Michael L Wong

US Patent:

6721796, Apr 13, 2004

Filed:

Jul 22, 1999

Appl. No.:

09/358869

Inventors:

Michael K. Wong - San Mateo CA

Assignee:

Cisco Technology, Inc. - San Jose CA

International Classification:

G06F 1516

US Classification:

709232, 709234, 709235, 3703954, 3703957, 370412, 370413, 370415, 370417

Abstract:

Multi-level buffer system dynamically allocates storage for data units arriving at network gateway, and retrieves stored data units according to hierarchical schedule. Minimum and maximum thresholds associated with system resource and storage availability determine acceptance and storage of data units. Data units are accepted preferably if all threshold criteria are met. Threshold criteria may be determined from reserved minimum buffer length, calculated maximum buffer length, or random early discard-type algorithm applied separately to each buffer level. Optionally, buffer management applies to non-hierarchical storage systems.

US Patent:

6754216, Jun 22, 2004

Filed:

May 8, 2000

Appl. No.:

09/566602

Inventors:

Michael Wong - Fremont CA
Rajesh Nair - San Jose CA
Milan Momirov - San Francisco CA

Assignee:

Nortel Networks Limited - St. Laurent

International Classification:

H04L 1228

US Classification:

37039542, 3703957

Abstract:

A system enables a cell of data to be transmitted one time over a high speed data bus to an Ethernet switch system where it is then distributed to each of the destinations for which it is intended. A first group of buffers is for temporarily storing data that is to be delivered to only one destination. A second group of buffers is for holding the cells that are to be transmitted to a plurality of devices. In the described embodiment of the invention, the unicast and multicast cells are transmitted over the same line or bus. The invention further includes a switch processor that is formed to detect congestion within the switch fabric and to transmit a congestion rating to each device coupled to transmit and receive data through the switch fabric.

US Patent:

6865155, Mar 8, 2005

Filed:

May 8, 2000

Appl. No.:

09/566604

Inventors:

Michael Wong - Fremont CA, US
Rajesh Nair - San Jose CA, US
Milan Momirov - San Francisco CA, US

Assignee:

Nortel Networks Ltd. - St. Laurent

International Classification:

G08C015/00

US Classification:

370235, 370418

Abstract:

A switching system includes a switch fabric that includes groups of buffers for each port interface device that communicates with external communication devices through an Ethernet plane. The port interface device is formed to receive congestion ratings reflecting buffer usage within the switch fabric for each device coupled to transmit and receive packets through the switching fabric and to determine whether to route to the switching fabric or drop the packet based upon the congestion rating of the destination device and upon a quality of service rating for the packet.

US Patent:

6901491, May 31, 2005

Filed:

Oct 16, 2002

Appl. No.:

10/272784

Inventors:

Leslie D. Kohn - Fremont CA, US
Michael K. Wong - San Mateo CA, US

Assignee:

Sun Microsystems, Inc. - Santa Clara CA

International Classification:

G06F012/16
G06F013/00

US Classification:

711157, 711 5, 712 10, 712 11, 712 14, 714 11, 709212, 709214, 709216

Abstract:

In one embodiment, a server is provided. The server includes multiple application processor chips. Each of the multiple application processor chips includes multiple processing cores. Multiple memories corresponding to the multiple processor chips are included. The multiple memories are configured such that one processor chip is associated with one memory. A plurality of fabric chips enabling each of the multiple application processor chips to access any of the multiple memories are included. The data associated with one of the multiple application processor chips is stored across each of the multiple memories. In one embodiment, the application processor chips include a remote direct memory access (RDMA) and striping engine. The RDMA and striping engine is configured to store data in a striped manner across the multiple memories. A method for allowing multiple processors to exchange information through horizontal scaling is also provided.

US Patent:

6938119, Aug 30, 2005

Filed:

Oct 18, 2002

Appl. No.:

10/273806

Inventors:

Leslie D. Kohn - Fremont CA, US
Kunle A. Olukotun - Stanford CA, US
Michael K. Wong - San Mateo CA, US

Assignee:

Sun Microsystems, Inc. - Santa Clara CA

International Classification:

G06F012/00

US Classification:

711105, 711154, 711156, 713320, 713321, 713322, 713323, 713330, 713340, 713400, 713401, 713500, 713501

Abstract:

A system and method for limiting power consumption of a computer memory system. The system and method includes selecting a memory access rate. The selected memory access rate corresponds to a desired average memory power consumption rate. A first time interval is started as a current time interval. A memory system is accessed. If the memory access rate has not been exceeded, then the access is applied to the memory system. Alternatively, if the memory access rate has been exceeded, then the access is delayed until the current time interval has expired and a subsequent time interval is started as the current time interval and the access is applied to the memory system.

US Patent:

6963563, Nov 8, 2005

Filed:

May 8, 2000

Appl. No.:

09/566603

Inventors:

Michael Wong - Fremont CA, US
Rajesh Nair - San Jose CA, US
Milan Momirov - San Francisco CA, US

Assignee:

Nortel Networks Limited - St. Laurent

International Classification:

H04L012/28

US Classification:

370389, 3703957, 370428, 370412, 370446

Abstract:

The present apparatus and method of use comprises a system that enables a cell of data to be transmitted one time over a high speed data bus to an switch system where it is then distributed to each of the destinations for which it is intended. A fabric access device and a multiplex devices are each formed to have groups of buffers for buffering signals according to type. The first group of buffers is for temporarily storing data that is to be delivered to only one destination. The second group of buffers is for holding the cells that are to be transmitted to a plurality of devices. In those embodiments in which the unicast and multicast cells are transmitted over the same line or bus a parsing unit examines a field within the header portion of each cell to determine whether the cell is a unicast or multicast cell. If the cell is a unicast cell, it is temporarily stored within the unicast receive buffer set. If the cell is a multicast cell, then it is temporarily stored in the multicast buffer set.

US Patent:

7209996, Apr 24, 2007

Filed:

Oct 16, 2002

Appl. No.:

10/272786

Inventors:

Leslie D. Kohn - Fremont CA, US
Kunle A. Olukotun - Stanford CA, US
Michael K. Wong - San Mateo CA, US

Assignee:

Sun Microsystems, Inc. - Santa Clara CA

International Classification:

G06F 13/16

US Classification:

710305, 711119, 710306, 710317, 710 52

Abstract:

In one embodiment, a processor is provided. The processor includes at least two cores, where each of the cores include a first level cache memory. Each of the cores are multi-threaded. In another embodiment, each of the cores includes four threads. In another embodiment a crossbar is included. A plurality of cache bank memories in communication with the at cores through the crossbar is provided. Each of the plurality of cache bank memories are in communication with a main memory interface. In another embodiment a buffer switch core in communication with each of the plurality of cache bank memories is also included. A server and a method for optimizing the utilization of a multithreaded processor core are also provided.

US Patent:

7248585, Jul 24, 2007

Filed:

Oct 16, 2002

Appl. No.:

10/272783

Inventors:

Leslie D. Kohn - Fremont CA, US
Michael K. Wong - San Mateo CA, US

Assignee:

Sun Microsystems, Inc. - Santa Clara CA

International Classification:

H04L 12/56

US Classification:

370392

Abstract:

In one embodiment, a method for efficiently classifying packets for a multi-processor/mutli-thread environment is provided. The method initiates with receiving a packet. Then, header information is extracted form the received packet. Next, a first hash value is calculated. Then, a field of interest in a lookup table is determined from the first hash value. Next, a second hash value is calculated. Then, the second hash value is compared to stored hash values in the field of interest of the lookup table to determine a match between the second hash value and one of the values in the field of interest of the lookup table. If there is a match, the received packet is transmitted to a processor corresponding to the one of the values in the row location of the lookup table. A network interface card and a system for efficiently classifying packets in a multicore/multithread environment are also provided.