`Patent No. 10,049,080
`
`
`UNITED STATES PATENT AND TRADEMARK OFFICE
`___________________
`
`BEFORE THE PATENT TRIAL AND APPEAL BOARD
`___________________
`
`
`
`MERCEDES-BENZ USA, LLC,
`
`Petitioner,
`
`v.
`
`DAEDALUS PRIME LLC,
`
`Patent Owner.
`_________________________________
`Case No.: IPR2023-01333
`
`U.S. Patent No. 10,049,080
`___________________
`
`
`DECLARATION OF MICHAEL C. BROGIOLI, PH.D.
`IN SUPPORT OF PATENT OWNER’S RESPONSE
`
`
`
`
`
`MB USA v. Daedalus Prime
`IPR 2023-01333
`Daedalus Ex. 2006
`
`
`
`
`
`
`
`
`
`
`
`
`
`TABLE OF CONTENTS
`
`INTRODUCTION .................................................................................... - 1 -
`I.
`QUALIFICATIONS ................................................................................. - 2 -
`II.
`III. LEGAL STANDARDS ............................................................................ - 7 -
`IV. A PERSON OF ORDINARY SKILL IN THE ART (“POSITA”) ........ - 10 -
`V. OVERVIEW OF THE ’080 PATENT ................................................... - 11 -
`A.
`Background .................................................................................. - 11 -
`B.
`The ’080 Patent ............................................................................ - 13 -
`C.
`The Challenged Claims ................................................................ - 16 -
`D.
`Claim Construction ...................................................................... - 18 -
`VI. THE CHALLENGES BASED ON SUTARDJA DO NOT
`DEMONSTRATE OBVIOUSNESS OF THE INDEPENDENT
`CLAIMS ................................................................................................. - 19 -
`A. Overview of Sutardja ’748 ........................................................... - 19 -
`B.
`Overview of Sutardja ’785 ........................................................... - 20 -
`C.
`In Sutardja, a Core Switching Module Transitions the System
`Between HP and LP Modes ......................................................... - 21 -
`In Sutardja, the Core Switching Module is Software ................... - 21 -
`Hypervisors in Sutardja ................................................................ - 22 -
`Inconsistencies in the Petition ...................................................... - 23 -
`Sutardja Does Not Teach or Suggest an Operating System
`that Controls Power Management Hardware ............................... - 24 -
`
`D.
`E.
`F.
`G.
`
`
`
`
`
`
`
`- - ii
`
`
`
`
`
`VII. THE CHALLENGES BASED ON MATHIESON-SUTARDJA DO
`NOT DEMONSTRATE OBVIOUSNESS OF THE
`INDEPENDENT CLAIMS .................................................................... - 26 -
`A. Overview of Mathieson ................................................................ - 26 -
`In Mathieson, Transfer Between Cores is Transparent to the
`B.
`Operating System ......................................................................... - 29 -
`C. Mathieson’s Fig. 6 Does Not Demonstrate a Second Plurality
`of Cores Consuming Less Power for a Same Applied Supply
`Voltage ......................................................................................... - 31 -
`The Fuses Referenced in Mathieson are Storage Elements ......... - 32 -
`D.
`E. Mathieson is Explaining a Difference in Power Consumption
`Based on Operating Frequency .................................................... - 33 -
`
`
`
`
`
`
`
`
`
`- - iii
`
`
`
`
`
`I, Michael C. Brogioli, declare as follows:
`
`I.
`
`Introduction
`
`1.
`
`I have been retained on behalf of Patent Owner, Daedalus Prime LLC
`
`(“Daedalus”), to provide my opinions in connection with this proceeding. In doing
`
`so, I have been asked to provide my independent review, analysis, and insights
`
`regarding the technology involved in U.S. Patent No 10,049,080 (“the ’080
`
`patent”) and how it compares with the art asserted in the petition challenging the
`
`patentability of its claims. In this Declaration, I set forth the independent opinions
`
`that I have reached and the basis for those opinions using the information currently
`
`available to me.
`
`2.
`
`I am being compensated for my work in this matter at my standard
`
`hourly rate for consulting services. My compensation is not dependent on the
`
`outcome of the case and does not affect the substance of my statements in this
`
`Declaration. I have no financial interest in the matter or the patents at issue.
`
`3.
`
`I have reviewed the specification and the claims of the ’080 patent. I
`
`am familiar with the technology involved therein, having worked in the field for
`
`over 20+ years.
`
`4.
`
`To prepare this Declaration, I have considered the petition and the
`
`references cited therein, including the declaration of Dr. Horst, the ’080 patent, my
`
`own experience and knowledge, and the documents cited herein. I provide my
`
`- 1 -
`MB USA v. Daedalus Prime
`
`IPR 2023-01333
`
`Daedalus Ex. 2006
`
`
`
`
`
`
`
`
`
`testimony from the perspective of a person of ordinary skill in the art at the time of
`
`the invention. I was at least a person of ordinary skill in the art at the time of the
`
`invention. I am familiar with the knowledge and skill level of these people at the
`
`time of the invention because I was involved in the industry at the time and had
`
`frequent interactions with such people at the relevant time. I have also been asked
`
`to construe the claims of the ’080 patent, applying the plain and ordinary meaning
`
`of the terms thereof from the perspective of a person of ordinary skill in the art in
`
`light of the specification and other intrinsic evidence.
`
`II. Qualifications
`
`5.
`
`Below is a summary of my education and experience. My curriculum
`
`vitae, included as Appendix A hereto, records my education, experience, and
`
`publications in greater detail.
`
`6.
`
`I am currently an Adjunct Professor of Electrical and Computer
`
`engineering at Rice University in Houston, Texas, and Managing Director of
`
`Polymathic Consulting in Austin, Texas. I received my Bachelor of Electrical
`
`Engineering from Rensselaer Polytechnic Institute in 1999, my Master of Science
`
`in Electrical and Computer Engineering from Rice University in 2003, and my
`
`Doctorate of Electrical and Computer Engineering from Rice University in 2007.
`
`7. While at Rice University, I developed various computer architecture
`
`designs for embedded and high-performance systems, including multi-core and
`
`- 2 -
`MB USA v. Daedalus Prime
`
`IPR 2023-01333
`
`Daedalus Ex. 2006
`
`
`
`
`
`
`
`
`
`low power systems. For example, from 1999 to 2003, I worked in the area of low-
`
`power computing, specifically focusing on dynamic power management and
`
`performance of configurable computing and memory systems. From 2002 to 2004,
`
`I developed Spinach, a computer architecture design and modeling toolset, which
`
`models system components common to all programmable computing
`
`environments, including memory systems, multi-core microprocessor systems, and
`
`related hardware components. From 2004 to 2009, I developed Spinach DSP-
`
`FPGA, a modular and composable simulator design infrastructure for
`
`programmable and reconfigurable embedded SOC architectures specifically
`
`targeting mobile, low-power, and embedded and portable computing devices.
`
`From 2005 to 2009, I developed and published a retargetable compiler
`
`infrastructure and hardware design exploration toolkit for systems microprocessor
`
`design, which facilitated the design space exploration of microprocessors and
`
`multi-core processor designs. Many of these tools have been used at United States
`
`universities in the area of electrical and computer engineering research.
`
`8.
`
`In the late 1990s, I was a hardware and software developer at
`
`Vicarious Visions in New York, developing third-party titles for Nintendo’s
`
`handheld consoles, in addition to various hardware interfaces and
`
`hardware/software optimizations related low power portable computing devices.
`
`
`
`
`
`
`
`- 3 -
`
`
`
`MB USA v. Daedalus Prime
`IPR 2023-01333
`Daedalus Ex. 2006
`
`
`
`
`
`During my career, I have served as Chief Technology Officer, advisory board, and
`
`board of directors’ member, often in co-founding roles.
`
`9.
`
`From June 2006 to August 2007, I worked as the Technical Co-
`
`Founder of Method Seven LLC in Boston, MA, working with high-performance
`
`software and hardware systems architecture. I am currently a co-founder, co-
`
`inventor, and Chief Technology Officer of Network Native, an Internet of Things
`
`technology company as well as others that are detailed on my CV (Appendix A).
`
`10.
`
`I have held the position of Adjunct Professor at Rice University since
`
`2009 and the position of Managing Director at Polymathic Consulting since 2011.
`
`At Rice University, I instruct graduate-level curriculum in the areas of computer
`
`architecture, hardware and software systems. I also advise on university research
`
`and various design initiatives. At Polymathic Consulting, I work with a range of
`
`technologists from early-stage start-ups to Fortune 500 companies on similar
`
`technologies.
`
`11. From 2008 to 2009, I was Senior Engineer working in high-
`
`performance compiler designs and next-generation multi-core microprocessors and
`
`architectures at Freescale Semiconductor in Austin, TX. From November 2009 to
`
`October 2011, I was Chief Architect, Senior Member Technical Staff, at Freescale
`
`Semiconductor in Austin, TX (formerly Motorola), responsible for management of
`
`technology, engineering roadmaps, design lead on software infrastructure, and
`
`- 4 -
`MB USA v. Daedalus Prime
`
`IPR 2023-01333
`
`Daedalus Ex. 2006
`
`
`
`
`
`
`
`
`
`next-generation multi-core microprocessor architectures. During my tenure at
`
`Freescale Semiconductor, I was in charge of system developer tools for processor
`
`design, both in the hardware and software spaces. These included tools used for
`
`the programming of processors, simulation and design of processors, and related
`
`technologies.
`
`12.
`
`I have previously worked for Texas Instruments’ Advanced
`
`Architecture and Chip Technology division in Houston, Texas, in the areas of
`
`high-performance mobile and embedded systems design at the hardware and
`
`systems software level, specifically around heterogeneous computing, high-speed
`
`bus, and interconnect technologies. I also have worked at Intel Corporation’s
`
`Microprocessor Research Labs in the areas of computer architecture and compiler
`
`technologies.
`
`13.
`
`I am recognized as an expert in the field of computer architecture,
`
`computer hardware and computer software systems as they relate to the subject
`
`matter at hand. I am a member of the Institute of Electrical and Electronics
`
`Engineers (IEEE) and Association for Computing Machinery (ACM) and have
`
`been a Program Committee member for the IEEE and ACM Design Automation
`
`Conference since 2011, and have previously held the role of Program Chair of
`
`Design Automation Conference in the area of Embedded Computing. I have also
`
`
`
`
`
`
`
`- 5 -
`
`
`
`MB USA v. Daedalus Prime
`IPR 2023-01333
`Daedalus Ex. 2006
`
`
`
`
`
`been a reviewer and contributor to a number of IEEE and ACM technical
`
`conferences during the course of my career.
`
`14. Over the past 20 plus years, I have authored numerous peer-reviewed
`
`publications, as well as engineering books in the area of computer hardware and
`
`software design. Many of these incorporate technologies specific to the subject
`
`matter at hand. These publications are disclosed in my curriculum vitae, attached
`
`as Appendix A. I have previously served as an engineering consultant and
`
`testifying witness on matters related to, and including, microprocessors and multi-
`
`core technology as well as various aspects of power management hardware and
`
`software, etc. My curriculum vitae contains more information on my background
`
`and experience, as well as the cases in which I have served as an expert witness
`
`over at least the past four years.
`
`15.
`
`I am also a named inventor on the following United States patents and
`
`patent applications: U.S. Pat. Nos. 11,526,180; 11,354,757; 11,340,877; U.S. Pat.
`
`Appl. No. 17/060,025 (U.S. Publication No. 2021/0099522); and U.S. Pat. Appl.
`
`No. 15/110,643 (U.S. Publication No. 2016/0328993). More information regarding
`
`the patent applications for which I am a named inventor can be found in my
`
`curriculum vitae, attached as Appendix A.
`
`
`
`
`
`
`
`- 6 -
`
`
`
`MB USA v. Daedalus Prime
`IPR 2023-01333
`Daedalus Ex. 2006
`
`
`
`
`
`III. Legal Standards
`
`16.
`
`I have been informed by counsel of general principles concerning
`
`patent law and the analysis of whether a patent claim is invalid. Below I set forth
`
`my understanding of the legal principles that may be relevant to my opinions in
`
`this Declaration.
`
`17.
`
`I understand that a party challenging a patent’s patentability in an IPR
`
`proceeding bears the burden of proof by a preponderance of evidence. I also
`
`understand that the anticipation of a claim requires that a single prior art reference
`
`discloses (explicitly or inherently) every element required by the claim; that is, the
`
`single reference must teach all of the features or limitation of the claim.
`
`18.
`
`I have been informed and understand that, under U.S. patent law, a
`
`patent may not issue “if the differences between the subject matter sought to be
`
`patented and the prior art are such that the subject matter as a whole would have
`
`been obvious at the time the invention was made to a person having ordinary skill
`
`in the art to which said subject matter pertains.” This has typically been referred to
`
`as the obviousness inquiry. I have been informed and understand that the
`
`obviousness inquiry requires that the claimed invention be considered as a whole
`
`and that in determining the differences between the prior art and the claims, the
`
`question is not whether the differences themselves would have been obvious, but
`
`whether the claimed invention as a whole would have been obvious.
`
`- 7 -
`MB USA v. Daedalus Prime
`
`IPR 2023-01333
`
`Daedalus Ex. 2006
`
`
`
`
`
`
`
`
`
`19.
`
`I have been informed and understand that obviousness is a question of
`
`law based on underlying factual determinations, including: (1) the scope and
`
`content of prior art; (2) differences between prior art and claims; (3) the level of
`
`ordinary skill in the art; and (4) objective indicia of non-obviousness.
`
`20.
`
`I have been informed and understand that a party asserting that a
`
`patent is obvious must demonstrate that a skilled artisan would have had reason to
`
`combine the teaching of the asserted prior art references to achieve the claimed
`
`invention, and that the skilled artisan would have had a reasonable expectation of
`
`success in doing so. I have been informed and understand that reasonable
`
`expectation of success is assessed as of the date of the invention described in the
`
`patent. That the inventors were ultimately successful is irrelevant to whether one of
`
`ordinary skill in the art, at the time the invention was made, would have reasonably
`
`expected success. Any finding to the contrary represents impermissible use of
`
`hindsight— using the inventors’ success as evidence that the success would have
`
`been expected.
`
`21.
`
`I have been informed that, as part of determining the scope and
`
`content of the prior art, it may be appropriate to consider whether there is evidence
`
`of a “teaching, suggestion, or motivation” to combine the prior art teachings in the
`
`prior art, the nature of the problem or the knowledge of a person having ordinary
`
`skill in the art. The question is not simply whether the prior art teaches the
`
`- 8 -
`MB USA v. Daedalus Prime
`
`IPR 2023-01333
`
`Daedalus Ex. 2006
`
`
`
`
`
`
`
`
`
`particular element of the invention. Knowledge of the goal does not render its
`
`achievement obvious.
`
`22.
`
`I have been informed that one indicator of non-obviousness is when
`
`prior art “teaches away” from combining certain known elements. A prior art
`
`reference teaches away from the patent’s particular combination if it leads in a
`
`different direction or discourages that combination, recommends steps that would
`
`not likely lead to the patent’s result, or otherwise indicates that a seemingly
`
`inoperative device would be produced. If the prior art indicated that the invention
`
`would not have worked for its intended purpose or otherwise taught away from the
`
`invention, the invention is not obvious.
`
`23.
`
`I have been informed that a patent that is composed of several
`
`elements is not made obvious merely because each of those elements was
`
`independently known in the prior art. Rather, the useful inquiry is whether there
`
`was a reason that would have prompted a person having ordinary skill in the art to
`
`combine the elements in the manner of the new invention. Thus, the challenger
`
`must show that at the time of the invention a person of ordinary skill in the art
`
`would have been motivated to combine the references, not just that the references
`
`could have been combined.
`
`24.
`
`I have been informed that one must remain vigilant of the distortion
`
`caused by hindsight bias and must be cautious of arguments reliant upon ex post
`
`- 9 -
`MB USA v. Daedalus Prime
`
`IPR 2023-01333
`
`Daedalus Ex. 2006
`
`
`
`
`
`
`
`
`
`reasoning. The selective hindsight combination of references that show various
`
`elements of the claim generally does not suffice to establish obviousness.
`
`IV. A person of ordinary skill in the art (“POSITA”)
`
`25.
`
`In its Institution Decision, the Board assessed MB USA’s proposed
`
`level of ordinary skill in the art as “consistent with the prior art.” That proposed
`
`level of ordinary skill in the art is: a person with “a bachelor’s degree in electrical
`
`engineering, computer science, computer engineering, material science, physics,
`
`applied physics, or a related field” who would have had “at least two years of
`
`experience in the research, design, development, or testing of electronic circuits or
`
`components or software for controlling electronic circuits or components, or the
`
`equivalent, with additional education substituting for experience and vice versa.” I
`
`have been asked to apply a somewhat modified version of MB USA’s definition of
`
`a person of ordinary skill in the art, limiting that individual’s work experience to
`
`two years in the designated field (that is, not “at least” two years expertise, which
`
`may encompass significantly more than that), so as not to include persons who
`
`might exceed the level of ordinary skill in the art for purposes of this declaration.
`
`26. At the time of the invention recited in the ’080 patent, I met and
`
`exceeded this definition of a POSITA.
`
`
`
`
`
`
`
`- 10 -
`
`
`
`MB USA v. Daedalus Prime
`IPR 2023-01333
`Daedalus Ex. 2006
`
`
`
`
`
`V. Overview of the ’080 Patent
`
`A. Background
`
`27. The ’080 patent is entitled “Asymmetric Performance Multicore
`
`Architecture With Same Instructions Set Architecture.” EX1001 at p. 1 (54). The
`
`‘080 patent purports to solve certain perceived issues in power processor
`
`management. Rather than simply making adjustments to operating frequency
`
`and/or voltage or employing low power states when a processor’s core is inactive,
`
`the ’080 patent describes the use of different types of processor cores that consume
`
`different amounts of power and offer different performance capabilities (for a
`
`given operating frequency and/or voltage), yet which support the same instruction
`
`set in order to manage power consumption of the processor. Id. at 3:50-62.
`
`28. Multi-core processors are those that combine a plurality of processor
`
`cores on a single semiconductor die. Id. at 1:22-25. The different processor cores
`
`are the functional units that read and execute instructions contained in a computer
`
`program. The presence of multiple cores on a single die means a multi-core
`
`processor can execute multiple such instructions simultaneously.
`
`29. The amount of power consumed by a processor is strongly correlated
`
`with its performance capabilities. Id. at 2:26-29. Generally, decreasing processing
`
`performance of a system will correspond to a power savings and vice-versa; hence,
`
`
`
`processor power management techniques often involve scaling up performance as
`- 11 -
`
`MB USA v. Daedalus Prime
`IPR 2023-01333
`
`Daedalus Ex. 2006
`
`
`
`
`
`
`
`
`
`a processor’s workload increases, and scaling down performance as that workload
`
`decreases. Id. at 2:22-27. Such scaling of processing performance (and
`
`concomitant power consumption) with workload was often done by enabling or
`
`disabling entire processor cores by raising or lowering their supply voltages and
`
`operating frequencies (and thereby their power consumption) in response to system
`
`workload. Id. at 2:30-34.
`
`30. The ’080 patent notes that such solutions have limitations. For
`
`example, merely adjusting the operating frequency (also called “clock”) and/or
`
`voltage at which a core operates to adjust its power consumption/processing
`
`capability balance presents core management difficulties. See, e.g., EX2004 (U.S.
`
`Patent No. 8,892,931 to Kruglick) at 1:31–37 (“Attempts to apply [Dynamic
`
`Voltage and Frequency Scaling] to individual cores within a multicore processor
`
`may encounter a number of difficulties as the number of cores increases.”);
`
`EX2005 (U.S. Patent No. 9,086,883 to Thomson) at 1:42-56 (“Conventional
`
`[Dynamic Clock and Voltage Scaling] solutions [for multi-core processors] exhibit
`
`a number of performance problems, and implementing an effective DCVS method
`
`… for each core … is an important and challenging design criterion.”).
`
`31. Other approaches to power management, such as the use of low-
`
`power states, also have drawbacks. For example, it can take time (and energy) to
`
`wake a processor from a low power state. See EX1033 at 1:26-33 (“One approach
`
`- 12 -
`MB USA v. Daedalus Prime
`
`IPR 2023-01333
`
`Daedalus Ex. 2006
`
`
`
`
`
`
`
`
`
`to reducing power in a computing platform when there is relatively little activity, is
`
`to place the processor in a low-power state. However, placing a processor in a low-
`
`power state or returning a processor from a low-power state may require a non-
`
`trivial amount of time. Therefore, it may or may not be worth the time required to
`
`place a processor in a low-power state or to return the processor from a low-power
`
`state.”).
`
`B.
`
`The ’080 Patent
`
`32. The power management techniques described in the ’080 patent do
`
`not rely solely on adjustments to operating frequency and/or voltage or employing
`
`low power states when a core is inactive. Instead, as mentioned above, in the ’080
`
`patent different types of processor cores that consume different amounts of power
`
`and offer different performance capabilities (for a given operating frequency and/or
`
`voltage), yet which support the same instruction set, are used in a single processor.
`
`EX1001 at 3:50-62. The different types of processor cores employ different power
`
`consumption design features, such as different transistor technologies. These
`
`different processor cores are considered at the design stage of the processor, and
`
`different technology libraries are used when the different cores of the processor are
`
`fabricated. Id. at 3:63 – 4:9. Consequently, while the different types of processor
`
`cores can each execute the same instructions, “low power” cores will “exhibit
`
`inherently lower power consumption (and processing performance) than the higher
`
`- 13 -
`MB USA v. Daedalus Prime
`
`IPR 2023-01333
`
`Daedalus Ex. 2006
`
`
`
`
`
`
`
`
`
`power core(s),” even when the low power core is operated at the same clock
`
`frequency as a higher power core. Id. at 3:53 – 4:16.
`
`33. Because the different types of processor cores employ different power
`
`consumption design features, they consume different amounts of power and offer
`
`different performance capabilities. Hence, the different cores are referred to as
`
`“asymmetric” cores. Id. at 1:17-20. The use of asymmetric cores in the ’080 patent
`
`allows a single processor to “scale up [it]s processing performance … as the
`
`system’s workload increases, and scale down [it]s processing performance … as
`
`the system’s workload decreases,”’ just as prior multicore processors did, but
`
`because processors designed according to the teachings of the ’080 patent
`
`incorporate inherently low power cores, those processors are able to “achieve an
`
`even lower power consumption in the lower/lowest performance/power states.” Id.
`
`at 2:19-29, 3:50-62, 4:20-46.
`
`34. One power management strategy presented in the ’080 patent is
`
`described with reference to the patent’s FIG. 6, reproduced below. At 601, a multi-
`
`core processor with multiple high power cores and at least one low power core is
`
`operating. Id. at 4:47–59. As demand on the processor decreases, the various high
`
`power cores are disabled one at a time, until all are disabled. Id. at 4:54 – 5:6. If all
`
`of the high power cores are disabled and the demand on the processor continues to
`
`drop, then the low power cores are disabled one by one until only one low power
`
`- 14 -
`MB USA v. Daedalus Prime
`
`IPR 2023-01333
`
`Daedalus Ex. 2006
`
`
`
`
`
`
`
`
`
`core remains enabled and the processor’s lowest power state is reached. Id. at
`
`5:25–35. During this process, before a core (high power or low power) is disabled,
`
`its individual operating frequency and/or supply voltage, or the operating
`
`frequencies and/or supply voltages of some or all of the enabled cores, may be
`
`reduced in response to the drop in demand. And, if only one low power core is
`
`operating, its operating frequency and/or supply voltage may similarly be reduced
`
`until the lowest power consumption state is reached. Id. at 5:7-24, 28-49.
`
`
`
`
`
`
`
`
`
`
`
`- 15 -
`
`
`
`MB USA v. Daedalus Prime
`IPR 2023-01333
`Daedalus Ex. 2006
`
`
`
`
`
`C. The Challenged Claims
`
`35. The petition challenges various ones of the claims of the ’080 patent
`
`under different theories, as represented in the following table:
`
`Claims Challenged
`1–4, 7–12, 15–20, 23, 24
`5, 6, 13, 14, 21, 22
`7, 15, 23
`1–4, 7–12, 15–20, 23, 24
`5, 6, 13, 14, 21, 22
`
`35 U.S.C. § References
`103
`“Sutardja”
`103
`Sutardja, Rychlik1
`103
`Sutardja, Carmack2
`103
`Mathieson,3 Sutardja
`103
`Mathieson, Sutardja, Rychlik
`
`
`Pet. at 5-6.
`
`36.
`
`I understand that for the challenges based on “Sutardja,” MB USA has
`
`considered the combined teachings of Sutardja US PGPUB 2008/0288748 A1
`
`(referred to as Sutardja ’748) and Sutardja US PGPUB 2008/0288748 A1 (referred
`
`to as Sutardja ’785) because the former incorporates by reference the latter. Pet. at
`
`5 n.2. I adopt a similar approach in my analysis. Likewise, I understand MB USA
`
`
`1 US PGPUB 2011/0145615 A1.
`
`2 US PGPUB 2009/0309243 A1.
`
`3 US PGPUB 2011/0213950 A1.
`
`
`
`
`
`
`
`- 16 -
`
`
`
`MB USA v. Daedalus Prime
`IPR 2023-01333
`Daedalus Ex. 2006
`
`
`
`
`
`considers Mathieson to incorporate by reference the entire disclosure of Carmack.
`
`Id. at 5 n.5. Therefore, I do likewise for purposes of this declaration.
`
`37.
`
`Independent claim 1 of the ’080 patent recites:
`
`A multi-core processor comprising:
`
`a first plurality of cores and a second plurality of cores
`
`that support a same instruction set, wherein the second
`
`plurality of cores consume less power, for a same applied
`
`operating frequency and supply voltage, than the first plurality
`
`of cores; and
`
`power management hardware to, from a state where the
`
`first plurality of cores and the second plurality of cores are
`
`enabled, disable all of the first plurality of cores for a drop in
`
`demand below a threshold without disabling any of the second
`
`plurality of cores, wherein an operating system to execute on
`
`the multi-core processor is to monitor a demand for the multi-
`
`core processor and control the power management hardware
`
`based on the demand.
`
`EX1001 at 7:56 – 8:3.
`
`38.
`
`Independent claim 9 recites a method that includes operating a multi-
`
`core processor such that a first plurality of cores and a second plurality of cores
`
`- 17 -
`MB USA v. Daedalus Prime
`
`IPR 2023-01333
`
`Daedalus Ex. 2006
`
`
`
`
`
`
`
`
`
`execute a same instruction set, wherein the second plurality of cores consume less
`
`power, for a same applied operating frequency and supply voltage, than the first
`
`plurality of cores; and disabling with power management hardware, from a state
`
`where the first plurality of cores and the second plurality of cores are enabled, all
`
`of the first plurality of cores for a drop in demand below a threshold without
`
`disabling any of the second plurality of cores, wherein an operating system
`
`executing on the multi-core processor monitors a demand for the multi-core
`
`processor and controls the power management hardware based on the demand. Id.
`
`at 8:38-53. Independent claim 17 recites a non-transitory machine-readable
`
`medium containing program code that, when processed by a machine, causes a
`
`method as recited in claim 9 to be performed. Id. at 9:24-41.
`
`D. Claim Construction
`
`39.
`
`I understand that when reviewing the patentability of a claim, the
`
`analysis requires that the claim must first be construed. As mentioned above, I
`
`have been asked to construe the claims of the ’080 patent by applying the plain and
`
`ordinary meaning of the terms thereof from the perspective of a person of ordinary
`
`skill in the art in light of the specification and other intrinsic evidence.
`
`40. Each of the independent claims of the ’080 patent recites the term
`
`“operating system.” In particular, the claims recite an operating system that
`
`executes on a multi-core processor to monitor demand for the multi-core processor
`
`- 18 -
`MB USA v. Daedalus Prime
`
`IPR 2023-01333
`
`Daedalus Ex. 2006
`
`
`
`
`
`
`
`
`
`and control power management hardware based on the demand. EX1001 at 7:66 –
`
`8:3 (claim 1); 8:49-53 (claim 9); 9:37-41 (claim 17). The specification of the ’080
`
`patent provides examples of “operating systems,” including Apple’s MacOS,
`
`Microsoft’s Windows, UNIX, and others. Id. at 7:31-35. Therefore, I conclude that
`
`a person of ordinary skill in the art would understand that the term “operating
`
`system” is used in the ’080 patent in its conventional sense. Specifically, in the
`
`context of the ’080 patent a person of ordinary skill in the art would understand an
`
`”operating system” to be, “The software that controls the allocation and usage of
`
`hardware resources such as memory, central processing unit (CPU) time, disk
`
`space, and peripheral devices.” Ex. 2007 at 378 (Microsoft Computer Dictionary).
`
`VI. The Challenges Based on Sutardja Do Not Demonstrate Obviousness of
`the Independent Claims
`
`A. Overview of Sutardja ’748
`
`41.
`
` US PGPUB 2008/0288747 A1 (“Sutardja ’748”) (EX1007) describes
`
`“core morphing,” a process in which “cores [of a multi-core processor] are
`
`dynamically enabled (i.e., activated) or disabled (i.e., deactivated) based on the
`
`system load.” EX1007 at [0218]. When one core is active, other cores may be
`
`disabled (i.e., deactivated), for example, by being put in a standby mode or
`
`completely shut down, to save power. Id.
`
`
`
`
`
`
`
`- 19 -
`
`
`
`MB USA v. Daedalus Prime
`IPR 2023-01333
`Daedalus Ex. 2006
`
`
`
`
`
`42.
`
`In particular, Sutardja ’748 describes core morphing in the case of a
`
`processing system having a low-speed, low-power (LP) core and a high-speed,
`
`high-power (HP) core. Id. at [0219] et seq. In the HP mode, the HP core is in the
`
`active state and processes threads. The LP core may also operate during the HP
`
`mode. In other words, the LP core may be in the active state during all or part of
`
`the HP mode. Id. at [0223]. “When applications demand still higher performance
`
`than that provided by one HP core, multiple HP cores may be used.” Id. at [0219].
`
`B. Overview of Sutardja ’785
`
`43.
`
` US PGPUB 2007/0083785 A1 (Sutardja ’785) (EX1008) describes a
`
`system on chip (“SOC”) that includes two processors, each of which has active and
`
`inactive states. EX1008 at [0009]. One processor (the “second” processor)
`
`consumes less power when operating in its active state than the other processor
`
`(the “first” processor) when operating in its active state. Id. The SOC also includes
`
`a control module that communicates with the two processors and selectively
`
`transfers threads from the first processor to the second processor and selects the
`
`inactive state of the first processor. Id. In programming, a “thread” is “a process
`
`that is part of a larger process or program.” EX2007 at 518.
`
`
`
`
`
`
`
`
`
`
`
`
`
`- 20 -
`
`
`
`MB USA v. Daedalus Prime
`IPR 2023-01333
`Daedalus Ex. 2006
`
`
`
`
`
`
`
`C.
`
`In Sutardja, a Co