Patents by Inventor David Tetzlaff
David Tetzlaff 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: 10331997Abstract: A first input is processed via a first configuration of a neural network to produce a first output. The first configuration defines attributes of the neural network, such as connections between neural elements of the neural network. If the neural network requires a context switch to process a second input, a second configuration is applied to the neural network to change the attributes, and the second input is processed via the second configuration of the neural network to produce a second output.Type: GrantFiled: October 23, 2014Date of Patent: June 25, 2019Assignee: Seagate Technology LLCInventors: Richard Esten Bohn, Peng Li, David Tetzlaff
-
Patent number: 10061747Abstract: A compressed format is selected for storage of a matrix based on a computation to be performed using the matrix and architecture of a storage compute device to which the matrix is stored. Data of the matrix is stored on the storage compute device according to the compressed format. The computation is performed using the data via a computation unit that resides within the storage compute device.Type: GrantFiled: May 5, 2015Date of Patent: August 28, 2018Assignee: Seagate Technology LLCInventors: Peng Li, Richard Esten Bohn, David Tetzlaff
-
Patent number: 10025533Abstract: A logical block address space of a storage compute device is reserved for use in executing commands from a host. First data is received at a first portion of the logical block address space, the first data causing a computation to be performed by the storage compute device. Second data is sent to the host via a second portion of the logical block address space, the second data describing a result of the computation.Type: GrantFiled: January 9, 2017Date of Patent: July 17, 2018Assignee: Seagate Technology LLCInventors: Richard Esten Bohn, Peng Li, David Tetzlaff
-
Patent number: 9976814Abstract: An apparatus includes a heat pipe with a fluid path. A first part of the fluid path is thermally coupled to a first region of a higher temperature and a second part of the fluid path thermally is coupled to a second region of a lower temperature. A difference between the higher temperature and the lower temperature induces a flow of a magnetic fluid in the fluid path. A switchable magnetic device is magnetically coupled to the fluid path. Activation of the switchable magnetic device reduces the flow of the magnetic fluid in the fluid path, which reduces heat transfer from the first region to the second region.Type: GrantFiled: January 27, 2015Date of Patent: May 22, 2018Assignee: Seagate Technology LLCInventors: Kevin Arthur Gomez, Jon Trantham, David Tetzlaff, Mai A. Ghaly
-
Patent number: 9639275Abstract: Commands associated with one or more logical block addresses are received via a host interface of a storage device. Based on a timing and sequence of the commands, an extent of a file that contains the logical block addresses is determined, the file being stored on the storage device. The logical block addresses are managed internally as a unitary data structure based on determining an association between the logical block addresses and the file.Type: GrantFiled: August 6, 2014Date of Patent: May 2, 2017Assignee: SEAGATE TECHNOLOGY LLCInventors: Peng Li, Richard Esten Bohn, David Tetzlaff
-
Publication number: 20170115934Abstract: A logical block address space of a storage compute device is reserved for use in executing commands from a host. First data is received at a first portion of the logical block address space, the first data causing a computation to be performed by the storage compute device. Second data is sent to the host via a second portion of the logical block address space, the second data describing a result of the computation.Type: ApplicationFiled: January 9, 2017Publication date: April 27, 2017Inventors: Richard Esten Bohn, Peng Li, David Tetzlaff
-
Patent number: 9632729Abstract: A data object is received at a storage compute device in response to a request from a host. A requirement of the data object is determined based on a computation to be performed on the data object. The requirement related to at least speed and capacity of media used to store the data object. A tier is selected from the storage compute device based on speed and capacity characteristics of the selected tier corresponding to the requirement of the data object. The data object is stored in the selected tier.Type: GrantFiled: May 5, 2015Date of Patent: April 25, 2017Assignee: SEAGATE TECHNOLOGY LLCInventors: Peng Li, Richard Esten Bohn, David Tetzlaff, Kevin Arthur Gomez
-
Patent number: 9542122Abstract: A logical block address space of a storage compute device is reserved for use in executing commands from a host. The logical block address space is not mapped to a physical address space. First data is received at a first portion of the logical block address space, the first data causing a computation to be performed by the storage compute device. Second data is sent to the host via a second portion of the logical block address space, the second data describing a result of the computation.Type: GrantFiled: October 23, 2014Date of Patent: January 10, 2017Assignee: SEAGATE TECHNOLOGY LLCInventors: Richard Esten Bohn, Peng Li, David Tetzlaff
-
Publication number: 20160216043Abstract: An apparatus includes a heat pipe with a fluid path. A first part of the fluid path is thermally coupled to a first region of a higher temperature and a second part of the fluid path thermally is coupled to a second region of a lower temperature. A difference between the higher temperature and the lower temperature induces a flow of a magnetic fluid in the fluid path. A switchable magnetic device is magnetically coupled to the fluid path. Activation of the switchable magnetic device reduces the flow of the magnetic fluid in the fluid path, which reduces heat transfer from the first region to the second region.Type: ApplicationFiled: January 27, 2015Publication date: July 28, 2016Inventors: Kevin Arthur Gomez, Jon Trantham, David Tetzlaff, Mai A. Ghaly
-
Publication number: 20160117121Abstract: A logical block address space of a storage compute device is reserved for use in executing commands from a host. The logical block address space is not mapped to a physical address space. First data is received at a first portion of the logical block address space, the first data causing a computation to be performed by the storage compute device. Second data is sent to the host via a second portion of the logical block address space, the second data describing a result of the computation.Type: ApplicationFiled: October 23, 2014Publication date: April 28, 2016Inventors: Richard Esten Bohn, Peng Li, David Tetzlaff
-
Publication number: 20160041778Abstract: Commands associated with one or more logical block addresses are received via a host interface of a storage device. Based on a timing and sequence of the commands, an extent of a file that contains the logical block addresses is determined, the file being stored on the storage device. The logical block addresses are managed internally as a unitary data structure based on determining an association between the logical block addresses and the file.Type: ApplicationFiled: August 6, 2014Publication date: February 11, 2016Inventors: Peng Li, Richard Esten Bohn, David Tetzlaff
-
Publication number: 20150324125Abstract: A data object is received at a storage compute device in response to a request from a host. A requirement of the data object is determined based on a computation to be performed on the data object. The requirement related to at least speed and capacity of media used to store the data object. A tier is selected from the storage compute device based on speed and capacity characteristics of the selected tier corresponding to the requirement of the data object. The data object is stored in the selected tier.Type: ApplicationFiled: May 5, 2015Publication date: November 12, 2015Inventors: Peng Li, Richard Esten Bohn, David Tetzlaff, Kevin Arthur Gomez
-
Publication number: 20150326245Abstract: A compressed format is selected for storage of a matrix based on a computation to be performed using the matrix and architecture of a storage compute device to which the matrix is stored. Data of the matrix is stored on the storage compute device according to the compressed format. The computation is performed using the data via a computation unit that resides within the storage compute device.Type: ApplicationFiled: May 5, 2015Publication date: November 12, 2015Inventors: Peng Li, Richard Esten Bohn, David Tetzlaff
-
Publication number: 20150324685Abstract: A first input is processed via a first configuration of a neural network to produce a first output. The first configuration defines attributes of the neural network, such as connections between neural elements of the neural network. If the neural network requires a context switch to process a second input, a second configuration is applied to the neural network to change the attributes, and the second input is processed via the second configuration of the neural network to produce a second output.Type: ApplicationFiled: October 23, 2014Publication date: November 12, 2015Inventors: Richard Esten Bohn, Peng Li, David Tetzlaff
-
Patent number: 8947992Abstract: Combined echo and crosstalk cancellation is provided. Frequency domain adaptive filters are used to remove or reduce the effects of echo and crosstalk for a multi-channel and full-duplex communications system. Data from each transmit channel is buffered and converted to the frequency domain. The frequency domain data is multiplied by crosstalk coefficients to obtain a frequency domain correction signal for each channel. Adaptation of the crosstalk coefficients is based on correlations between the error signals and the data from each of the transmit channels. A single frequency domain transform engine, such as a Fast Fourier Transform engine, is employed for all calculations to save power and area.Type: GrantFiled: July 3, 2013Date of Patent: February 3, 2015Assignee: Vintomie Networks B.V., LLCInventors: James M. Little, Marwan Hassoun, David Tetzlaff, Chang-Chi Liu
-
Publication number: 20140010066Abstract: Combined echo and crosstalk cancellation is provided. Frequency domain adaptive filters are used to remove or reduce the effects of echo and crosstalk for a multi-channel and full-duplex communications system. Data from each transmit channel is buffered and converted to the frequency domain. The frequency domain data is multiplied by crosstalk coefficients to obtain a frequency domain correction signal for each channel. Adaptation of the crosstalk coefficients is based on correlations between the error signals and the data from each of the transmit channels. A single frequency domain transform engine, such as a Fast Fourier Transform engine, is employed for all calculations to save power and area.Type: ApplicationFiled: July 3, 2013Publication date: January 9, 2014Inventors: James M. Little, Marwan Hassoun, David Tetzlaff, Chang-Chi Liu
-
Patent number: 8488438Abstract: Combined echo and crosstalk cancellation is provided. Frequency domain adaptive filters are used to remove or reduce the effects of echo and crosstalk for a multi-channel and full-duplex communications system. Data from each transmit channel is buffered and converted to the frequency domain. The frequency domain data is multiplied by crosstalk coefficients to obtain a frequency domain correction signal for each channel. Adaptation of the crosstalk coefficients is based on correlations between the error signals and the data from each of the transmit channels. A single frequency domain transform engine, such as a Fast Fourier Transform engine, is employed for all calculations to save power and area.Type: GrantFiled: February 11, 2011Date of Patent: July 16, 2013Assignee: Vintomie Networks B.V., LLCInventors: James M. Little, Marwan Hassoun, David Tetzlaff, Chang-Chi Liu
-
Publication number: 20110141875Abstract: Combined echo and crosstalk cancellation is provided. Frequency domain adaptive filters are used to remove or reduce the effects of echo and crosstalk for a multi-channel and full-duplex communications system. Data from each transmit channel is buffered and converted to the frequency domain. The frequency domain data is multiplied by crosstalk coefficients to obtain a frequency domain correction signal for each channel. Adaptation of the crosstalk coefficients is based on correlations between the error signals and the data from each of the transmit channels. A single frequency domain transform engine, such as a Fast Fourier Transform engine, is employed for all calculations to save power and area.Type: ApplicationFiled: February 11, 2011Publication date: June 16, 2011Applicant: Vintomie Networks B.V., LLCInventors: James M. Little, Marwan Hassoun, David Tetzlaff, Chang-Chi Liu
-
Patent number: 7920461Abstract: Combined echo and crosstalk cancellation is provided. Frequency domain adaptive filters are used to remove or reduce the effects of echo and crosstalk for a multi-channel and full-duplex communications system. Data from each transmit channel is buffered and converted to the frequency domain. The frequency domain data is multiplied by crosstalk coefficients to obtain a frequency domain correction signal for each channel. Adaptation of the crosstalk coefficients is based on correlations between the error signals and the data from each of the transmit channels. A single frequency domain transform engine, such as a Fast Fourier Transform engine, is employed for all calculations to save power and area.Type: GrantFiled: September 25, 2008Date of Patent: April 5, 2011Assignee: Vintomie Networks B.V., LLCInventors: James M. Little, Marwan Hassoun, David Tetzlaff, Chang-Chi Liu
-
Publication number: 20090196161Abstract: Combined echo and crosstalk cancellation is provided. Frequency domain adaptive filters are used to remove or reduce the effects of echo and crosstalk for a multi-channel and full-duplex communications system. Data from each transmit channel is buffered and converted to the frequency domain. The frequency domain data is multiplied by crosstalk coefficients to obtain a frequency domain correction signal for each channel. Adaptation of the crosstalk coefficients is based on correlations between the error signals and the data from each of the transmit channels. A single frequency domain transform engine, such as a Fast Fourier Transform engine, is employed for all calculations to save power and area.Type: ApplicationFiled: September 25, 2008Publication date: August 6, 2009Inventors: James M. Little, Marwan Hassoun, David Tetzlaff, Chang-Chi Liu