公开(公告)号：US10715174B1

公开(公告)日：2020-07-14

申请号：US16751737

申请日：2020-01-24

Applicant: International Business Machines Corporation

Inventor： Bulent Abali ,  Ashutosh Misra ,  Suneel Pusarla

IPC: H03M7/30 ,  H03M7/40 ,  G06F5/01

Abstract: Embodiments of the invention are directed to a DEFLATE compression accelerator and to a method for reducing a latch count required for symbol sorting when generating a dynamic Huffman table. The accelerator includes an input buffer and a Lempel-Ziv 77 (LZ77) compressor communicatively coupled to an output of the input buffer. The accelerator further includes a Huffman encoder communicatively coupled to the LZ77 compressor. The Huffman encoder includes a bit translator. The accelerator further includes an output buffer communicatively coupled to the Huffman encoder.

公开(公告)号：US10585626B2

公开(公告)日：2020-03-10

申请号：US15834133

申请日：2017-12-07

Applicant: International Business Machines Corporation

Inventor： Jonathan Bradbury ,  Matthias Klein ,  Ashutosh Misra ,  Anthony Sofia

IPC: G06F3/06 ,  H03M7/30

Abstract: A system and method to manage a non-universal encoder and a universal encoder for compression of data include receiving the data. The data includes symbols. The method also includes subdividing the data into a first set of data blocks and a second set of data blocks and generating a non-universal encoder using the first set of data blocks. The non-universal encoder includes first codes. Each of the first codes corresponds to one of the symbols in the first set of data blocks only and at least one of the first codes includes fewer bits than the symbol corresponding to the at least one of the first codes. The method further includes compressing the second set of data blocks using at least the non-universal encoder.

公开(公告)号：US10033778B2

公开(公告)日：2018-07-24

申请号：US14868495

申请日：2015-09-29

Applicant: International Business Machines Corporation

Inventor： Suchismita Banerjee ,  Girish G. Kurup ,  Ashutosh Misra ,  Niranjan Vaish

IPC: H04L29/06 ,  H04L12/26 ,  G06F7/02

Abstract: A method for analyzing streaming data includes providing a streaming accumulator comprising an addition module and two multiplexers, receiving one or more data streams, continuously calculating a set of basic statistical elements, receiving a request to calculate a set of statistical descriptors, calculating the set of statistical descriptors, and providing the set of statistical descriptors. An apparatus for analyzing streaming data includes a first multiplexer configured to receive a first summation, a second summation, and a current data item, and forward the first summation on cycles 1 and 3, forward the second summation on cycle 4, and forward the current data item on cycle 2, a second multiplexer configured to receiver the second summation, a third summation, and a previous data item, and forward the previous data item on cycles 1 and 3, forward the second summation on cycle 4, and forward the third summation on cycle 2.

公开(公告)号：US09715564B2

公开(公告)日：2017-07-25

申请号：US14924898

申请日：2015-10-28

Applicant: International Business Machines Corporation

Inventor： Jason R. Baumgartner ,  Raj K. Gajavelly ,  Ashutosh Misra ,  Pradeep K. Nalla

IPC: G06F17/50 ,  G06F9/45

CPC classification number: G06F17/505

Abstract: A method for performing redundancy identification in an integrated circuit design. An optimized gate in a logic circuit is identified. A first netlist with a representation of the logic circuit is generated. An error is induced on the optimized gate. A second netlist is generated from a copy of the first netlist incorporating changes based on the error. Fan-out boundaries of the logic circuit are propagated for the first and second netlists. A redundancy report representing optimization steps performed to obtain the original logic circuit is analyzed to identify which steps are adequate to cause unobservability of the optimized gate. This is done by representing the optimization steps as constraints over the first and second netlists. Responsive to the error becoming undetectable under the constraints derived from the redundancy report, a minimal set of reductions is identified from the first netlist as the reason for unobservability of the optimized gate.

公开(公告)号：US20170093655A1

公开(公告)日：2017-03-30

申请号：US14868495

申请日：2015-09-29

Applicant: International Business Machines Corporation

Inventor： Suchismita Banerjee ,  Girish G. Kurup ,  Ashutosh Misra ,  Niranjan Vaish

IPC: H04L12/26

CPC classification number: H04L65/4069 ,  G06F7/02 ,  H04L43/08 ,  H04L43/50

Abstract: A method for analyzing streaming data includes providing a streaming accumulator comprising an addition module and two multiplexers, receiving one or more data streams, continuously calculating a set of basic statistical elements, receiving a request to calculate a set of statistical descriptors, calculating the set of statistical descriptors, and providing the set of statistical descriptors. An apparatus for analyzing streaming data includes a first multiplexer configured to receive a first summation, a second summation, and a current data item, and forward the first summation on cycles 1 and 3, forward the second summation on cycle 4, and forward the current data item on cycle 2, a second multiplexer configured to receiver the second summation, a third summation, and a previous data item, and forward the previous data item on cycles 1 and 3, forward the second summation on cycle 4, and forward the third summation on cycle 2.

公开(公告)号：US20210374049A1

公开(公告)日：2021-12-02

申请号：US16886943

申请日：2020-05-29

Applicant: INTERNATIONAL BUSINESS MACHINES CORPORATION

Inventor： Deepankar Bhattacharjee ,  Girish Gopala Kurup ,  Ashutosh Misra ,  Puja Sethia

IPC: G06F12/02 ,  G06F12/0855 ,  G06F9/54

Abstract: One or more units of decompressed data of a plurality of units of decompressed data is written to a target location for subsequent writing to memory. The plurality of units of decompressed data includes a plurality of symbol outputs and has associated therewith a plurality of decompression headers. A determination is made that the subsequent writing to memory of at least a portion of another unit of decompressed data to be written to the target location is to be stalled. A symbol start position of the other unit of decompressed data and a decompression header of a selected unit of the one or more units of decompressed data written to the target location are provided to a component of the computing environment. The decompression header is used for the subsequent writing of the other unit of decompressed data to memory.

公开(公告)号：US11119928B2

公开(公告)日：2021-09-14

申请号：US16286861

申请日：2019-02-27

Applicant: International Business Machines Corporation

Inventor： Matthias Klein ,  Ashutosh Misra ,  Girish Gopala Kurup

IPC: G06F12/08 ,  G06F12/0831 ,  G06F13/28 ,  G06F9/38

Abstract: A system architecture is provided and includes an on-chip coherency unit, a processing unit, an accelerator and dedicated wiring. The processing unit is communicative with the on-chip coherency unit via a first interface. The accelerator is communicative with the on-chip coherency unit via a second interface. The accelerator is configured to be receptive of a request to execute lossless data compression or decompression from the processing unit and to responsively execute the lossless data compression or decompression faster than the processing unit. The processing unit and the accelerator are directly communicative by way of the dedicated wiring.

公开(公告)号：US10754781B2

公开(公告)日：2020-08-25

申请号：US15443607

申请日：2017-02-27

Applicant: International Business Machines Corporation

Inventor： James J. Bonanno ,  Michael J. Cadigan, Jr. ,  Adam B. Collura ,  Daniel Lipetz ,  Ashutosh Misra ,  Brian R. Prasky

IPC: G06F9/30 ,  G06F12/0875 ,  G06F12/0897 ,  G06F9/38 ,  G06F11/34 ,  G06F11/30

Abstract: Embodiments are directed to a method for optimizing performance of a microprocessor. The method includes monitoring the performance of the microprocessor in each of a plurality of performance modes. The method further includes choosing a performance mode based on the monitoring. Thereafter, using the performance mode for a predetermined amount of time. Each of the plurality of performance modes is a branch prediction mode.

公开(公告)号：US09697001B2

公开(公告)日：2017-07-04

申请号：US15286752

申请日：2016-10-06

Applicant: International Business Machines Corporation

Inventor： Narasimha R. Adiga ,  James J. Bonanno ,  Ashutosh Misra ,  Anthony Saporito

IPC: G06F9/30 ,  G06F9/38

CPC classification number: G06F9/3806 ,  G06F9/30058 ,  G06F9/30061 ,  G06F9/30145 ,  G06F9/3844 ,  G06F9/3848

Abstract: Embodiments relate to variable branch prediction. An aspect includes determining a branch selection of an execution unit of a processor and determining whether a present prediction state of the state machine correctly predicted the branch selection by the execution unit. The aspect includes determining whether a predetermined condition is met for performing an alternative state transition and, based on determining that the predetermined condition is met, changing the present prediction state of the branch prediction state machine from the one state to another state according to an alternative state transition process based on the branch selection of the execution unit and the determination whether the present prediction state of the state machine correctly predicted the branch selection by the execution unit.

公开(公告)号：US20170093656A1

公开(公告)日：2017-03-30

申请号：US15215635

申请日：2016-07-21

Applicant: International Business Machines Corporation

Inventor： Suchismita Banerjee ,  Girish G. Kurup ,  Ashutosh Misra ,  Niranjan Vaish

IPC: H04L12/26 ,  G06F7/02 ,  H04L29/06

CPC classification number: H04L65/4069 ,  G06F7/02 ,  H04L43/08 ,  H04L43/50

Abstract: A method for analyzing streaming data includes providing a streaming accumulator comprising an addition module and two multiplexers, receiving one or more data streams, continuously calculating a set of basic statistical elements, receiving a request to calculate a set of statistical descriptors, calculating the set of statistical descriptors, and providing the set of statistical descriptors. An apparatus for analyzing streaming data includes a first multiplexer configured to receive a first summation, a second summation, and a current data item, and forward the first summation on cycles 1 and 3, forward the second summation on cycle 4, and forward the current data item on cycle 2, a second multiplexer configured to receiver the second summation, a third summation, and a previous data item, and forward the previous data item on cycles 1 and 3, forward the second summation on cycle 4, and forward the third summation on cycle 2.