Cisco Technology, Inc.

Amdahl Corporation

Lsi Corporation

Kent Wayne Wendorf

Hariprasad Ginjpalli

Gary S Goldman

Thinh Hau

Earl T Cohen

Leonid Baryudin

Durai Chinnaiah

Jayakumar Jayakumar

Hariprasada Rao Ginjpalli

Jeffrey L Baker

Man Pak Yip

Ronald N Hilton

John Sandoval

Quang Phung

James Ding

Alfred Platt

Kham X Nguyen

Nicholas Y Pang

Nitin Garg

Jose C Shen

Phong Bui

Simon Kwong

Described embodiments provide a lookup engine that receives lookup requests including a requested key and a speculative add requestor. Iteratively, for each one of the lookup requests, the lookup engine searches each entry of a lookup table for an entry having a key matching the requested key of the lookup request. If the lookup table does not include an entry having a key matching the requested key, the lookup engine sends a miss indication corresponding to the lookup request to the control processor. If the speculative add requestor is set, the lookup engine speculatively adds the requested key to a free entry in the lookup table. Speculatively added keys are searchable in the lookup table for subsequent lookup requests to maintain coherency of the lookup table without creating duplicate key entries, comparing missed keys with each other or stalling the lookup engine to insert missed keys.

Lookup engine with pipelined access, speculative add and lock-in-hit function

A system and method for distributing guaranteed bandwidth among service groups in a network node are described. A position on a time scale of a buffer containing multiple data units is determined. A signal prompting selection of said buffer for release of at least one data unit of the data units is modified based on the position determined on the time scale.

System and method for distributing guaranteed bandwidth among service groups in a network node

A system and method to multicast guaranteed and best-effort traffic in a communications network are disclosed. According to one embodiment, incoming traffic is separated into unicast traffic and multicast traffic. Each root cell of the multicast traffic is classified based on multiple corresponding classes of service. Each root cell is then stored into a root cell buffer of multiple root cell buffers within an egress memory, each root cell buffer being associated with a corresponding class of service. According to one embodiment, each root cell is retrieved from the corresponding root cell buffer within the egress memory according to the associated class of service. Each root cell is then stored in a corresponding replication queue of multiple replication queues based on its associated class of service, with predetermined replication parameters assigned to each replication queue. Each root cell is subsequently replicated according to one or more associated replication parameters to obtain multiple leaf cells for each replicated root cell. According to one embodiment, the unicast traffic is multiplexed with the replicated leaf cells of the multicast traffic to obtain egress arrival cells. Each egress arrival cell is then stored hierarchically into a queuing buffer of multiple queuing buffers within a queuing memory according to queuing parameters corresponding to each egress arrival cell.

System and method to multicast guaranteed and best-effort traffic in a communications network

A system and method for writing the same data field to multiple RAM copies during a single write cycle that fans out write data, address data, and control data to multiple RAMs. The multiple copies of data held at the same address in the multiple RAM copies are also read during a single write cycle and the data from each RAM copy is concatenated into a single word that is read during a single read cycle.

Accessing multiple copies of RAM distributed throughout an ASIC/FPGA and maintaining their content consistency

A method and system for the fair allocation of unused guaranteed bandwidth. Data segments of at least one class of service are received at each of a plurality of ingress line cards. Each class of service has a guaranteed percentage of transmission bandwidth. The system uses a must-serve bit to mark a number of data cells within each class of service. The number of cells marked depends on the guaranteed bandwidth percentage for the particular class of service. The mark is referred to as 'must-serve' since the scheduler must serve the particular class of service to the extent of the marked cells in order to meet the guaranteed bandwidth. The generic switch fabric monitors the cells and the switch CPU reallocates bandwidth so that only the marked cells are provided preferential transmission. Non-marked cells compete equally for excess bandwidth.

In-band must-serve indication from scheduler to switch fabric

A system and method for efficient packetization of cells transmitted over a packet network are described. Multiple virtual connections are received, each virtual connection carrying multiple cells. One or more cells of the cells carried by each virtual connection are stored for a predetermined time period. A cell counter of the stored cells is modified to indicate a number of the stored cells. Finally, a packet containing the stored cells is formed, the packet to be transmitted to a destination node over a network.

System and method for efficient packetization of ATM cells transmitted over a packet network

A system and method for testing the random access memory of a computer system is disclosed. A memory-testing engine is embedded in the utility bus controller of an application specific integrated circuit, which is coupled to a random access memory in need of testing. Upon receiving an initiation signal over a bus from the central processing unit, the memory-testing engine begins writing data to a targeted area of the memory, and then reading back the stored data and comparing the data to what was sent. Having the memory-testing engine distributed to the memory's being tested allows several memory devices to be tested simultaneously.

System and method to asynchronously test RAMs

A system and method for allocating bandwidth in a network node are described. Multiple data units are received along a plurality of virtual connections in a network, each virtual connection having a predetermined unit rate. The data units are subsequently stored. Finally, a data unit of the stored data units having a data unit rate higher than the predetermined unit rate of each virtual connection is transmitted at predetermined time intervals.

System and method for allocating bandwidth in a network node

An interface between a storage unit and a system control unit maintains a sequential processing when retrying linestores by providing a single piece of information that the linestore is a 'first' in a series. An initial request flag accompanies linestore requests to the system controller and is returned to the storage unit with linestore replies. A mode indicator coupled with the storage unit pipeline sets a retry mode latch in response to a linestore restart reply, and resets the retry mode latch in response to a linestore read reply by the system controller to an initial request. Logic in the pipeline suppresses requests for the system resource in the pipeline, other than initial requests, when the retry mode latch is set, such as by flagging data subject of the second flow as locked or otherwise invalid. The port includes a state machine which preserves the order of pending requests for the system resource.

Retry request system in a pipeline data processing system where each

An error handling and reporting mechanism is capable of taking advantage of sophisticated error analysis performed after clocks have been stopped in response to an error detected in a controller. The controller provides services in a data processing system in response to requests for controller services from a plurality of requestors. The controller includes a plurality of ports for storing requests for controller services. A plurality of servers is coupled to the plurality of ports, and perform separate services associated with the requests for controller services stored in the plurality of ports. An error reporting mechanism is included which is responsive to a detected error in a particular server associated with a request in a particular port, for posting error status in the particular port and causing clock stoppage within a clock stop latency period. An error analysis mechanism analyzes the detected errors during the clock stoppage. Error handling logic is coupled with the error analysis mechanism, and is responsive to the posted error status in the ports, for notifying a requestor of an error status posted with a request in the particular port. The error handling logic includes a stall counter, which stalls the error handling mechanism in response to the posted error status for at least the clock latency period so that the clock stoppage occurs and the error analysis mechanism completes error analysis before the requestor is notified.

Growing community of inventors

San Jose, CA, United States of America

Kent Wayne Wendorf