Cisco Technology, Inc.

David Richard Barach

David Delano Ward

Ijsbrand Wijnands

Neale David Raymond Ranns

Mohammed Ismael Tatar

Kenneth Harvey Potter, Jr

Naader Hasani

Jeffery B Scott

John Mitten

Sreedhar Ravipalli

Shishir Gupta

Kent H Hoult

Shahin Habibi

Christopher G Riedle

James Neil Guichard

Hendrikus G P Bosch

Cullen F Jennings

Pradeep Kumar Kathail

Alessandro Duminuco

Humberto Jose La Roche

Peter Bosch

Sape Jurrien Mullender

Robert Goguen

Michael E Lipman

Sape Mullender

Kirk Stewart Lougheed

John A DeNisco

Eliot Michael Dresselhaus

Jim McKie

Philip Winterfield

Bobby Batacharia

Jim Mckie

In one embodiment, an offload platform is an compute platform, adjunct to a router or other packet switching device, that performs packet processing operations including determining an egress forwarding value corresponding to the next-hop node of the packet switching device to which to send an offload-platform processed packet. The offload platform downloads forwarding information from the router, and augments it, such as, but not limited to, representing interfaces of the router as identifiable virtual interface(s) on the offload platform, and including each of one or more next-hop nodes of the router represented as an identifiable virtual adjacency and identifiable tunnel (e.g., identified by the egress forwarding value). In one embodiment, the egress forwarding value is an Multiprotocol Label Switching (MPLS) label or Segment Routing Identifier. The router identifies packets of certain packet flows to send to the adjunct offload platform, rather than processing per its routing information base.

Processing packets by an offload platform adjunct to a packet switching device

One embodiment performs longest prefix matching operations in one or more different manners that provides packet processing and/or memory efficiencies in the processing of packets. In one embodiment, a packet switching device determines a set of one or more mask lengths of a particular conforming entry of a multibit trie or other data structure that matches a particular address of a packet via a lookup operation in a mask length data structure. A conforming entry refers to an entry which has less than or equal to a maximum number of different prefix lengths, with this maximum number corresponding to the maximum number of prefix lengths which can be searched in parallel in the address space for a longest matching prefix by the implementing hardware. The packet switching device then performs corresponding hash table lookup operation(s) in parallel in determining an overall longest matching prefix for the particular address.

Longest prefix matching providing packet processing and/or memory efficiencies in processing of packets

A baseboard management controller (BMC) can physically partition the computing resources of a physical host into different resource groups for concurrently running a different operating system per resource group. The BMC can allocate a first processor of the host to a first resource group and a second processor of the host to a second resource group. The BMC can separate the memory of the host into a first memory range for the first processor and a second memory range for the second processor, and the BMC can limit access to the first memory range to the first processor and limit access to the second memory range to the second processor. The BMC can also distribute physical or virtual peripheral devices of the host between the first processor and the second processor.

Physical partitioning of computing resources for server virtualization

An example method for load balancing in a network environment is provided and includes receiving a packet from a first stage load-balancer in a network environment, where the packet is forwarded from the first stage load-balancer to one of a plurality of second stage load-balancers in the network according to a hash based forwarding scheme, and routing the packet from the second stage load-balancer to one of a plurality of servers in the network according to a per-session routing scheme. The per-session routing scheme includes retrieving a session routing state from a distributed hash table in the network. In a specific embodiment, the hash based forwarding scheme includes equal cost multi path routing. The session routing state can include an association between a next hop for the packet and the packet's 5-tuple representing a session to which the packet belongs.

Hierarchical load balancing in a network environment

Disclosed are, inter alia, methods, apparatus, computer-storage media, mechanisms, and means associated with load balancing across multiple network address translation (NAT) instances and/or processors. N network address translation (NAT) processors and/or instances are each assigned a portion of the source address traffic in order to load balance the network address translation among them. Additionally, the address space of translated addresses is partitioned and uniquely assigned to the NAT processors and/or instances such that the identification of the assigned NAT processor and/or instance associated with a received translated address can be readily determined there from, and then used to network address translate that received packet.

Load balancing across multiple network address translation (NAT) instances and/or processors

An intermediate network node is configured to forward a plurality of packets concurrently, e.g., as a vector, rather than one packet at a time. As such, the node can load a single sequence of forwarding instructions that may be repeatedly executed for packets in the vector. In addition, the intermediate network node adaptively controls the rate at which it processes data packets through a directed forwarding graph. To that end, the intermediate node is configured to dynamically select the number of packets per vector, i.e., vector size, processed at each node of the forwarding graph. Further, the intermediate node also may be configured to dynamically select timing intervals for one or more 'node timers' used to control the rate at which packets traverse through the graph. Illustratively, the vector size and node-timer intervals are selected so that the average latency through the forwarding graph is less than a predetermined target latency, e.g., 50 microseconds (&#x3bc;s).

Vectorized software packet forwarding

A router that includes a plurality of processors (SMPs) where there is 'affinity' between particular processors and particular interfaces: Each of the router's interfaces are assigned to one of the processors. A packet arriving at a particular interface will be handled by the processor having an affinity to that particular interface. If the packet's egress is on an interface assigned to the same processor, then the output process will also be handled by that processor. If the egress interface has an affinity to a different processor, then the packet is handed over to the other processor for egress. The data structures that must be retrieved from memory to handle a packet are often associated with the interfaces through which the packet passes. Thus, having a particular processor handle all the packets that pass through a particular interface insures that the data structures needed to handle the packets will more likely be stored in the processor's cache and less likely be the object of inter-processor lock contention.

Accelerated data switching on symmetric multiprocessor systems using port affinity

The present invention provides a technique for efficiently looking up address-routing information in an intermediate network node, such as a router. To that end, the node locates routing information stored in its memory using one or more 'lookup' tables (LUT) which can be searched using a small, bounded number of dependent lookups, thereby reducing the number of dependent lookups conventionally performed. The LUTs are arranged so each table provides routing information for network addresses whose subnet mask lengths are within a different range ('stride') of mask lengths. According to the technique, the node locates a network address's routing information by searching the LUTs, in order of decreasing prefix lengths, until the routing information is found. Preferably, several tables are searched in parallel. A match in a LUT may further point to a small MTRIE that enables the final bits of a prefix to be matched. That final MTRIE is searched using a relatively small, bounded number of dependent lookups.

Bounded index extensible hash-based IPv6 address lookup method

An intermediate network node is configured to drop or reject new client sessions when its available resources are below a predetermined level. In this manner, the intermediate node can efficiently process a large number of new session attempts at substantially the same time. The intermediate node monitors the availability of its resources by calculating a load metric. The load metric is based on one or more partial load metrics, each corresponding to a different system resource. The load metric is compared with a predetermined value to determine whether the node has enough available resources to continue establishing new client sessions. Alternatively, the intermediate node rejects new client sessions when a total number of allocated 'abstract resource units' rises above a predetermined level. That is, client sessions are assigned a predetermined number of abstract resource units on a per-protocol basis and a resource counter stores the number of abstract resource units allocated by the intermediate node. Each protocol is assigned a different time interval for a timing wheel, after which time the number of abstract resource units assigned for the protocol is subtracted from the resource counter. The intermediate node actively or passively rejects new client sessions until the timing wheel sufficiently decreases the counter below the predetermined level.

Flexible WAN protocol call admission control algorithm

A data path protocol eliminates most of the conventional read transactions required to transfer data between devices interconnected by a split transaction bus, such as a HyperTransport (HPT) bus. To that end, each device is configured to manage its own set of buffer descriptors, unlike previous data path protocols in which only one device managed all the buffer descriptors. As such, neither device has to perform a read transaction to retrieve a 'free' buffer descriptor from the other device. As a result, only write transactions are performed for transferring descriptors across the HPT bus, thereby decreasing the amount of traffic over the bus and eliminating conventional latencies associated with read transactions. In addition, because descriptors are separately managed in each device, the data path protocol also conserves processing bandwidth that is traditionally consumed by managing ownership of the buffer descriptors within a single device.

Hypertransport data path protocol

A buffer-management technique efficiently manages a set of data buffers accessible to first and second devices interconnected by a split transaction bus, such as a Hyper-Transport (HPT) bus. To that end, a buffer manager controls access to a set of 'free' buffer descriptors, each free buffer descriptor referencing a corresponding buffer in the set of data buffers. Advantageously, the buffer manager ensures that the first and second devices are allocated a sufficient number of free buffer descriptors for use in a HPT data path protocol in which the first and second devices have access to respective sets of free buffer descriptors. Because buffer management over the HPT bus is optimized by the buffer manager, the amount of processing bandwidth traditionally consumed managing descriptors can be reduced.

Buffer management technique for a hypertransport data path protocol

A system controller on a half-duplex communication bus architecture gathers statistical data from at least one line card. The line card automatically collects statistical data for the system controller in a controlled manner in an information buffer. Then, the line card reports a subset of the statistical data in the buffer to the system controller when polled either for the statistical data or for some other reason and the statistical data is selected to be reported. The line card reports the statistical data in place of reporting a null response in the preferred embodiment whenever possible.

Growing community of inventors

Boxborough, MA, United States of America

David Richard Barach