Huawei Technologies Co., Limited

Future Wei Technologies, Inc.

Integrated Device Technology, Inc.

Pmc-sierra Us, Inc.

Zixiong Wang

Mohammad Shahanshah Akhter

Gregory Edward Lund

David Clifton Bond

Xiaozhong Wang

Wenyong Zhang

Yuanming Zheng

Alan Coady

A method for performing a remote direct memory access (RDMA) read command includes reserving a memory space in a buffer to receive data associated with an address range, sending a RDMA read instruction to a remote module to request the data, and receiving the data from the remote module. A method for performing a RDMA write command includes receiving from a remote module a request to send command requesting to send data associated with an address range, reserving a memory space in a buffer to receive data associated with the address range, sending a RDMA read instruction to the remote module, and receiving the data from the remote module.

Methods and apparatuses for remote direct memory access page fault handling

Methods and systems for collecting congestion information at a switch of a leaf-spine network. The switch processes a path finding packet. Congestion information is obtained relating to congestion between ports of the switch. The congestion information is inserted into the payload of the path finding packet. The switch multicasts the path finding packet.

Methods and systems for data center load balancing

The method of creating a key entry includes inserting a routing instance identifier (RII) after at least a portion of a key entry of a routing instance (RI) FIB, in accordance with an encoding scheme. In other words, at least a portion of bits of the RI FIB key entry is located before bit(s) of the RII in the resulting, merged FIB key entry. Depending on the encoding scheme, the RII can be inserted at the end of the RI FIB key entry, or at an intermediary location within the RI FIB key entry (after a given number of bits). To form the merged FIB, the method is repeated multiple times on corresponding key entries of the RI FIB. There is also provided a method of creating a search key to lookup the merged FIB.

Method of forwarding data packets, method of creating merged FIB key entry and method of creating a search key

The computer can include a memory system having a plurality of memory cells readable and writable by the processing unit and including a least a first group of memory cells of a same speed grade. A plurality of copy regions each having a corresponding portion of the memory cells of the first group, and a distinct combination of copy unit and copy factor, the copy unit corresponding to a given amount of memory cells. The processing unit can be configured to obtain an indication to copy a data structure stored in the memory system; associate the data structure to one of the copy regions based on the corresponding combination of copy unit and copy factor; copy the data structure to the associated copy region in a number of copies equal to the corresponding copy factor; and successively access different ones of the copies of the data structure.

Computer memory management method and system

An apparatus comprising a ternary content-addressable memory (TCAM) configured to store a root node for each of a plurality of tries resulting in a plurality of root nodes, and at least one non-TCAM memory configured to store associated data that identifies each of the plurality of root nodes, and store a plurality of network address prefixes represented by one or more subtries which belong to a trie in the plurality of tries, wherein each subtrie is locatable at least partially based on the associated data for the root node.

Internet protocol and Ethernet lookup via a unified hashed trie

A trie comprising a plurality of subtries may be balanced by storing, in a first memory stage, a first root that identifies a first subtrie of a trie and a second root that identifies a second subtrie, which is a direct or indirect child of the first subtrie. A plurality of network address prefixes representing vertexes in the plurality of subtries may be stored in at least one additional memory stage. As the first subtrie is located on a top subtrie level which may contain relatively fewer network address prefixes, promoting the second subtrie to the top subtrie level may help improve memory utilization. Further, looking up any received network address may have less memory access latency.

Trie stage balancing for network address lookup

A method and apparatus for a configurable packet routing, buffering and scheduling scheme to optimize throughput with deadlock prevention in SRIO-to-PCIe Bridges have been described. A single level enqueue method with dynamic buffering and dequeuing based on packet re-ordering is disclosed. Single level packet routing and scheduling to meet SRIO and PCIe rules to enqueue packets based on FType/TType is disclosed. Backpressure based on ingress watermarks for different packet types is disclosed. Use of a circular-reorder queue (CRQ) for both ingress and egress allows packet reordering and packet passing.

Method and apparatus for a configurable packet routing, buffering and scheduling scheme to optimize throughput with deadlock prevention in SRIO-to-PCIe bridges

A method of packet classification implemented by a network router, the method comprising obtaining a packet whose header comprises a plurality of dimensional fields specified by a set of rules, matching one or more bits in at least two of the dimensional fields with a corresponding entry stored in a ternary content-addressable memory (TCAM), generating a hash key based on data associated with the corresponding entry in the TCAM, wherein the associated data is stored in at least one non-TCAM memory, searching, using the hash key, a multi-dimensional trie stored in the at least one non-TCAM memory, and mapping the packet to one or more of the set of rules based on the search.

Ternary content-addressable memory assisted packet classification

One embodiment of the present invention provides a router. The router includes a storage, a forwarding table management mechanism, and a longest prefix matching mechanism. The storage stores, in a data structure corresponding to a rich prefix node associated with a trie, a plurality of network address prefixes represented by the trie. The forwarding table management mechanism expresses a respective network address prefix using one or more of: 1) a traversal path on the trie from a root node to a subtrie of the trie; and 2) an encoded representation of a node in the subtrie. The longest prefix matching mechanism matches a network address with a longest prefix from the network address prefixes stored in the data structure corresponding to the prefix node.

Optimized trie-based address lookup

Delay measurement and delay calibration methods and apparatus are described for use within distributed wireless base stations employing a remote radio head topology. The methods and apparatus are usable in any system that requires accurate delay measurement and/or constant delay through an electronic device. The methods and apparatus for measuring delay embody a highly accurate distributed delay measurement architecture that handles multiple delay paths within distributed wireless base stations employing a remote radio head topology. The method and apparatus are amenable to implementation with current integrated circuit technology. The methods and apparatus for calibrating electronic delay within distributed base stations employing a remote radio head topology are useful for implementing distributed wireless base stations where transmit diversity is desired. Using the methods disclosed herein, delay within a distributed wireless base station can be measured and calibrated to achieve very deterministic delay characteristics at the system level.

Growing community of inventors

Ottawa, Canada

Zixiong Wang