David F Robles

US Patent:

7979686, Jul 12, 2011

Filed:

Feb 23, 2010

Appl. No.:

12/710839

Inventors:

David Morgan Robles - Alameda CA, US

Assignee:

NetApp, Inc. - Sunnyvale CA

International Classification:

G06F 9/46

US Classification:

712244, 718108, 718107

Abstract:

A multiplexed hierarchical array of interrupt controllers is configured to enable low latency task switching of a processor. The hierarchical array comprises a plurality of interrupt controllers coupled to a root interrupt controller. For each task that the processor is configured to execute, a corresponding interrupt controller is provided. To switch the processor to a task, the corresponding interrupt controller signals the root interrupt controller which, in turn, sends an interrupt and a Task Identifier to the processor. The root interrupt controller also cooperates with an access multiplexer/demultiplexer to select the corresponding interrupt controller for communication with the processor. By providing interrupt controller selection as well as task identification, the hierarchical array offloads arbitration and context switching overhead from the processor. That is, in response to the interrupt, the processor switches to the identified task and may access a memory address space dedicated to the task.

US Patent:

7984306, Jul 19, 2011

Filed:

Nov 28, 2006

Appl. No.:

11/605159

Inventors:

David Morgan Robles - Alameda CA, US

Assignee:

NetApp, Inc. - Sunnyvale CA

International Classification:

G06F 11/30

US Classification:

713189, 713193, 726 30, 726 13, 380269, 380 45

Abstract:

A system and method enables efficient transfer and cryptographic processing of data within a security appliance. A network adapter of the security appliance initiates a random access data transfer to a system memory that is illustratively organized into a plurality of blocks of predetermined size. Mapping circuitry of the appliance is configured to track the progress of the data transfer between the adapter and memory such that, once data is stored in a memory block, an encryption processor of the appliance may process the data or the network adapter may retrieve the processed data without having to wait for completion of the data transfer between the adapter and memory, thereby reducing processing latency in the security appliance.

US Patent:

7698541, Apr 13, 2010

Filed:

Oct 31, 2006

Appl. No.:

11/590205

Inventors:

David Morgan Robles - Alameda CA, US

Assignee:

NetApp, Inc. - Sunnyvale CA

International Classification:

G06F 9/46

US Classification:

712244, 718108, 718107

Abstract:

A multiplexed hierarchical array of interrupt controllers is configured to enable low latency task switching of a processor. The hierarchical array comprises a plurality of interrupt controllers coupled to a root interrupt controller. For each task that the processor is configured to execute, a corresponding interrupt controller is provided. To switch the processor to a task, the corresponding interrupt controller signals the root interrupt controller which, in turn, sends an interrupt and a Task Identifier to the processor. The root interrupt controller also cooperates with an access multiplexer/demultiplexer to select the corresponding interrupt controller for communication with the processor. By providing interrupt controller selection as well as task identification, the hierarchical array offloads arbitration and context switching overhead from the processor. That is, in response to the interrupt, the processor switches to the identified task and may access a memory address space dedicated to the task.

US Patent:

20150370600, Dec 24, 2015

Filed:

Aug 28, 2015

Appl. No.:

14/839257

Inventors:

- Sunnyvale CA, US
David Morgan Robles - Alameda CA, US

International Classification:

G06F 9/48
G06F 21/60

Abstract:

A system and method for prioritized queues is provided. A plurality of queues are organized to enable long-running operations to be directed to a long running queue operation, while faster operations are directed to a non-long running operation queue. When an operation request is received, a determination is made whether it is a long-running operation, and, if so, the operation is placed in a long-running operation queue. When the processor core that is executing long-running operations is ready for the next operation, it removes an operation from the long-running operation queue and processes the operation.