Patents by Inventor Avinash N. Ananthakrishnan

Avinash N. Ananthakrishnan 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).

  • Publication number: 20180059757
    Abstract: In one embodiment, a processor includes: at least one core to execute instructions; a power controller to control power consumption of the processor; and a storage to store a plurality of entries to associate a dynamic capacitance with a time duration for which a current spike is to be exposed to a power delivery component. Other embodiments are described and claimed.
    Type: Application
    Filed: August 31, 2016
    Publication date: March 1, 2018
    Inventors: Avinash N. Ananthakrishnan, Jorge P. Rodriguez
  • Publication number: 20180060123
    Abstract: In one embodiment, a processor includes: a first storage to store a set of common performance state request settings; a second storage to store a set of thread performance state request settings; and a controller to control a performance state of a first core based on a combination of at least one of the set of common performance state request settings and at least one of the set of thread performance state request settings. Other embodiments are described and claimed.
    Type: Application
    Filed: August 31, 2016
    Publication date: March 1, 2018
    Inventors: Eliezer Weissmann, Israel Hirsh, Efraim Rotem, Doron Rajwan, Avinash N. Ananthakrishnan, Natanel Abitan, Ido Melamed, Guy M. Therien
  • Publication number: 20180060085
    Abstract: In one embodiment, a processor includes a plurality of cores and a power controller. This power controller in turn may include a voltage ramp logic to pre-empt a voltage ramp of a voltage regulator from a first voltage to a second voltage, responsive to a request for a second core to exit a low power state. Other embodiments are described and claimed.
    Type: Application
    Filed: August 31, 2016
    Publication date: March 1, 2018
    Inventors: Avinash N. Ananthakrishnan, Jeremy J. Shrall, Anupama Suryanarayanan, Ameya Ambardekar, Craig Topper, Eric R. Heit, Joseph M. Alberts
  • Patent number: 9760409
    Abstract: In one embodiment, the present invention includes a multicore processor having a power controller with logic to dynamically switch a power management policy from a power biased policy to a performance biased policy when a utilization of the processor exceeds a threshold level. Thus at low utilizations, reduced power consumption can be realized, while at higher utilizations, greater performance can be realized. Other embodiments are described and claimed.
    Type: Grant
    Filed: May 24, 2016
    Date of Patent: September 12, 2017
    Assignee: Intel Corporation
    Inventors: Krishnakanth V. Sistla, Mark Rowland, Ankush Varma, Ian M. Steiner, Matthew Bace, Daniel Borkowski, Vivek Garg, Chelsea Akturan, Avinash N. Ananthakrishnan
  • Patent number: 9753516
    Abstract: According to one embodiment of the invention, an integrated circuit device comprises one or more processor cores and a control unit coupled to the processor core(s). The control unit is adapted to control an operating frequency of at least one processor core based on an estimated activity level in lieu of a power level. The estimated activity level differing from an estimated power level by being independent of leakage power and voltage characteristics particular to that integrated circuit device.
    Type: Grant
    Filed: December 22, 2011
    Date of Patent: September 5, 2017
    Assignee: Intel Corporation
    Inventors: Ryan D. Wells, Avinash N. Ananthakrishnan, Efraim Rotem, Doron Rajwan, Gary A. Andrew
  • Patent number: 9734116
    Abstract: Techniques and mechanisms for configuring an integrated circuit (IC) chip to implement a protocol stack. In an embodiment, a transaction layer of the IC chip is operable to exchange with a link layer of the IC chip transaction layer packets (TLPs) having a format compatible with one defined in a Peripheral Component Interconnect Express™ (PCIe™) specification. Configuration circuitry of the IC chip provides for configuration of a first protocol stack including the transaction layer, circuitry of the link layer and a first physical layer of the IC chip. The configuration circuitry further provides for an alternative configuration of a second protocol stack including the transaction layer, circuitry of the link layer and a second physical layer of the IC chip. In another embodiment, the first protocol stack supports single-ended signaling to communicate TLP information, whereas the second protocol stack supports differential signaling to communicate TLP information.
    Type: Grant
    Filed: March 13, 2015
    Date of Patent: August 15, 2017
    Assignee: Intel Corporation
    Inventors: Bryan L. Spry, Marcus W. Song, Deepak M. Rangaraj, Avinash N. Ananthakrishnan, Robert J. Hayes, Aimee D. Wood, Adam E. Letendre, Brent R. Boswell
  • Publication number: 20170115716
    Abstract: One embodiment of an apparatus includes a semiconductor chip having a processor and an on-die non-volatile storage resource. The on-die non-volatile storage may store different, appropriate performance related information for different configurations and usage cases of the processor for a same performance state of the processor.
    Type: Application
    Filed: August 16, 2016
    Publication date: April 27, 2017
    Inventors: ANKUSH VARMA, KRISHNAKANTH V. SISTLA, MARTIN T. ROWLAND, CHRIS POIRIER, ERIC J. DEHAEMER, AVINASH N. ANANTHAKRISHNAN, JEREMY J. SHRALL, XIUTING C. MAN, STEPHEN H. GUNTHER, KRISHNA K. RANGAN, DEVADATTA V. BODAS, DON SOLTIS, HANG T. NGUYEN, CYPRIAN W. WOO, THI DANG
  • Patent number: 9618997
    Abstract: In one embodiment, the present invention includes a multicore processor with a power controller to control a frequency at which the processor operates. More specifically, the power controller can limit a maximum operating frequency of the processor to less than a configured maximum operating frequency to enable a reduction in a number of frequency transitions occurring responsive to power state events, thus avoiding the overhead of operations performed in handling such transitions. Other embodiments are described and claimed.
    Type: Grant
    Filed: December 10, 2014
    Date of Patent: April 11, 2017
    Assignee: Intel Corporation
    Inventors: Avinash N. Ananthakrishnan, Efraim Rotem, Doron Rajwan, Eliezer Weissmann, Ryan Wells, Nadav Shulman
  • Publication number: 20170097668
    Abstract: In one embodiment, the present invention includes a processor having multiple domains including at least a core domain and a non-core domain that is transparent to an operating system (OS). The non-core domain can be controlled by a driver. In turn, the processor further includes a memory interconnect to interconnect the core domain and the non-core domain to a memory coupled to the processor. Still further, a power controller, which may be within the processor, can control a frequency of the memory interconnect based on memory boundedness of a workload being executed on the non-core domain. Other embodiments are described and claimed.
    Type: Application
    Filed: December 16, 2016
    Publication date: April 6, 2017
    Inventors: Avinash N. Ananthakrishnan, Inder M. Sodhi, Efraim Rotem, Doron Rajwan, Eliezer Weissmann, Ryan Wells
  • Patent number: 9606595
    Abstract: Dynamic runtime calibration of a processor with respect to a specific voltage regulator that powers the processor or a memory subsystem coupled to the processor can reduce or eliminate the need for guardbands in power management computations. The processor receives a current measurement from the voltage regulator and computes a calibration factor based on the measured value and a stored expected value. The calibration factor can be used in making power management decisions instead of adding the guardband to power readings. A manufacturer or distributor of the processor can compute the stored values with a controlled voltage supply that has a higher precision than typical commercial power supplies used in computing systems. The computed, stored values indicate the expected value, which can be used to determine a calibration factor relative to a voltage regulator of an active system.
    Type: Grant
    Filed: December 30, 2011
    Date of Patent: March 28, 2017
    Assignee: Intel Corporation
    Inventors: Ankush Varma, Jeremy J. Shrall, Krishnakanth Venkata Sistla, Avinash N. Ananthakrishnan, Vivek Garg, Christopher A. Poirier, Sr., Martin T. Rowland, Edward R. Stanford
  • Publication number: 20170083076
    Abstract: In one embodiment, the present invention includes a processor having a core and a power controller to control power management features of the processor. The power controller can receive an energy performance bias (EPB) value from the core and access a power-performance tuning table based on the value. Using information from the table, at least one setting of a power management feature can be updated. Other embodiments are described and claimed.
    Type: Application
    Filed: December 2, 2016
    Publication date: March 23, 2017
    Inventors: Krishnakanth V. Sistla, Jeremy Shrall, Stephen H. Gunther, Efraim Rotem, Alon Naveh, Eliezer Weissmann, Anil Aggarwal, Martin T. Rowland, Ankush Varma, Ian M. Steiner, Matthew Bace, Avinash N. Ananthakrishnan, Jason Brandt
  • Patent number: 9594560
    Abstract: In an embodiment, a processor includes a first logic to calculate a scalability value for a processor domain based at least in part on an active state residency, a stall duration, and a memory bandwidth of the domain, and to determine an operating frequency update for the domain based at least in part on a current operating frequency of the domain and the scalability value. Other embodiments are described and claimed.
    Type: Grant
    Filed: September 27, 2013
    Date of Patent: March 14, 2017
    Assignee: Intel Corporation
    Inventors: Avinash N. Ananthakrishnan, Stephen H. Gunther, Jeremy J. Shrall, Jay D. Schwartz
  • Publication number: 20170010656
    Abstract: In one embodiment, the present invention is directed to a processor having a plurality of cores and a cache memory coupled to the cores and including a plurality of partitions. The processor can further include a logic to dynamically vary a size of the cache memory based on a memory boundedness of a workload executed on at least one of the cores. Other embodiments are described and claimed.
    Type: Application
    Filed: September 20, 2016
    Publication date: January 12, 2017
    Inventors: Avinash N. Ananthakrishnan, Efraim Rotem, Eliezer Weissmann, Doron Rajwan, Nadav Shulman, Alon Naveh, Hisham Abu-Salah
  • Patent number: 9535487
    Abstract: In one embodiment, the present invention includes a processor having a core and a power controller to control power management features of the processor. The power controller can receive an energy performance bias (EPB) value from the core and access a power-performance tuning table based on the value. Using information from the table, at least one setting of a power management feature can be updated. Other embodiments are described and claimed.
    Type: Grant
    Filed: September 16, 2015
    Date of Patent: January 3, 2017
    Assignee: Intel Corporation
    Inventors: Krishnakanth V. Sistla, Jeremy Shrall, Stephen H. Gunther, Efraim Rotem, Alon Naveh, Eliezer Weissmann, Anil Aggarwal, Martin T. Rowland, Ankush Varma, Ian M. Steiner, Matthew Bace, Avinash N. Ananthakrishnan, Jason Brandt
  • Patent number: 9513688
    Abstract: A scalability algorithm causes a processor to initialize a performance indicator counter, operate at an initial frequency of the first clock signal for a first duration, and determine, based on the performance indicator counter, an initial performance of the first processing core. The algorithm may then cause the processor to operate at a second frequency of the first clock signal for a second duration and determine, based on the performance indicator counter, a second performance of the first processing core. A performance scalability of the first processing core may be determined based on the initial performance and the second performance and an operational parameter, such as one or more clock frequencies and/or supply voltage(s), may be changed based on the determined scalability.
    Type: Grant
    Filed: March 16, 2013
    Date of Patent: December 6, 2016
    Assignee: Intel Corporation
    Inventors: Ankush Varma, Krishnakanth V. Sistla, Jeremy J. Shrall, Avinash N. Ananthakrishnan
  • Patent number: 9501299
    Abstract: A processor may include a cause agnostic frequency dither filter (FD filter), which may cause reduction in the frequency transitions while maintaining the performance levels. The FD Filter may minimize the performance loss, which may otherwise accrue from these frequency transitions, while trying to maximize the peak frequency of the processor. The FD filter may determine a minimum and maximum limit, which may be used by a power management unit (PMU) to restrict the number of frequency transitions to be within a specified threshold. The FD filter may determine the maximum and minimum limits based on transition data stored in internal tables captured during one or more time windows (or observation windows). Based on an average system behavior, the PMU may either apply the minimum or the maximum limit over the subsequent time window.
    Type: Grant
    Filed: June 16, 2014
    Date of Patent: November 22, 2016
    Assignee: Intel Corporation
    Inventors: Ameya Ambardekar, Avinash N. Ananthakrishnan, Ian M. Steiner
  • Publication number: 20160313778
    Abstract: In one embodiment, the present invention includes a processor having multiple domains including at least a core domain and a non-core domain that is transparent to an operating system (OS). The non-core domain can be controlled by a driver. In turn, the processor further includes a memory interconnect to interconnect the core domain and the non-core domain to a memory coupled to the processor. Still further, a power controller, which may be within the processor, can control a frequency of the memory interconnect based on memory boundedness of a workload being executed on the non-core domain. Other embodiments are described and claimed.
    Type: Application
    Filed: April 26, 2016
    Publication date: October 27, 2016
    Inventors: Avinash N. Ananthakrishnan, Inder M. Sodhi, Efraim Rotem, Doron Rajwan, Eliezer Weissmann, Ryan Wells
  • Patent number: 9471490
    Abstract: In one embodiment, the present invention is directed to a processor having a plurality of cores and a cache memory coupled to the cores and including a plurality of partitions. The processor can further include a logic to dynamically vary a size of the cache memory based on a memory boundedness of a workload executed on at least one of the cores. Other embodiments are described and claimed.
    Type: Grant
    Filed: August 31, 2015
    Date of Patent: October 18, 2016
    Assignee: Intel Corporation
    Inventors: Avinash N. Ananthakrishnan, Efraim Rotem, Eliezer Weissmann, Doron Rajwan, Nadav Shulman, Alon Naveh, Hisham Abu-Salah
  • Publication number: 20160266941
    Abstract: In one embodiment, the present invention includes a multicore processor having a power controller with logic to dynamically switch a power management policy from a power biased policy to a performance biased policy when a utilization of the processor exceeds a threshold level. Thus at low utilizations, reduced power consumption can be realized, while at higher utilizations, greater performance can be realized. Other embodiments are described and claimed.
    Type: Application
    Filed: May 24, 2016
    Publication date: September 15, 2016
    Inventors: Krishnakanth V. Sistla, Mark Rowland, Ankush Varma, Ian M. Steiner, Matthew Bace, Daniel Borkowski, Vivek Garg, Cagdas Akturan, Avinash N. Ananthakrishnan
  • Patent number: 9417681
    Abstract: One embodiment of an apparatus includes a semiconductor chip having a processor and an on-die non-volatile storage resource. The on-die non-volatile storage may store different, appropriate performance related information for different configurations and usage cases of the processor for a same performance state of the processor.
    Type: Grant
    Filed: April 27, 2015
    Date of Patent: August 16, 2016
    Assignee: Intel Corporation
    Inventors: Ankush Varma, Krishnakanth V. Sistla, Martin T. Rowland, Chris Poirier, Eric J. Dehaemer, Avinash N. Ananthakrishnan, Jeremy J. Shrall, Xiuting C. Man, Stephen H. Gunther, Krishna K. Rangan, Devadatta V. Bodas, Don Soltis, Hang T. Nguyen, Cyprian W. Woo, Thi Dang