Cisco Technology, Inc.

Samar Sharma

Mouli Vytla

Rajendra Kumar Thirumurthi

Avni Baveja

Sanjaya Kumar

Mitali Parthasarathy

Dinesh Ganapathy Dutt

Ashish Purushottam Attarde

Chao Feng

Rina Panigrahy

Umesh Mahajan

Nikhil Dhar

Fabio Rodolfo Maino

Abhijit Patra

Biju Mathews Mammen

Thomas James Edsall

Diheng Qu

Roy D'Cruz

Nitesh G Soni

Balaji Sundararajan

Arindam Paul

Gaurav Rastogi

Prashant Billore

Jhaanaki M Krishnan

Pramod Cherukumilli

Rajesh Bhandari

Suresh Kumar Nalluru

Rakesh B Goudar

Ronak K Desai

Murali Basavaiah

Jeevan Madhu Kamisetty

Chandramouli Radhakrishnan

Ankur Goyal

Ravinder Reddy Amanaganti

Manish Chandra Agrawal

Venkatabalakrishnan Krishnamurthy

Sachin Vijay Vishwarupe

Praveen Parthasarathy Iyengar

Rahul Chari

Devendra Mohan Badhani

Manuwela Kanade

Srinivas Avasarala

Puneet Kumar

Mahadeva Swamy Chandra

Deven Walia

Kalyan Venkataramana

Shyam N Kapadia

Silvano Gai

Sanjay Sane

Anand Girish Parthasarathy

Lukas Krattiger

Chandan Mishra

Sameer Merchant

Ashwath Nagaraj

John Andrew Fingerhut

Suran Saminda De Silva

Kushal A Patel

Venkataraman Natham

Yoichi Hariguchi

Raghavendra J Rao

Sriram Chidambaram

Rajesh Sharma

Shriram Velaga

Anand Akella

Walter Song

Goran Saradzic

Deepak Cherian

Krishna Bharadwaj Dharwada

Vivek Purushotham

Pavani Baddepudi

Manjunath Masanagi

Christopher Jon Bucci

Jon Wei

Chandrashekarappa Surekha Puttasubbappa

Gopinath Durairaj

Bala Nagesh

In an example, a network switch is configured to natively act as a high-speed load balancer. Numerous load-balancing techniques may be used, including one that bases the traffic 'bucket' on a source IP address of an incoming packet. This particular technique provides a network administrator a powerful tool for shaping network traffic. For example, by assigning certain classes of computers on the network particular IP addresses, the network administrator can ensure that the traffic is load balanced in a desirable fashion. To further increase flexibility, the network administrator may apply a bit mask to the IP address, and expose only a portion, selected from a desired octet of the address.

Adjustable bit mask for high-speed native load balancing on a switch

In one embodiment, load balancing criteria and an indication of a plurality of network nodes is received. A plurality of forwarding entries are created based on the load balancing criteria and the indication of the plurality of nodes. A content addressable memory of a network element is programmed with the plurality of forwarding entries. The network element selectively load balances network traffic by applying the plurality of forwarding entries to the network traffic, wherein network traffic meeting the load balancing criteria is load balanced among the plurality of network nodes.

Selective load balancing of network traffic

Systems, methods, and computer-readable media for creating service chains for inter-cloud traffic. In some examples, a system receives domain name system (DNS) queries associated with cloud domains and collects DNS information associated the cloud domains. The system spoofs DNS entries defining a subset of IPs for each cloud domain. Based on the spoofed DNS entries, the system creates IP-to-domain mappings associating each cloud domain with a respective IP from the subset of IPs. Based on the IP-to-domain mappings, the system programs different service chains for traffic between a private network and respective cloud domains. The system routes, through the respective service chain, traffic having a source associated with the private network and a destination matching the IP in the respective IP-to-domain mapping.

Service chains for inter-cloud traffic

In an example, a network switch is configured to operate natively as a load balancer. The switch receives incoming traffic on a first interface communicatively coupled to a first network, and assigns the traffic to one of a plurality of traffic buckets. This may include looking up a destination IP of an incoming packet in a fast memory such as a ternary content-addressable memory (TCAM) to determine whether the packet is directed to a virtual IP (VIP) address that is to be load balanced. If so, part of the source destination IP address may be used as a search tag in the TCAM to assign the incoming packet to a traffic bucket or IP address of a service node.

TCAM-based load balancing on a switch

In an example, there is disclosed a network switch or other computing apparatus comprising: an ingress interface; a plurality of egress interfaces; and one or more logic elements, including at least a content addressable memory (CAM), comprising a channel selection engine to provide persistent channel selection comprising: receive a packet on the ingress interface; inspect a layer 2 (L2) attribute of the packet; lookup the L2 attribute in the CAM; and assign the packet to an egress interface communicatively coupled to a network service.

Layer 2 channel selection

The disclosed technology relates to a load balancing system. A load balancing system is configured to receive health monitoring metrics, at a controller, from a plurality of leaf switches. The load balancing system is further configured to determine, based on the health monitoring metrics, that a server has failed and modify a load balancing configuration for the network fabric. The load balancing system is further configured to transmit the load balancing configuration to each leaf switch in the network fabric and update the tables in each leaf switch to reflect an available server.

Distributed load balancer health management using data center network manager

Presented herein are service chaining techniques for selective traffic redirection based on Access Control List (ACL) configurations on switches. Network traffic forwarded along one or more service chains may be monitored on the basis of individual segments of the service chains. In one example, the network traffic forwarded along individual segments may be counted on a per-segment basis.

Service chaining segmentation analytics

Presented herein are techniques to support service chains in a network, such as a spine-leaf switch fabric network, that also uses overlay networking technology. More specifically, in accordance with the techniques presented herein, a linecard at an ingress network node for an overlay network is configured to receive a packet. Using information obtained from the packet, a hardware lookup is performed at the linecard to identify a service chain with which the packet is associated. An entry corresponding to the identified service chain is identified within a memory location of the linecard, where the entry includes overlay network information for forwarding packets along the identified service chain via an overlay network. Using the overlay network information included in the identified entry, the packet is encapsulated with an overlay packet header for use in forwarding the packet via the overlay network.

Overlay network hardware service chaining

A method including: in a network element that includes one or more hardware memory resources of fixed storage capacity for storing data used to configure a plurality of networking features of the network element and a utilization management process running on the network element, the utilization management process performing operations including: obtaining a plurality of entries of the one or more hardware memory resources representing utilization of the one or more hardware memory resources by network traffic passing through the network element; sorting the plurality of entries of the one or more hardware memory resources by statistics associated with the network traffic passing through the network element to produce sorted entries; and sending the extracted to a network management application for display is disclosed. An apparatus and one or more non-transitory computer readable storage media to execute the method are also provided.

Sorting and displaying network traffic analytics

In an embodiment, a method is provided. The method includes: storing, in at least one hardware module of a network device having a plurality of ports, attributes for at least one access control list and associated actions that cause network packets received at one of the plurality of ports that match the attributes for the at least one access control list, to be directed into a service chain that includes at least a first network processing application specified according to a port and a second network processing application specified according to an internet protocol (IP) address; and directing a received network packet that matches the attributes for the at least one access control list into the service chain.

Hybrid services insertion

A method including: storing, in at least one hardware module of a network device having a plurality of ports, attributes for at least one access control list and associated actions that cause network packets received at one of the plurality of ports that match the attributes for the at least one access control list, to be directed into a service chain that includes at least a first application performed by any one of a first plurality of redundant network processing appliance elements connected to another port of the plurality of ports; directing received network packets that match the attributes for the at least one access control list into the service chain; and load balancing network packets among the first plurality of redundant network processing appliance elements for the first application based on the attributes stored in the at least one hardware module of the network device.

Load balancing in a service chain

A switch/switching fabric is configured to load balance traffic. The switch fabric includes a plurality of switches. A packet is received at a first switch of the plurality of switches. The first switch load balances the packet to a particular entity among a plurality of entities. Each of the entities is connected to one of the plurality of switches. The first switch determines a particular switch of the plurality of switches to which the packet should be directed, the particular entity being connected to the particular switch of the plurality of switches. The particular switch receives the packet, and determines which interface of the particular switch to direct the packet to the particular entity. The plurality of entities include servers and network appliances as physical devices or virtual machines.

Switch fabric based load balancing

Growing community of inventors

San Jose, CA, United States of America

Samar Sharma