Patents by Inventor Frederick R. Gruner
Frederick R. Gruner has filed for patents to protect the following inventions. This listing includes patent applications that are pending as well as patents that have already been granted by the United States Patent and Trademark Office (USPTO).
-
Patent number: 9294085Abstract: A low-inductance, air-insulated gas switch uses a de-enhanced annular trigger ring disposed between two opposing high voltage electrodes. The switch is DC chargeable to 200 kilovolts or more, triggerable, has low jitter (5 ns or less), has pre-fire and no-fire rates of no more than one in 10,000 shots, and has a lifetime of greater than 100,000 shots. Importantly, the switch also has a low inductance (less than 60 nH) and the ability to conduct currents with less than 100 ns rise times. The switch can be used with linear transformer drives or other pulsed-power systems.Type: GrantFiled: December 6, 2013Date of Patent: March 22, 2016Assignee: Sandia CorporationInventors: Frederick R. Gruner, William A. Stygar
-
Patent number: 8424012Abstract: A method for context switching on a video processor having a scalar execution unit and a vector execution unit. The method includes executing a first task and a second task on a vector execution unit. The first task in the second task can be from different respective contexts. The first task and the second task are each allocated to the vector execution unit from a scalar execution unit. The first task and the second task each comprise a plurality of work packages. In response to a switch notification, a work package boundary of the first task is designated. A context switch from the first task to the second task is then executed on the work package boundary.Type: GrantFiled: November 4, 2005Date of Patent: April 16, 2013Assignee: Nvidia CorporationInventors: Ashish Karandikar, Shirish Gadre, Frederick R. Gruner, Franciscus W. Sijstermans
-
Patent number: 7934198Abstract: A prefix matching apparatus for directing information to a destination port includes a memory configured to store a piece of data including an address and a plurality of levels each including a plurality of memory locations, the levels each representing a unique address space. A controller is coupled to the memory and to the plurality of levels, and is configured to read the data address and to direct the data to the next level associated with a unique address space associated with the data address. In one embodiment, the controller is configured to match the data address prefix to a plurality of addresses associated with the unique address spaces. Advantages of the invention include fast switch decisions and low switch latency.Type: GrantFiled: October 18, 2004Date of Patent: April 26, 2011Assignee: NetLogic Microsystems, Inc.Inventors: Frederick R. Gruner, Gaurav Singh, Elango Ganesan, Samir C. Vora, Christopher M. Eccles, Brian Hang Wai Yang
-
Patent number: 7562196Abstract: A precedence determination system including a first type memory bank configured to receive a first search signal and to provide first search result indications, a second type memory bank configured to receive a second search signal and to provide second search result indications, a precedence number table coupled to the first and second type memory banks and configured to provide programmable precedence numbers, and a precedence determination circuit coupled to the first and second type memory banks and the precedence number table and configured to provide a third search result indication is disclosed. In one embodiment, the first type memory bank can be a static random access memory (SRAM) and the second type memory bank can be a ternary content addressable memory (TCAM).Type: GrantFiled: March 23, 2007Date of Patent: July 14, 2009Assignee: RMI CorporationInventors: Sophia W. Kao, Puneet Agarwal, Frederick R. Gruner
-
Patent number: 7213111Abstract: A precedence determination system including a first type memory bank configured to receive a first search signal and to provide first search result indications, a second type memory bank configured to receive a second search signal and to provide second search result indications, a precedence number table coupled to the first and second type memory banks and configured to provide programmable precedence numbers, and a precedence determination circuit coupled to the first and second type memory banks and the precedence number table and configured to provide a third search result indication is disclosed. In one embodiment, the first type memory bank can be a static random access memory (SRAM) and the second type memory bank can be a ternary content addressable memory (TCAM).Type: GrantFiled: February 27, 2004Date of Patent: May 1, 2007Assignee: Raza Microelectronics, Inc.Inventors: Sophia W. Kao, Puneet Agarwal, Frederick R. Gruner
-
Patent number: 7174441Abstract: A configurable lookup table extension system including a plurality of lookup tables arranged in an internal memory, an external memory, and a flexible controller configured to couple at least one of the plurality of lookup tables to the external memory through a single memory interface is disclosed. Implementations of this system can support the flexible allocation of IP and MAC table entries so that a router/switch can flexibly support applications suited to a particular allocation. This approach provides an efficient scheme for extending multiple internal tables to external memory via a single external interface. Further, such extensibility is also programable to allow the size and number of external tables to be configured by software. This solution can provide the flexibility of customizing table sizes for different markets and/or customer requirements.Type: GrantFiled: October 17, 2003Date of Patent: February 6, 2007Assignee: Raza Microelectronics, Inc.Inventors: Gaurav Singh, Frederick R. Gruner, Brian Hang Wai Yang
-
Patent number: 6598154Abstract: A method of reducing the branch penalty in a microprocessor includes predecoding the instruction to determine whether an instruction is a branch, the length of the instruction, and prediction marker information for the instruction should it be a branch. The target of the branch is relayed to the align stage of the microprocessor to readjust the read pointer to point to the target of the branch if the instruction is a branch. An apparatus for reducing the branch penalty in a microprocessor includes a branch predecode and taken resolution unit which determines whether an instruction is a predicted taken branch, and relays that information to the align stage of the microprocessor to deliver the target of the branch to the align stage as early as possible.Type: GrantFiled: December 29, 1998Date of Patent: July 22, 2003Assignee: Intel CorporationInventors: Kushagra Vaid, Frederick R. Gruner
-
Patent number: 6066959Abstract: A logic array includes an AND plane, a first OR plane, and a second OR plane. The AND plane is adapted to receive a plurality of logic array inputs and provide a plurality of minterms. Each minterm represents a logical combination of a subset of the plurality of logic array inputs. The first OR plane is adapted to receive the minterms and provide a plurality of intermediate outputs. Each intermediate output represents a logical combination of a subset of the minterms. The second OR plane is adapted to receive the intermediate outputs and provide a plurality of logic array outputs. Each logic array output represents a logical combination of a subset of the intermediate outputs. A method for programming a logic array includes providing a plurality of minterms. A plurality of subsets of the minterms are logically combined to define a plurality of intermediate outputs. A plurality of subsets of the intermediate outputs are logically combined to define a plurality of logic array outputs.Type: GrantFiled: December 9, 1997Date of Patent: May 23, 2000Assignee: Intel CorporationInventors: Frederick R. Gruner, Ralph Portillo
-
Patent number: 5977794Abstract: A logic array includes a first logic plane, a second logic plane, and a third logic plane. The first logic plane has a first plurality of intermediate outputs, and the second logic plane has a second plurality of intermediate outputs. The third logic plane has first and second opposing sides and is adapted to receive the first and second pluralities of intermediate outputs. The first plurality of intermediate outputs intersect the third logic plane through the first side, and the second plurality of intermediate outputs intersect the third logic plane through the second side. A method for increasing the density of a logic array includes providing a first logic plane, a second logic plane, and a third logic plane. The first logic plane has a first plurality of intermediate outputs, and the second logic plane has a second plurality of intermediate outputs. The first plurality of intermediate outputs and the second plurality of intermediate outputs are interleaved in the third logic plane.Type: GrantFiled: December 9, 1997Date of Patent: November 2, 1999Assignee: Intel CorporationInventors: Frederick R. Gruner, Ralph Portillo