104] The centralized control plane 776 transmits relevant messages to the data plane 780 based on CCP application layer 880 calculations and middleware layer mapping for each flow. A flow may be defined as a set of packets whose headers match a given pattern of bits; in this sense, traditional IP forwarding is also flow-based forwarding where the flows are defined by the destination IP address for example; however, in other implementations, the given pattern of bits used for a flow definition may include more fields (e.g., 10 or more) in the packet headers. Different NDs/NEs/VNEs of the data plane 780 may receive different messages, and thus different forwarding information. The data plane 780 processes these messages and programs the appropriate flow information and corresponding actions in the forwarding tables (sometime referred to as flow tables) of the appropriate NE/VNEs, and then the NEs/VNEs map incoming packets to flows represented in the forwarding tables and forward packets based on the matches in the forwarding tables.

[00105] Standards such as OpenFlow define the protocols used for the messages, as well as a model for processing the packets. The model for processing packets includes header parsing, packet classification, and making forwarding decisions. Header parsing describes how to interpret a packet based upon a well-known set of protocols. Some protocol fields are used to build a match structure (or key) that will be used in packet classification (e.g., a first key field could be a source media access control (MAC) address, and a second key field could be a destination MAC address).

[00106] Packet classification involves executing a lookup in memory to classify the packet by determining which entry (also referred to as a forwarding table entry or flow entry) in the forwarding tables best matches the packet based upon the match structure, or key, of the forwarding table entries. It is possible that many flows represented in the forwarding table entries can correspond/match to a packet; in this case the system is typically configured to determine one forwarding table entry from the many according to a defined scheme (e.g., selecting a first forwarding table entry that is matched). Forwarding table entries include both a specific set of match criteria (a set of values or wildcards, or an indication of what portions of a packet should be compared to a particular value/values/wildcards, as defined by the matching capabilities - for specific fields in the packet header, or for some other packet content), and a set of one or more actions for the data plane to take on receiving a matching packet. For example, an action may be to push a header onto the packet, for the packet using a particular port, flood the packet, or simply drop the packet. Thus, a forwarding table entry for IPv4/IPv6 packets with a particular transmission control protocol (TCP) destination port could contain an action specifying that these packets should be dropped. [00107] Making forwarding decisions and performing actions occurs, based upon the forwarding table entry identified during packet classification, by executing the set of actions identified in the matched forwarding table entry on the packet.

[00108] However, when an unknown packet (for example, a "missed packet" or a "match- miss" as used in OpenFlow parlance) arrives at the data plane 780, the packet (or a subset of the packet header and content) is typically forwarded to the centralized control plane 776. The centralized control plane 776 will then program forwarding table entries into the data plane 780 to accommodate packets belonging to the flow of the unknown packet. Once a specific forwarding table entry has been programmed into the data plane 780 by the centralized control plane 776, the next packet with matching credentials will match that forwarding table entry and take the set of actions associated with that matched entry.

[00109] A network interface (NI) may be physical or virtual; and in the context of IP, an interface address is an IP address assigned to a NI, be it a physical NI or virtual NI. A virtual NI may be associated with a physical NI, with another virtual interface, or stand on its own (e.g., a loopback interface, a point-to-point protocol interface). A NI (physical or virtual) may be numbered (a NI with an IP address) or unnumbered (a NI without an IP address). A loopback interface (and its loopback address) is a specific type of virtual NI (and IP address) of a NE/VNE (physical or virtual) often used for management purposes; where such an IP address is referred to as the nodal loopback address. The IP address(es) assigned to the NI(s) of a ND are referred to as IP addresses of that ND; at a more granular level, the IP address(es) assigned to NI(s) assigned to a NE/VNE implemented on a ND can be referred to as IP addresses of that NE/VNE.

[00110] While the invention has been described in terms of several embodiments, those skilled in the art will recognize that the invention is not limited to the embodiments described, can be practiced with modification and alteration within the spirit and scope of the appended claims. The description is thus to be regarded as illustrative instead of limiting.

Claims

CLAIMS What is claimed is:

1. A method implemented by a network device, the method for autonomous dynamic load balancing at the network device functioning as a data plane node (DPN) in a software defined networking (SDN) network, the DPN forwarding data traffic to a set of load- balanced virtual network function (VNF) instances proportionate to a weighting of load balance for the set of load-balanced VNF instances, the method comprising:

analyzing (309) collected timestamps for data traffic that traversed the set of load- balanced VNF instances to determine processing delay for each VNF instance in the set of load -balanced VNF instances; and

updating (311) the weighting of load balance to reflect determined delay times for each VNF instance.

2. The method of claim 1, further comprising:

updating (313) the weighting of load balance to reflect a current status for each VNF instance.

3. The method of claim 1, wherein the analyzing the collected timestamps determines network congestion between the DPN and each VNF instance.

4. The method of claim 1, further comprising:

selecting (303) a subset of data traffic for the set of load-balanced VNF instances, where timestamping is limited to the subset of data traffic and marking identifies the subset of data traffic upon return from the set of load-balanced VNF instances.

5. The method of claim 1, further comprising:

generating a delay map with a running average delay for each VNF instance in the set of load-balanced VNF instances; and

weighting (311, 313) the load balance for each VNF instance in the set of load- balanced VNF instances inverse to the delay map.

6. A network device configured to implement a method for autonomous dynamic load balancing at the network device functioning as a data plane node (DPN) in a software defined networking (SDN) network, the DPN forwarding data traffic to a set of load- balanced virtual network function (VNF) instances proportionate to a weighting of load balance for the set of load-balanced VNF instances, the network device comprising:

a non-transitory computer readable medium (718) having stored therein a delay analyzer;

a flow control pipeline to forward the data traffic according to a load balance; and a processor (712) configured to execute the delay analyzer (764), the delay analyzer to analyze collected timestamps for data traffic that traversed the set of load- balanced VNF instances to determine processing delay for each VNF instance in the set of load-balanced VNF instances, and to update the weighting of load balance to reflect determined delay times for each VNF instance.

7. The network device of claim 6, wherein the delay analyzer is further to update the weighting of load balance to reflect a current status for each service.

8. The network device of claim 6, wherein the analyzing the collected timestamps determines network congestion between the DPN and each VNF instance.

9. The network device of claim 6, wherein the processor is further configured to execute a time stamper and marker to select a subset of data traffic for the set of load- balanced VNF instances, where timestamping is limited to the subset of data traffic and marking identifies the subset of data traffic upon return from the set of load-balanced VNF instances.

10. The network device of claim 6, wherein the delay analyzer is further configured to generate a delay map with a running average delay for each VNF instance in the set of load-balanced VNF instances, and

to weight the load balance for each VNF instance in the set of load-balanced VNF instances inverse to the delay map.

11. A computing device in communication with a network device, the computing device to execute a plurality of virtual machines for implementing network function virtualization (NFV), the computing device configured to execute a method for autonomous dynamic load balancing at the network device functioning as a data plane node (DPN) in a software defined networking (SDN) network, the DPN forwarding data traffic to a set of load- balanced virtual network function (VNF) instances proportionate to a weighting of load balance for the set of load-balanced VNF instances, the computing device comprising:

a non-transitory computer readable medium (748) having stored therein a delay analyzer; and

a processor (742) configured to execute a virtual machine from the plurality of virtual machines 762, the virtual machine to execute the delay analyzer (764), the delay analyzer to analyze collected timestamps for data traffic that traversed the set of load-balanced VNF instances to determine processing delay for each VNF instance in the set of load-balanced VNF instances, and to update the weighting of load balance to reflect determined delay times for each VNF instance.

12. The computing device of claim 11, wherein the delay analyzer is further to update the weighting of load balance to reflect a current status for each VNF instance.

13. The computing device of claim 11, wherein the analyzing the collected timestamps determines network congestion between the DPN and each VNF instance.

14. The computing device of claim 11, wherein the virtual machine is further configured to execute a time stamper and marker to select a subset of data traffic for the set of load- balanced VNF instances, where timestamping is limited to the subset of data traffic and marking identifies the subset of data traffic upon return from the set of load-balanced VNF instances.

15. The computing device of claim 11, wherein the delay analyzer is further configured to

generate a delay map with a running average delay for each VNF instance in the set of load-balanced VNF instance, and

16. A control plane device in communication with a network device, the control plane device configured to execute a method for autonomous dynamic load balancing at the network device functioning as a data plane node (DPN) in a software defined networking (SDN) network, the DPN forwarding data traffic to a set of load-balanced virtual network function (VNF) instances proportionate to a weighting of load balance for the set of load- balanced VNF instances, the control plane device comprising:

a non-transitory computer readable medium (848) having stored therein a delay analyzer (881); and

a processor (842) configured to execute the delay analyzer, the delay analyzer to analyze collected timestamps for data traffic that traversed the set of load- balanced VNF instances to determine processing delay for each VNF instance in the set of load-balanced VNF instances, and to update the weighting of load balance to reflect determined delay times for each VNF instance.

17. The control plane device of claim 16, wherein the delay analyzer is further to update the weighting of load balance to reflect a current status for each VNF instance.

18. The control plane device of claim 16, wherein the analyzing the collected timestamps determines network congestion between the DPN and each VNF instance.

19. The control plane device of claim 16, wherein the processor is further configured to execute a time stamper and marker to select a subset of data traffic for the set of load- balanced VNF instances, where timestamping is limited to the subset of data traffic and marking identifies the subset of data traffic upon return from the set of load-balanced VNF instances.

20. The control plane device of claim 16, wherein the delay analyzer is further configured to

generate a delay map with a running average delay for each VNF instance in the set of load-balanced VNF instances, and