CN118715766A - System and method for dynamically selecting a load balancing algorithm - Google Patents

Abstract

In one embodiment, a method includes: traffic is received and one or more attributes associated with the traffic are identified. The method further comprises the steps of: the load balancing algorithm is dynamically selected based on one or more attributes according to a load balancing scheme. The method further comprises the steps of: load balancing is performed on the traffic according to a load balancing algorithm, and traffic is transferred from the first network element to the second network element according to the load balancing.

Description

System and method for dynamically selecting a load balancing algorithm

CROSS-REFERENCE TO RELATED APPLICATIONS

This application claims the benefit of U.S. Provisional Patent Application No. 63/332,840, filed on April 20, 2022 by Arul Murugan Manickam et al., entitled “DYNAMIC LOAD BALANCING ALGORITHM SELECTION BASED ON APPLICATION OR POLICY,” which is incorporated herein by reference as if reproduced in its entirety.

Technical Field

The present disclosure relates generally to communication networks, and more particularly to systems and methods for dynamically selecting a load balancing algorithm.

Background Art

In a typical router, load balancing is automatically activated if the routing table has multiple paths to the destination. A typical enterprise edge router may provide one or more different types of load balancing algorithms. Currently, for global or per-virtual private network (VPN) configuration, a user may specify a single load balancing algorithm that is applied to all different applications, even though different applications may require different load balancing algorithms for efficiency and correctness.

BRIEF DESCRIPTION OF THE DRAWINGS

FIG. 1 illustrates a load balancing system according to some embodiments.

FIG. 2 illustrates another load balancing system according to some embodiments.

FIG. 3 illustrates a load balancing system for dynamically selecting a load balancing algorithm according to some embodiments.

FIG. 4 illustrates a method for dynamically selecting a load balancing algorithm according to some embodiments.

FIG. 5 illustrates a computer system that may be used by the systems and methods described herein, according to some embodiments.

DETAILED DESCRIPTION

Overview

Various aspects of the invention are set out in the independent claims and preferred features are set out in the dependent claims. Features of one aspect may be applicable to each aspect alone or in combination with the other aspects.

According to an embodiment, a first network element includes one or more processors and one or more computer-readable non-transitory storage media, the one or more computer-readable non-transitory storage media being coupled to the one or more processors and including instructions that, when executed by the one or more processors, cause the first network element to perform operations. The operations include: receiving traffic and identifying one or more attributes associated with the traffic. The operations also include: dynamically selecting a load balancing algorithm based on the one or more attributes according to a load balancing scheme. The operations also include: performing load balancing on the traffic according to the load balancing algorithm, and transmitting the traffic to a second network element according to the load balancing.

In some embodiments, the operations include dynamically selecting a load balancing algorithm from the following load balancing algorithms: a source IP address only load balancing algorithm, an L3 load balancing algorithm, an L4 load balancing algorithm, and a per-packet load (PPL) balancing algorithm.

In some embodiments, one or more attributes associated with the traffic include an application type. The application type may be one of: a file transfer protocol (FTP) application, a software as a service (SaaS) application, an elephant flow application, and a mouse flow application. In some embodiments, the load balancing scheme is an application-based load balancing scheme that matches the application type with a load balancing algorithm.

In some embodiments, the load balancing scheme is a user-defined application-to-algorithm mapping load balancing scheme that matches application types with user-defined load balancing algorithms.

In some embodiments, the load balancing scheme is a policy-based load balancing scheme that uses one or more attributes and policies to select a load balancing algorithm. In some embodiments, the one or more attributes include one or more of the following attributes: application type, source Internet Protocol (IP) address, destination IP address, subnet, IP differentiated service code point (DSCP), and protocol type.

According to another embodiment, a method includes receiving traffic and identifying one or more attributes associated with the traffic. The method also includes dynamically selecting a load balancing algorithm based on the one or more attributes according to a load balancing scheme. The method also includes performing load balancing on the traffic according to the load balancing algorithm, and transmitting the traffic from a first network element to a second network element according to the load balancing.

According to yet another embodiment, one or more computer-readable non-transitory storage media may include instructions that, when executed by a processor, cause the processor to perform operations. The operations include: receiving traffic and identifying one or more attributes associated with the traffic. The operations also include: dynamically selecting a load balancing algorithm based on the one or more attributes according to a load balancing scheme. The operations also include: performing load balancing on the traffic according to the load balancing algorithm, and transmitting the traffic from a first network element to a second network element according to the load balancing.

The technical advantages of certain embodiments of the present disclosure may include one or more of the following technical advantages. Certain embodiments of the present disclosure include network elements (e.g., edge routers) that dynamically select a load balancing solution based on application/policy requirements without any user intervention. Certain embodiments of the present disclosure provide a dynamic load balancing algorithm selection based on an application or policy that goes beyond the user-configured default load balancing algorithm. This method can meet the needs of recent cloud networking and SaaS applications, and improve network paths, bandwidth, and overall efficiency. Embodiments of the present disclosure can be applied to any application that requires a special hash.

Other technical advantages will be apparent to those skilled in the art from the following drawings, descriptions and claims.In addition, although specific advantages have been listed above, various embodiments may include all, some or none of the listed advantages.

Example Embodiments

The present disclosure describes systems and methods for dynamically selecting a load balancing algorithm. A typical enterprise edge router may provide one or more of the following different types of load balancing algorithms: (1) source IP address only load balancing algorithm, (2) L3 load balancing algorithm; (3) L4 load balancing algorithm; and (4) PPL balancing algorithm. Currently, for global or per-VPN configuration, a user may specify a single load balancing algorithm that is applied to all different applications. This may not be efficient because different types of applications require different load balancing for efficiency and correctness.

FIG. 1 shows a load balancing system 100 according to some embodiments. System 100 or a portion thereof may be associated with an entity, which may include any entity that performs load balancing, such as a business, company, or enterprise. In some embodiments, the entity may be a service provider that performs load balancing. The components of system 100 may include any appropriate combination of hardware, firmware, and software. For example, the components of system 100 may use one or more elements of the computer system of FIG. 5 .

In the embodiment shown in Figure 1, the system 100 includes a network 110, the Internet 112, edge routers 120 (edge router 120a and edge router 120b), a load balancer 122, a load balancing algorithm 124 (source IP only load balancing algorithm 124a, L3 load balancing algorithm 124b, L4 load balancing algorithm 124c, and PPL balancing algorithm 124d), a branch 130, a host 132, an application 134 (FTP application 134a, SaaS application 134b, elephant flow application 134c, and mouse flow application 134d), a data center 140, a server 142, an Internet circuit 150 (Internet gold circuit 150a and Internet silver circuit 150b), and a file transfer protocol (FTP) connection 160 (FTP control connection 160a and FTP data connection 160b).

The network 110 of the system 100 is any type of network that facilitates communication between components of the system 100. The network 110 can connect one or more components of the system 100. One or more parts of the network 110 can include an ad hoc network, the Internet, an intranet, an extranet, a VPN, an Ethernet VPN (EVPN), a local area network (LAN), a wireless LAN (WLAN), a virtual LAN (VLAN), a wide area network (WAN), a wireless WAN (WWAN), an SD-WAN, a metropolitan area network (MAN), a portion of a public switched telephone network (PSTN), a cellular telephone network, a digital subscriber line (DSL), a multi-protocol label switching (MPLS) network, a 3G/4G/5G network, a long-term evolution (LTE) network, a cloud network, a combination of two or more of these networks, or other suitable types of networks. The network 110 can include one or more different types of networks. In the embodiment shown in FIG. 1 , the network 110 includes the Internet 112. The Internet 112 is a global system of interconnected computer networks that use the Internet protocol group (Transmission Control Protocol (TCP)/IP) to communicate between network components.

The network 110 can be any communication network, such as a private network, a public network, a connection through the Internet 112, a mobile network, a WI-FI network, etc. The network 110 can include a core network, an access network of a service provider, an Internet service provider (ISP) network, etc. The access network is a part of the network that provides access to services to users. The core network is a part of the network 110 that acts like a backbone network for connecting different parts of the access network. One or more components of the system 100 can communicate through the network 110. In the embodiment shown in Figure 1, the network 110 is a software defined wide area network (SD-WAN).

The network 110 of the system 100 may include one or more nodes. A node is a connection point within the network 110 that receives, creates, stores, and/or sends data along a path. A node may include one or more redistribution points that identify, process, and forward data to other nodes of the network 110. A node may include virtual and/or physical nodes. For example, a node may include one or more virtual machines, bare metal servers, and the like. As another example, a node may include a data communication device, such as a computer, a router, a server, a printer, a workstation, a switch, a bridge, a modem, a hub, and the like. In the embodiment shown in FIG. 1 , the nodes of the network 110 include an edge router 120, a host 132, and a server 142.

The edge routers 120 (edge router 120a and edge router 120b) are specialized routers residing at the edge of the network 110. In certain embodiments, the edge routers 120 use static and/or dynamic routing to send data to and/or receive data from one or more networks 110 of the system 100. The edge routers 120 may include one or more hardware devices, one or more servers including routing software, etc. In certain embodiments, the edge router 120a of the system 100 is a branch edge router that sends data to and/or receives data from a host 132 of a branch 130. In the embodiment shown in FIG. 1 , the edge router 120a transmits data to and/or receives data from the edge router 120b via the Internet 112, and the edge router 120b transmits data to and/or receives data from the data center 140.

In some embodiments, edge router 120a receives one or more packets from host 132. The header of each packet may include the appropriate protocol, originating address (IP address of host 132), destination address (IP address of server 142), and packet number (e.g., packet number 1, 2, 3, 4, etc.) In some embodiments, edge router 120a identifies the destination address in the header and compares it to a lookup table to determine where to send the packet.

In some embodiments, edge router 120a includes load balancer 122. Load balancer 122 is hardware and/or software that distributes traffic across multiple nodes of network 110. In some embodiments, load balancer 122 identifies specific criteria (e.g., source IP address, destination IP address, source port, destination port, protocol, etc.) associated with incoming packets. When packets arrive, load balancer 122 can match the criteria with entries in the table. In some embodiments, load balancer 122 relies on a hash function calculated on each incoming packet. For example, load balancer 122 can use a general hash function (e.g., a cryptographic hash function) to generate a summary of source address, destination address, source port, and/or destination port. As another example, load balancer 122 can use a cyclic redundancy check (CRC) on source address, destination address, source port, and/or destination port.

In some embodiments, the load balancer 122 of the edge router 120a uses one or more load balancing algorithms 124 to distribute traffic through the Internet 112. The load balancing algorithms 124 are rules that the load balancer 122 uses to distribute traffic through the Internet 112. The load balancing algorithms 124 include the following algorithms: source-only IP load balancing algorithm 124a, L3 load balancing algorithm 124b, L4 load balancing algorithm 124c, and PPL balancing algorithm 124d.

The source IP-only load balancing algorithm 124a makes routing decisions based only on the source IP address. For example, the load balancer 122 of the edge router 120a can be configured to make load balancing decisions based only on the source IP address of the incoming traffic. Using only the source IP limits the criteria used in the hash, which makes it more likely that a specific circuit will be selected to forward traffic.

L3 load balancing algorithm 124b makes routing decisions based only on source IP address and destination IP address. In certain embodiments, L3 load balancing algorithm 124b uses multiple paths to achieve load sharing across multiple source-destination host pairs. Even if multiple paths are available, it is also possible to ensure that the packets of a given source-destination host pair adopt the same path.

The L4 load balancing algorithm 124c makes routing decisions based on the source IP address, destination IP address, source port, and destination port associated with the traffic. In some embodiments, the L4 load balancing algorithm 124cc is beneficial for traffic flows running on equal cost paths that are not load shared because most of the traffic is between peer addresses using different port numbers, such as real-time protocol (RTP) streams.

The PPL balancing algorithm 124d transmits consecutive data packets over different paths without regard to individual hosts or user sessions. For example, the PPL balancing algorithm 124d may transmit a first packet to a destination (e.g., a server 142 of a data center 140) over a first path, transmit a second packet to the same destination over a second path, and so on. In some embodiments, the PPL balancing algorithm 124d ensures that the load on all links is equal. However, because there may be differential delays within the network 110, packets may arrive at the destination out of order. The PPL balancing algorithm 124d may use a round-robin approach to determine which path each packet takes to reach the destination.

The branch 130 of the system 100 is part of an enterprise network infrastructure that provides users at geographically dispersed remote sites with access to the same network services as users in the enterprise campus. The branch 130 may include one or more buildings, offices, stores, homes, etc. In some embodiments, the branch 130 and the data center 140 are established and maintained by the same business enterprise. In the embodiment shown in FIG. 1 , the branch 130 includes a host 132.

The host 132 of the branch 130 is a device (e.g., a server or client) connected to the network 110. In some embodiments, the host 132 provides resources, data services, and/or programs to the client. The host 132 can be associated with a unique IP address, host name, etc. The IP address of the host 132 can be used as the physical address of the host 132 in the network 110. In some embodiments, the IP address is a 32-bit number. The host name is a label assigned to the host 132. The host 132 can provide processing, memory, local storage, and/or network connections that drive the application. In the embodiment shown in Figure 1, the host 132 is associated with the application 134.

Application 134 is a computer program designed to perform a specific task. Applications may include word processing software, spreadsheet software, presentation software, web browsers, multimedia software, graphics software, educational software, business application software, productivity software, etc. In some embodiments, application 134 is classified according to type (e.g., SaaS application), according to protocol (e.g., FTP traffic), according to flow size (e.g., elephant flow or mouse flow), etc. In some embodiments, application 134 may be associated with an application list. For example, a SaaS application list may include a WebEx audio application, a WebEx video application, etc. In the embodiment shown in FIG. 1 , application 134 includes FTP application 134a, SaaS application 134b, elephant flow application 134c, and mouse flow application 134d.

FTP application 134a is an application that uses FTP software. FTP software uses a client-server model that requires an FTP client (e.g., host 132) and an FTP server (e.g., server 142). FTP application 134a can be used to upload and/or download files for archiving, share files that are too large for email, share files securely between offices and/or externally, transfer data back to a disaster recovery site, etc.

SaaS applications 134b are applications managed by SaaS providers. SaaS is a delivery and licensing model in which software is accessed on the web via a subscription rather than being installed on a local computer. Typically, entities do not manage SaaS applications 134b or invest in hardware to run SaaS applications 134b. Instead, providers host and manage infrastructure to support SaaS applications 134b. In some embodiments, SaaS is a cloud-based application accessed over the Internet. SaaS applications 134b may include email tools, calendar tools, office tools (e.g., Microsoft Office 365), conference tools (Webex), etc.

The elephant flow application 134c is an application associated with the elephant flow. The elephant flow is a single session, a relatively long running network connection (e.g., a TCP connection) that consumes a large amount (total bytes) or a disproportionate amount of bandwidth. The elephant flow can be associated with virtual machine (VM) migration, backup data transmission, video, data files, etc.

The mouse flow application 134d is an application associated with a mouse flow. A mouse flow is a short (total byte) flow established by a protocol (e.g., TCP) flow measured on a network link. In some embodiments, a mouse flow has less than "n" packets, while an elephant flow has at least "n" packets, where the constant "n" is a degree of freedom in the analysis. A mouse flow may include queries, responses, control messages, etc. exchanged between application hosts. In TCP applications, mouse flows do not tolerate packet drops well. Because the number of packets in a mouse flow is small, most mouse flows are completed in the TCP slow start phase with a small TCP window size. If there are not enough packets in transit to trigger TCP fast retransmission of lost packets, the edge router 120a may not detect packet loss until the TCP retransmission timer expires. Retransmission timeouts may be detrimental to mouse flow performance and may severely degrade overall application performance. Therefore, packet losses in mouse flows must be avoided.

Different applications 134 are more suitable for different load balancing algorithms 124. For example, the mouse flow application 134d may be best suited for the L3 load balancing algorithm 124b to reduce and/or avoid packet loss. As another example, the FTP application 134a may be suitable for the source IP only load balancing algorithm 124a to avoid packet drops at the server 142. As another example, the SaaS application 134b opens multiple sub-flows. If multiple Internet circuits are involved, each sub-flow may receive a different public IP address. This may cause the flow to be rejected by the SaaS application. In this case, the source IP address only load balancing algorithm 124a may be required to avoid packet drops at the server 142. As another example, the elephant flow application 134c may be best suited for the PPL balancing algorithm 124d to effectively utilize bandwidth by distributing packets among multiple available circuits and avoid pinning the elephant flow to one circuit. As another example, the multipath transmission control protocol (MPTCP) opens multiple connections to achieve end-to-end efficient bandwidth utilization. The MPTCP application may be best suited for the L4 load balancing algorithm 124c.

The data center 140 of the system 100 is a physical facility used by an organization to house its critical applications and data. The data center 140 may include routers, switches, firewalls, storage systems, servers, application delivery controllers, etc. These components of the data center 140 may store and/or manage business critical data, applications, etc. The data center 140 may be an enterprise data center, a managed service data center, a colocation data center, a cloud data center, a combination thereof, or any other suitable type of data center.

The servers 142 of the data center 140 are computers or other devices that provide resources, data services, and/or programs to the data center 140. The servers 142 may provide processing, memory, local storage, and network connections to drive applications. The servers 142 may be physical servers or virtualized servers. In the embodiment shown in FIG. 1 , the servers 142 transmit traffic to and/or receive traffic from the Internet circuit 150 of the Internet 112.

The Internet circuits 150 (Internet Gold Circuit 150a and Internet Silver Circuit 150b) of the system 100 are available paths between the edge router 120a to the edge router 120b. In some embodiments, the Internet circuits 150 are connections that deliver Internet services to sites (e.g., branch 130 and data center 140). The Internet circuits 150 may include leased lines, MPLS, IP security (IPsec), VPN tunnels, optical fibers, T1 connections (e.g., bonded T1 connections), digital subscriber lines (DSL) (e.g., asymmetric DSL, symmetric DSL, high data rate DSL, very high data rate DSL, etc.), cable modems, dedicated Internet access (DIA) circuits, Ethernet circuits, LAN circuits, etc. The Internet circuits 150 may include point-to-point circuits, point-to-multipoint circuits, and/or multipoint-to-multipoint circuits.

1, Internet Gold circuit 150a is associated with an Internet speed of 200 megabits per second (Mbps), while Internet Silver circuit 150b is associated with a relatively slower Internet speed of 100 Mbps. Mbps is a unit of measurement for network bandwidth and throughput.

The FTP connections 160 (FTP control connection 160a and FTP data connection 160b) of the system 100 are connections used by FTP to transfer data between a client (e.g., host 132 of branch 130) and a server (e.g., server 142 of data center 140) of the network 110. FTP is an application protocol in the IP group. FTP supports file transfers between different hosts in different Internet environments. Using FTP, files can be moved from one computer to another even if each computer runs a different operating system and uses a different file storage format.

The FTP control connection 160a is the main FTP connection for transferring commands between the host 132 of the branch 130 and the server 142 of the data center 140. In some embodiments, the host 132 of the branch 130 and the server 142 of the data center 140 use the FTP control connection 160a to create an FTP data connection 160b. The FTP data connection 160b is used by the host 132 of the branch 130 and the server 142 of the data center 140 to transfer data.

In the embodiment shown in FIG. 1 , a host 132 in a branch 130 performs FTP to a server 142 in a data center 140 and forwards the FTP traffic to an edge router 120a. The edge router 120a performs network address translation (NAT) enabled on the Internet Gold Circuit 150a and the Internet Silver Circuit 150b. The FTP control connection 160a is a main stream from an IP address 30.30.30.30, while the FTP data connection 160b is a substream from an IP address 20.20.20.20. Because the main stream tuple load balancing hash algorithm and the substream tuple load balancing hash algorithm are different, the FTP control connection 160a can be transmitted on the interface of the Internet Gold Circuit 150a, and the FTP data connection 160b can be transmitted on the interface of the Internet Silver Circuit 150b due to different load balancing algorithms. These different sources may cause packet drops at the server 142 of the data center 140 because the server 142 expects a substream from the source IP address 30.30.30.30. SaaS applications 134b with multiple flows (main and sub-flows) also face the same problem. The dynamic load balancing solution discussed below in FIG. 3 solves these problems by dynamically selecting the source-only IP load balancing algorithm 124a because it is most suitable for FTP applications 134a and SaaS applications 134b.

Although FIG. 1 shows a specific number of networks 110, edge routers 120 (edge router 120a and edge router 120b), load balancers 122, load balancing algorithms 124 (source IP only load balancing algorithm 124a, L3 load balancing algorithm 124b, L4 load balancing algorithm 124c, and PPL balancing algorithm 124d), branches 130, hosts 132, applications 134 (FTP application 134a, SaaS application 134b, elephant flow application 134c, and mouse flow application 134d), , data center 140, server 142, Internet circuit 150 (Internet gold circuit 150a and Internet silver circuit 150b), and FTP connection 160 (FTP control connection 160a and FTP data connection 160b), but the present disclosure contemplates any suitable number of networks 110, edge routers 120, load balancers 122, load balancing algorithms 124, branches 130, hosts 132, applications 134, data center 140, server 142, Internet circuit 150, and FTP connection 160. For example, system 100 can include more or less than four load balancing algorithms 124. As another example, system 100 can include more or less than four types of applications 134. As yet another example, system 100 can include more than two Internet circuits 150.

Although FIG. 1 shows a network 110, edge routers 120 (edge router 120a and edge router 120b), a load balancer 122, a load balancing algorithm 124 (source IP only load balancing algorithm 124a, L3 load balancing algorithm 124b, L4 load balancing algorithm 124c, and PPL balancing algorithm 124d), a branch 130, a host 132, an application 134 (FTP application 134a, SaaS application 134b, elephant flow application 134c, and mouse flow application 134d), a data center Although the present disclosure describes a specific arrangement of network 110, edge router 120, load balancer 122, load balancing algorithm 124, branch 130, host 132, application 134, data center 140, server 142, Internet circuit 150 (Internet gold circuit 150a and Internet silver circuit 150b), and FTP connection 160 (FTP control connection 160a and FTP data connection 160b), the present disclosure contemplates any suitable arrangement of network 110, edge router 120, load balancer 122, load balancing algorithm 124, branch 130, host 132, application 134, data center 140, server 142, Internet circuit 150, and FTP connection 160. For example, branch 130 and/or data center 140 may be replaced with different types of site locations.

Furthermore, although FIG. 1 describes and illustrates particular components, devices, or systems that perform particular actions, this disclosure contemplates any suitable combination of any suitable components, devices, or systems that perform any suitable actions.

FIG. 2 shows an example load balancing system 200 according to some embodiments. System 200 or a portion thereof may be associated with an entity, which may include any entity that performs load balancing, such as a business, company, or enterprise. In some embodiments, the entity may be a service provider that performs load balancing. The components of system 200 may include any appropriate combination of hardware, firmware, and software. For example, the components of system 200 may use one or more elements of the computer system of FIG. 5 .

In the embodiment shown in FIG. 2 , system 200 includes network 110, edge routers 120 (edge router 120a and edge router 120b), load balancer 122, load balancing algorithms 124 (source IP only load balancing algorithm 124a, L3 load balancing algorithm 124b, L4 load balancing algorithm 124c, and PPL balancing algorithm 124d), branches 130, hosts 132, applications 134 (FTP application 134a, SaaS application 134b, elephant flow application 134c, and mouse flow application 134d), data center 140, server 142, and Internet circuits 150 (Internet gold circuit 150a and Internet silver circuit 150b), which are described above in FIG. 1 .

In the embodiment shown in FIG. 2 , the elephant flow application 134 c at the host 132 of the branch 130 attempts to synchronize data to the server 142 of the data center 140. Due to the hash selection of the existing configuration (e.g., the L3 load balancing algorithm 124 b), the elephant flow packets (e.g., packets 1 to 4) may follow the Internet Gold Circuit 150 a or the Internet Silver Circuit 150 b, even though the PPL balancing algorithm 124 d is most suitable for the elephant flow application 134 c. The PPL balancing algorithm 124 d will not be selected for all traffic outgoing from the edge router 120 a because PPL load balancing may cause out-of-order delivery and retransmission of packets of the mouse traffic. The dynamic load balancing scheme discussed below in FIG. 3 solves these problems by dynamically selecting the PPL balancing algorithm 124 d for the elephant flow application 134 c to more efficiently utilize the bandwidth of the Internet Gold Circuit 150 a and the Internet Silver Circuit 150 b.

Although FIG. 2 shows a specific number of networks 110, edge routers 120 (edge router 120a and edge router 120b), load balancers 122, load balancing algorithms 124 (source IP only load balancing algorithm 124a, L3 load balancing algorithm 124b, L4 load balancing algorithm 124c, and PPL balancing algorithm 124d), branches 130, hosts 132, applications 134 (FTP application 134a, SaaS application 134b, elephant flow application 134c, and mouse flow application 134d), , data center 140, server 142, Internet circuit 150 (Internet gold circuit 150a and Internet silver circuit 150b), and FTP connection 160 (FTP control connection 160a and FTP data connection 160b), but the present disclosure contemplates any suitable number of networks 110, edge routers 120, load balancers 122, load balancing algorithms 124, branches 130, hosts 132, applications 134, data center 140, server 142, Internet circuit 150, and FTP connection 160. For example, system 200 can include more or less than three load balancing algorithms 124. As another example, system 200 can include more or less than four types of applications 134. As yet another example, system 200 can include more than two Internet circuits 150.

2 shows a network 110, an edge router 120 (edge router 120a and edge router 120b), a load balancer 122, a load balancing algorithm 124 (source IP only load balancing algorithm 124a, L3 load balancing algorithm 124b, L4 load balancing algorithm 124c, and PPL balancing algorithm 124d), a branch 130, a host 132, an application 134 (FTP application 134a, SaaS application 134b, elephant flow application 134c, and mouse flow application 134d), a data center Although the present disclosure describes a specific arrangement of network 110, edge router 120, load balancer 122, load balancing algorithm 124, branch 130, host 132, application 134, data center 140, server 142, Internet circuit 150 (Internet gold circuit 150a and Internet silver circuit 150b), and FTP connection 160 (FTP control connection 160a and FTP data connection 160b), the present disclosure contemplates any suitable arrangement of network 110, edge router 120, load balancer 122, load balancing algorithm 124, branch 130, host 132, application 134, data center 140, server 142, Internet circuit 150, and FTP connection 160. For example, branch 130 and/or data center 140 may be replaced with different types of site locations.

Furthermore, although FIG. 2 describes and illustrates particular components, devices, or systems that perform particular actions, this disclosure contemplates any suitable combination of any suitable components, devices, or systems that perform any suitable actions.

FIG. 3 illustrates an example load balancing system 300 for dynamically selecting a load balancing algorithm 124. The system 300 or a portion thereof may be associated with an entity, which may include any entity, such as a business, company, or enterprise, that dynamically selects a load balancing algorithm 124. In some embodiments, the entity may be a service provider that dynamically selects a load balancing algorithm 124. The components of the system 300 may include any suitable combination of hardware, firmware, and software. For example, the components of the system 300 may use one or more elements of the computer system of FIG. 5.

In the embodiment shown in Figure 3, the system 300 includes a network 110, edge routers 120 (edge router 120a and edge router 120b), a load balancer 122, load balancing algorithms 124 (source IP only load balancing algorithm 124a, L3 load balancing algorithm 124b, L4 load balancing algorithm 124c and PPL balancing algorithm 124d), a branch 130, a host 132, an application 134 (FTP application 134a, SaaS application 134b, elephant flow application 134c and mouse flow application 134d), a data center 140, a server 142 and an Internet circuit 150 (Internet gold circuit 150a and Internet silver circuit 150b), which are described above in Figure 1.

The system 300 also includes a load balancing scheme 310. The load balancing scheme 310 is a way of distributing within the network 110. The load balancing scheme 310 includes an application-based load balancing scheme 310a, an application-to-algorithm mapping load balancing scheme 310b, a policy-based load balancing scheme 310c, and a default load balancing scheme 310d.

The application-based load balancing scheme 310a dynamically selects a load balancing algorithm 124 (source-only IP load balancing algorithm 124a, L3 load balancing algorithm 124b, L4 load balancing algorithm 124c, and PPL balancing algorithm 124d) based on the type of application 134. Because different types of applications 134 are more suitable for different load balancing algorithms 124, the edge router 120a can use the application-based load balancing scheme 310a to select the load balancing algorithm 124 that is most suitable for the type of application 134. For example, the edge router 120a can select the source-only IP load balancing algorithm 124a for the FTP application 134a and the SaaS application 134b to prevent discarded packets. As another example, the edge router 120a can select the PPL balancing algorithm 124d for the elephant flow application 134c to effectively utilize bandwidth. As another example, the edge router 120a can select the L3 load balancing algorithm 124b for the mouse flow application 134d.

In some embodiments, the edge router 120a can use an application recognition engine such as network-based application recognition (NBAR) to recognize the type of application 134. NBAR uses deep packet inspection (DPI) to identify and classify the application 134. Once the edge router 120a has classified the application 134, the edge router 120a can select an appropriate load balancing algorithm 170 based on the type of application 134 according to the application-based load balancing scheme 310a.

The application-to-algorithm mapping load balancing scheme 310b selects a load balancing algorithm 124 (e.g., source-only IP load balancing algorithm 124a, L3 load balancing algorithm 124b, L4 load balancing algorithm 124c, and PPL balancing algorithm 124d) based on a global application-to-algorithm mapping table configured by a user. For example, a user may map a group of voice/video applications 134 to an L3 load balancing algorithm 124b. As another example, a user may map a group of social network applications 134 to a source-only IP load balancing algorithm 124a. As yet another example, a user may map a hypertext transfer protocol (HTTP) application 134 to an L4 load balancing algorithm 170c.

The policy-based load balancing scheme 310c selects a load balancing algorithm 124 (e.g., source-only IP load balancing algorithm 124a, L3 load balancing algorithm 124b, L4 load balancing algorithm 124c, and PPL balancing algorithm 124d) based on any standard matching criteria. The matching criteria may include one or more of the following attributes: type of application, source IP address, destination IP address, subnet, IP DSCP, protocol type, site identification, VPN identification, sequence number, etc.

In some embodiments, the edge router 120a may use a policy-based load balancing scheme 310c to select a load balancing algorithm 124 based on a numbered (ordered) sequence of match-action pairs that are evaluated in order. For example, a data policy may be configured to first identify whether the application 134 is a SaaS application 134b, next identify whether the SaaS application 134b is associated with a specific VPN list, and then next identify whether the SaaS application 134b is associated with a specific list of SaaS applications 134b (e.g., WebEx audio, WebEx video, etc.). If all of these conditions are met, the edge router 120a matches the SaaS application 134b to the source-only IP load balancing algorithm 124a configured by the policy. The policy-based load balancing scheme 310c is flexible and can be used to define and select different actions (e.g., load sharing) based on different criteria.

If the application-based load balancing scheme 310a, the application-to-algorithm mapping load balancing scheme 310b, and the policy-based load balancing scheme 310c are not selected, the default load balancing scheme 310d is automatically selected. The default load balancing scheme 310d may be based on a global/system-level load balancing configuration, a VPN-level load balancing configuration, a feature-level load balancing configuration, etc.

In operation, a user selects an application-based load balancing scheme 310a for an edge router 120a of the system 300. The edge router 120a receives FTP traffic from a host 132 of a branch 130 via an FTP control connection 160a and an FTP data connection 160b. The edge router 120a associates the FTP traffic with an FTP application 134a using NBAR. The edge router 120a uses the application-based load balancing scheme 310a to dynamically select a source-only IP load balancing algorithm 124a associated with the FTP application 134a, which ensures that the FTP traffic will follow the same route (internet gold circuit 150a) to a server 142 of a data center 140. The edge router 120a of the system 100 also receives elephant flow traffic from a host 132 of a branch 130. The edge router 120a associates the elephant flow traffic with an elephant flow application 134c using NBAR. The edge router 120a uses the application-based load balancing scheme 310a to select the PPL balancing algorithm 124d, which effectively distributes the elephant flow packets 1, 2, 3, and 4 between the Internet circuits 150. For example, as shown in FIG3, the edge router 120a can distribute the elephant flow packets 1, 2, 3, and 4 in a round-robin manner, so that the elephant flow packets 1 and 3 follow the Internet silver circuit 150b, and the elephant flow packets 2 and 4 follow the Internet gold circuit 150a. In this way, the system 300 can be configured to dynamically select a load balancing scheme based on a certain criterion such as the application type, which improves the bandwidth and overall efficiency of the system 300.

3 shows a specific number of networks 110, edge routers 120 (edge router 120a and edge router 120b), load balancers 122, load balancing algorithms 124 (source IP only load balancing algorithm 124a, L3 load balancing algorithm 124b, L4 load balancing algorithm 124c, and PPL balancing algorithm 124d), branches 130, hosts 132, applications 134 (FTP application 134a, SaaS application 134b, elephant flow application 134c, and mouse flow application 134d), data centers 140, servers 142, Internet circuits 150 (Internet gold circuit 150a and Internet silver circuit 150b), FTP connections 160 (FTP control connection 160a and FTP data connection 160b), and load balancing schemes 310 (application-based load balancing scheme 310a, application-to-algorithm mapping load balancing scheme 310b, policy-based load balancing scheme 310c, and default load balancing scheme 310d), but the present disclosure contemplates any appropriate number of networks 110, edge routers 120, load balancers 122, load balancing algorithms 124, branches 130, hosts 132, applications 134, data centers 140, servers 142, Internet circuits 150, FTP connections 160, and load balancing schemes 310. For example, system 300 may include more or less than four load balancing schemes 310. As another example, system 300 may include more or less than four types of applications 134. As yet another example, system 300 may include more than two Internet circuits 150.

3 shows a network 110, edge routers 120 (edge router 120a and edge router 120b), a load balancer 122, load balancing algorithms 124 (source IP only load balancing algorithm 124a, L3 load balancing algorithm 124b, L4 load balancing algorithm 124c, and PPL balancing algorithm 124d), a branch 130, a host 132, an application 134 (FTP application 134a, SaaS application 134b, elephant flow application 134c, and mouse flow application 134d), a data center 140, a server 142, and an Internet circuit 150 (Internet gold circuit 150a and Internet silver circuit 150b). , FTP connections 160 (FTP control connection 160a and FTP data connection 160b), and load balancing schemes 310 (application-based load balancing scheme 310a, application-to-algorithm mapping load balancing scheme 310b, policy-based load balancing scheme 310c, and default load balancing scheme 310d), but the present disclosure contemplates any suitable arrangement of network 110, edge router 120, load balancer 122, load balancing algorithm 124, branch 130, host 132, application 134, data center 140, server 142, Internet circuit 150, FTP connection 160, and load balancing scheme 310. For example, branch 130 and/or data center 140 may be replaced with different types of site locations.

Furthermore, although FIG. 3 describes and illustrates particular components, devices, or systems that perform particular actions, this disclosure contemplates any suitable combination of any suitable components, devices, or systems that perform any suitable actions.

FIG4 illustrates a method 400 for dynamically selecting a load balancing algorithm according to some embodiments. The method 400 uses an application-based load balancing scheme to dynamically select a load balancing algorithm (e.g., a source IP-only load balancing algorithm, an L3 load balancing algorithm, an L4 load balancing algorithm, a PPL balancing algorithm, etc.) based on one or more criteria such as application type, which improves the efficiency of the network because different types of applications are more suitable for different load balancing algorithms. The method 400 of FIG4 can be used by the system 300 of FIG3.

Method 400 begins at step 410. At step 420 of method 400, an edge router of the network receives traffic from a host of a branch. For example, referring to system 300 of FIG. 3, edge router 120a of network 110 may receive traffic, such as FTP traffic or elephant flow traffic, from host 132 of branch 130. In some embodiments, the network is an SD-WAN network. Then, method 400 moves from step 420 to step 430.

At step 430 of method 400, the edge router identifies one or more attributes associated with the traffic. For example, referring to the system 300 of FIG. 3, the edge router 120a of network 110 can identify the type of application 134 associated with the traffic (e.g., FTP application 134a, SaaS application 134b, elephant flow application 134c, or mouse flow application 134d). In some embodiments, the edge router uses an application recognition engine such as NBAR to identify the type of application 134. As another example, the edge router 120a of network 110 can identify attributes associated with the traffic, such as source IP address, destination IP address, subnet, IP DSCP, protocol type, etc. Then, method 400 moves from step 430 to step 440.

At step 440 of method 400, the edge router determines whether a particular load balancing scheme has been selected. For example, referring to the system 300 of FIG. 3, the edge router 120a of the network 110 may determine whether an application-based load balancing scheme 310a, an application-to-algorithm mapping load balancing scheme 310b, or a policy-based load balancing scheme 310c has been selected. If the edge router determines that a particular load balancing scheme has been selected, the method 400 moves from step 440 to step 450, where the edge router dynamically selects a load balancing algorithm based on one or more attributes according to the load balancing scheme. For example, referring to the system 300 of FIG. 3, the edge router 120a of the network 110 may determine that an application-based load balancing scheme 310a has been selected. Based on the application-based load balancing scheme 310a, the edge router 120a may dynamically select a source-only IP load balancing algorithm 124a, an L3 load balancing algorithm 124b, an L4 load balancing algorithm 124c, or a PPL balancing algorithm 124d based on the type of the identified application 134.

If the edge router determines at step 440 of method 400 that a particular load balancing scheme is not selected, then the method 400 proceeds from step 440 to step 470, where the edge router performs load balancing on the traffic according to a default load balancing scheme. For example, referring to the system 300 of FIG. 3 , the edge router 120 a of the network 110 may determine that the application-based load balancing scheme 310 a, the application-to-algorithm mapping load balancing scheme 310 b, and the policy-based load balancing scheme 310 c are not selected. The edge router 120 a may use a default load balancing scheme 310 d, which may be configured to use a source-only IP load balancing algorithm 124 a, an L3 load balancing algorithm 124 b, an L4 load balancing algorithm 124 c, or a PPL balancing algorithm 124 d for all traffic types.

Then, the method 400 moves from step 460 and step 470 to step 480, where the edge router transmits the traffic to the second network element according to the load balancing scheme. For example, referring to the system 300 of FIG. 3 , the edge router 120a of the network 110 can transmit the traffic to the edge router 120b according to the source-only IP load balancing algorithm 124a, the L3 load balancing algorithm 124b, the L4 load balancing algorithm 124c, or the PPL balancing algorithm 124d. Then, the method 400 moves from step 480 to step 490, where the method 400 ends. In this way, the method 400 can be configured to dynamically select a load balancing scheme according to a certain criterion such as application type, which improves the bandwidth and overall efficiency of the network.

Although this disclosure describes and illustrates particular steps of method 400 of FIG. 4 occurring in a particular order, this disclosure contemplates any suitable steps of method 400 of FIG. 4 occurring in any suitable order. Although this disclosure describes and illustrates an example method for dynamically selecting a load balancing algorithm that includes particular steps of method 400 of FIG. 4 , this disclosure contemplates any suitable method for dynamically selecting a load balancing algorithm that includes any suitable steps, which may include all, some, or none of the steps of method 400 of FIG. 4 , where appropriate. Although FIG. 4 describes and illustrates particular components, devices, or systems that perform particular actions, this disclosure contemplates any suitable combination of any suitable components, devices, or systems that perform any suitable actions.

FIG. 5 shows an example computer system 500. In a particular embodiment, one or more computer systems 500 perform one or more steps of one or more methods described or illustrated herein. In a particular embodiment, one or more computer systems 500 provide the functionality described or illustrated herein. In a particular embodiment, software running on one or more computer systems 500 performs one or more steps of one or more methods described or illustrated herein, or provides the functionality described or illustrated herein. A particular embodiment includes one or more parts of one or more computer systems 500. Where appropriate, references to computer systems herein may cover computing devices, and vice versa. In addition, where appropriate, references to computer systems may cover one or more computer systems.

The present disclosure contemplates any suitable number of computer systems 500. The present disclosure contemplates computer systems 500 in any suitable physical form. By way of example and not limitation, computer system 500 may be an embedded computer system, a system on a chip (SOC), a single board computer system (SBC) (e.g., a computer on a module (COM) or a system on a module (SOM)), a desktop computer system, a laptop or notebook computer system, an interactive kiosk, a mainframe, a computer system grid, a mobile phone, a personal digital assistant (PDA), a server, a tablet computer system, an augmented/virtual reality device, or a combination of two or more thereof. Where appropriate, computer system 500 may include one or more computer systems 500; be unified or distributed; span multiple locations; span multiple machines; span multiple data centers; or reside in a cloud, which may include one or more cloud components in one or more networks. Where appropriate, one or more computer systems 500 may perform one or more steps of one or more methods described or illustrated herein without substantial spatial or temporal limitations. By way of example and not limitation, one or more computer systems 500 may perform one or more steps of one or more methods described or illustrated herein in real time or in a batch mode. Where appropriate, one or more computer systems 500 may perform at different times or at different locations one or more steps of one or more methods described or illustrated herein.

In a particular embodiment, computer system 500 includes processor 502, memory 504, storage 506, input/output (I/O) interface 508, communication interface 510, and bus 512. Although this disclosure describes and illustrates a particular computer system having particular numbers of particular components in a particular arrangement, this disclosure contemplates any suitable computer system having any suitable numbers of any suitable components in any suitable arrangement.

In a particular embodiment, the processor 502 includes hardware for executing instructions such as those constituting a computer program. By way of example and not limitation, to execute instructions, the processor 502 may retrieve (or fetch) instructions from an internal register, an internal cache, a memory 504, or a storage device 506; decode and execute them; and then write one or more results to an internal register, an internal cache, a memory 504, or a storage device 506. In a particular embodiment, the processor 502 may include one or more internal caches for data, instructions, or addresses. Where appropriate, the present disclosure contemplates a processor 502 including any appropriate number of any appropriate internal caches. By way of example and not limitation, the processor 502 may include one or more instruction caches, one or more data caches, and one or more translation lookaside buffers (TLBs). The instructions in the instruction cache may be copies of instructions in the memory 504 or the storage device 506, and the instruction cache may speed up the retrieval of these instructions by the processor 502. The data in the data cache may be a copy of data in the memory 504 or storage device 506, and instructions are executed at the processor 502 to operate on the data; it may be the result of a previous instruction executed by the processor 502, which is used for access by subsequent instructions executed at the processor 502 or for writing to the memory 504 or storage device 506; or it may be other appropriate data. The data cache can speed up the read or write operation of the processor 502. The TLB can speed up the virtual address translation of the processor 502. In a particular embodiment, the processor 502 may include one or more internal registers for data, instructions, or addresses. Where appropriate, the present disclosure contemplates a processor 502 including any appropriate number of any appropriate internal registers. Where appropriate, the processor 502 may include one or more arithmetic logic units (ALUs); may be a multi-core processor, or may include one or more processors 502. Although the present disclosure describes and illustrates a particular processor, the present disclosure contemplates any appropriate processor.

In certain embodiments, the memory 504 includes a main memory for storing instructions executed by the processor 502 or for storing data operated by the processor 502. As an example and not limitation, the computer system 500 may load instructions from the storage device 506 or another source (e.g., another computer system 500) to the memory 504. The processor 502 may then load the instructions from the memory 504 to an internal register or internal cache. To execute the instructions, the processor 502 may retrieve the instructions from the internal register or internal cache and decode them. During or after the execution of the instructions, the processor 502 may write one or more results (which may be intermediate or final results) to the internal register or internal cache. The processor 502 may then write one or more of these results to the memory 504. In certain embodiments, the processor 502 executes only instructions in one or more internal registers or internal caches or memory 504 (as opposed to the storage device 506 or elsewhere) and operates only on data in one or more internal registers or internal caches or memory 504 (as opposed to the storage device 506 or elsewhere). One or more memory buses (which may each include an address bus and a data bus) may couple the processor 502 to the memory 504. The bus 512 may include one or more memory buses, as described below. In a particular embodiment, one or more memory management units (MMUs) are located between the processor 502 and the memory 504 and facilitate access to the memory 504 requested by the processor 502. In a particular embodiment, the memory 504 includes a random access memory (RAM). Where appropriate, the RAM may be a volatile memory. Where appropriate, the RAM may be a dynamic RAM (DRAM) or a static RAM (SRAM). In addition, where appropriate, the RAM may be a single-port or multi-port RAM. The present disclosure contemplates any appropriate RAM. Where appropriate, the memory 504 may include one or more memories 404. Although the present disclosure describes and illustrates specific memories, the present disclosure contemplates any appropriate memory.

In a particular embodiment, the storage device 506 includes a large-capacity storage device for data or instructions. As an example and not limitation, the storage device 506 may include a hard disk drive (HDD), a floppy disk drive, a flash memory, an optical disk, a magneto-optical disk, a tape, or a universal serial bus (USB) drive, or a combination of two or more thereof. Where appropriate, the storage device 506 may include a removable or non-removable (or fixed) medium. Where appropriate, the storage device 506 may be located inside or outside the computer system 500. In a particular embodiment, the storage device 506 is a non-volatile solid-state memory. In a particular embodiment, the storage device 506 includes a read-only memory (ROM). Where appropriate, the ROM may be a mask-programmed ROM, a programmable ROM (PROM), an erasable PROM (EPROM), an electrically erasable PROM (EEPROM), an electrically alterable ROM (EAROM) or a flash memory, or a combination of two or more thereof. The present disclosure contemplates a large-capacity storage device 506 in any appropriate physical form. Where appropriate, the storage device 506 may include one or more storage device control units that facilitate communication between the processor 502 and the storage device 506. Where appropriate, storage 506 may include one or more storage devices 506. Although this disclosure describes and illustrates particular storage devices, this disclosure contemplates any suitable storage devices.

In a particular embodiment, the I/O interface 508 includes hardware, software, or both, providing one or more interfaces for communication between the computer system 500 and one or more I/O devices. Where appropriate, the computer system 500 may include one or more of these I/O devices. One or more of these I/O devices may enable communication between a person and the computer system 500. As an example and not limitation, the I/O device may include a keyboard, a keypad, a microphone, a display, a mouse, a printer, a scanner, a speaker, a camera, a stylus, a tablet computer, a touch screen, a trackball, a camera, another suitable I/O device, or a combination of two or more thereof. The I/O device may include one or more sensors. The present disclosure contemplates any suitable I/O device and any suitable I/O interface 408 for them. Where appropriate, the I/O interface 508 may include one or more devices or software drivers that enable the processor 502 to drive one or more of these I/O devices. Where appropriate, the I/O interface 508 may include one or more I/O interfaces 408. Although this disclosure describes and illustrates a particular I/O interface, this disclosure contemplates any suitable I/O interface.

In certain embodiments, the communication interface 510 includes hardware, software, or both, providing one or more interfaces for communication (e.g., packet-based communication) between the computer system 500 and one or more other computer systems 500, or between the computer system 500 and one or more networks. By way of example and not limitation, the communication interface 510 may include a network interface controller (MIC) or a network adapter for communicating with an Ethernet or other wire-based networks; or the communication interface 510 may include a wireless NIC (WNIC) or a wireless adapter for communicating with a wireless network (e.g., a WI-FI network). The present disclosure contemplates any appropriate network and any appropriate communication interface 510 for the network. By way of example and not limitation, the computer system 500 may communicate with one or more portions of an ad hoc network, a personal area network (PAN), a LAN, a WAN, a MAN, or the Internet, or a combination of two or more of these networks. One or more portions of one or more of these networks may be wired or wireless. As examples, computer system 500 may communicate with a wireless PAN (WPAN) (e.g., a Bluetooth WPAN), a WI-FI network, a WI-MAX network, a cellular telephone network (e.g., a Global System for Mobile Communications (GSM) network, a 3G network, a 4G network, a 5G network, an LTE network, or other suitable wireless networks, or a combination of two or more thereof. Where appropriate, computer system 500 may include any suitable communication interface 510 for any of these networks. Where appropriate, communication interface 510 may include one or more communication interfaces 510. Although this disclosure describes and illustrates particular communication interfaces, this disclosure contemplates any suitable communication interface.

In a particular embodiment, bus 512 includes hardware, software, or both that couple the components of computer system 500 to each other. As an example and not limitation, bus 512 may include accelerated graphics port (AGP) or other graphics bus, enhanced industry standard architecture (EISA) bus, front side bus (FSB), HYPERTRANSPORT (HT) interconnect, industry standard architecture (ISA) bus, INFINIBAND interconnect, low pin count (LPC) bus, memory bus, micro channel architecture (MCA) bus, peripheral component interconnect (PCI) bus, PCI-Express (PCIe) bus, serial advanced technology attachment (SATA) bus, video electronics standard association local (VLB) bus, or other appropriate bus, or a combination of two or more buses thereof. Where appropriate, bus 512 may include one or more buses 512. Although the present disclosure describes and illustrates a specific bus, the present disclosure contemplates any appropriate bus or interconnect.

Where appropriate, the computer-readable non-transitory storage medium herein may include one or more semiconductor-based circuits or other integrated circuits (ICs) (e.g., field programmable gate arrays (FPGAs) or application-specific ICs (ASICs)), hard disk drives (HDDs), hybrid hard disk drives (HMDs), optical disks, optical disk drives (ODDs), magneto-optical disks, magneto-optical drives, floppy disks, floppy disk drives (FDDs), magnetic tapes, solid-state drives (SSDs), RAM drives, secure digital (SECURE DITIGAL) cards or drives, any other suitable computer-readable non-transitory storage media, or any suitable combination of two or more of these media. Where appropriate, the non-transitory computer-readable storage medium may be volatile, non-volatile, or a combination of volatile and non-volatile.

As used herein, "or" is inclusive and not exclusive, unless expressly stated otherwise or the context indicates otherwise. Thus, as used herein, "A or B" means "A, B, or both," unless expressly stated otherwise or the context indicates otherwise. Additionally, "and" means both jointly and severally, unless expressly stated otherwise or the context indicates otherwise. Thus, as used herein, "A and B" means "A and B, jointly or severally," unless expressly stated otherwise or the context indicates otherwise.

The scope of the present disclosure covers all changes, substitutions, variations, alterations and modifications to the example embodiments described or illustrated herein that will be understood by those of ordinary skill in the art. The scope of the present disclosure is not limited to the example embodiments described or illustrated herein. In addition, although the present disclosure describes and illustrates various embodiments herein as including specific components, elements, features, functions, operations or steps, any of these embodiments may include any combination or arrangement of any components, elements, features, functions, operations or steps described or illustrated anywhere in this article that will be understood by those of ordinary skill in the art. In addition, in the attached claims, references to devices or systems or components of devices or systems being adapted to, arranged to, capable of, configured to, enabled to, operable to or operated to perform specific functions cover whether the device, system, component or the specific function is activated, turned on or unlocked, as long as the device, system or component is so adapted, arranged, capable of, configured to, enabled to, operable to or operated. In addition, although the present disclosure describes or illustrates specific embodiments as providing specific advantages, specific embodiments may not provide these advantages, provide some advantages or provide all of these advantages.

Claims (23)

1. A first network element comprising one or more processors and one or more computer-readable non-transitory storage media coupled to the one or more processors and comprising instructions that, when executed by the one or more processors, cause the first network element to perform operations comprising:

Receiving traffic;

identifying one or more attributes associated with the traffic;

dynamically selecting a load balancing algorithm based on the one or more attributes according to a load balancing scheme;

performing load balancing on the traffic according to the load balancing algorithm; and

And transmitting the traffic to a second network element according to the load balancing.

2. The first network element of claim 1, the operations further comprising: the load balancing algorithm is dynamically selected from the following load balancing algorithms:

A source-only IP address load balancing algorithm;

An L3 load balancing algorithm;

An L4 load balancing algorithm; and

Load balancing algorithms per packet.

3. The first network element of claim 1 or 2, wherein:

The one or more attributes associated with the traffic include an application type; and

The application type is one of the following:

File Transfer Protocol (FTP) applications;

software as a service (SaaS) applications;

Elephant flow application; and

Mouse flow application.

4. A first network element according to any one of claims 1 to 3, wherein:

The one or more attributes associated with the traffic include an application type; and

The load balancing scheme is an application-based load balancing scheme that matches the application type with the load balancing algorithm.

5. The first network element of any of claims 1 to 4, wherein the load balancing scheme is a user defined application-to-algorithm mapping load balancing scheme matching application types with user defined load balancing algorithms.

6. The first network element of any of claims 1-5, wherein the load balancing scheme is a policy-based load balancing scheme that uses the one or more attributes and policies to select the load balancing algorithm.

7. The first network element of any of claims 1 to 6, wherein the one or more attributes comprise one or more of the following set of attributes:

An application type;

A source Internet Protocol (IP) address;

a destination IP address;

A sub-network;

IP Differentiated Services Code Points (DSCP); and

Protocol type.

8. A method, comprising:

Receiving traffic;

identifying one or more attributes associated with the traffic;

dynamically selecting a load balancing algorithm based on the one or more attributes according to a load balancing scheme;

performing load balancing on the traffic according to the load balancing algorithm; and

The traffic is transferred from the first network element to the second network element according to the load balancing.

9. The method of claim 8, further comprising: the load balancing algorithm is dynamically selected from the following load balancing algorithms:

A source-only IP address load balancing algorithm;

An L3 load balancing algorithm;

An L4 load balancing algorithm; and

Load balancing algorithms per packet.

10. The method according to claim 8 or 9, wherein:

The one or more attributes associated with the traffic include an application type; and

The application type is one of the following:

File Transfer Protocol (FTP) applications;

software as a service (SaaS) applications;

Elephant flow application; and

Mouse flow application.

11. The method of any one of claims 8 to 10, wherein:

The one or more attributes associated with the traffic include an application type; and

The load balancing scheme is an application-based load balancing scheme that matches the application type with the load balancing algorithm.

12. The method of any of claims 8 to 11, wherein the load balancing scheme is a user defined application-to-algorithm mapping load balancing scheme matching application types to user defined load balancing algorithms.

13. The method of any of claims 8 to 12, wherein the load balancing scheme is a policy-based load balancing scheme that uses the one or more attributes and policies to select the load balancing algorithm.

14. The method of any of claims 8 to 13, wherein the one or more attributes comprise one or more of the following set of attributes:

An application type;

A source Internet Protocol (IP) address;

a destination IP address;

A sub-network;

IP Differentiated Services Code Points (DSCP); and

Protocol type.

15. One or more computers

A readable non-transitory storage medium comprising instructions that, when executed by a processor, cause the processor to perform operations comprising:

Receiving traffic;

identifying one or more attributes associated with the traffic;

dynamically selecting a load balancing algorithm based on the one or more attributes according to a load balancing scheme;

performing load balancing on the traffic according to the load balancing algorithm; and

The traffic is transferred from the first network element to the second network element according to the load balancing.

16. The one or more computer-readable non-transitory storage media of claim 15, the operations further comprising: the load balancing algorithm is dynamically selected from the following load balancing algorithms:

A source-only IP address load balancing algorithm;

An L3 load balancing algorithm;

An L4 load balancing algorithm; and

Load balancing algorithms per packet.

17. The one or more computer-readable non-transitory storage media of claim 15 or 16, wherein:

The one or more attributes associated with the traffic include an application type; and

The application type is one of the following:

File Transfer Protocol (FTP) applications;

software as a service (SaaS) applications;

Elephant flow application; and

Mouse flow application.

18. The one or more computer-readable non-transitory storage media of any of claims 15-17, wherein:

The one or more attributes associated with the traffic include an application type; and

The load balancing scheme is an application-based load balancing scheme that matches the application type with the load balancing algorithm.

19. The one or more computer-readable non-transitory storage media of any of claims 15 to 18, wherein the load balancing scheme is a user-defined application-to-algorithm mapping load balancing scheme that matches application types to user-defined load balancing algorithms.

20. The one or more computer-readable non-transitory storage media of any of claims 15-19, wherein the load balancing scheme is a policy-based load balancing scheme that uses the one or more attributes and policies to select the load balancing algorithm.

21. An apparatus, comprising:

A module for receiving traffic;

means for identifying one or more attributes associated with the traffic;

means for dynamically selecting a load balancing algorithm based on the one or more attributes according to a load balancing scheme;

Means for performing load balancing on the traffic according to the load balancing algorithm; and

And means for transferring the traffic from the first network element to the second network element in accordance with the load balancing.

22. The apparatus of claim 21, further comprising means for implementing the method of any one of claims 9 to 14.

23. A computer program, computer program product or computer readable medium comprising instructions which, when executed by a computer, cause the computer to perform the steps of the method according to any of claims 8 to 14.