DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Status of Claims
Claims 1, 8 and 15 have been amended and are hereby entered.
Claims 1-20 are pending and have been examined.
This action is made FINAL.
Response to Arguments
Applicant's arguments filed July 17, 2026 have been fully considered but they are not persuasive.
Regarding the applicant's arguments against the 101 rejection of pending claims on pages 7 -15: Applicant’s arguments directed to 101 analysis were considered. However, these arguments are not persuasive and the examiner respectfully disagrees for the following reasons:
For Step 2A-Prong 1 starting in p. 7: The Applicant argues that the pending claims are not directed to any of the abstract ideas identified, firstly, because “the claims are NOT "related to legal obligations” since the Applicant removed the limitations steps related to determining “scope 2 CO2” emissions and its costs per asset identified being more than a threshold value cost. However, the Examiner finds this argument unpersuasive and respectfully disagrees. Because even if Applicant remove these limitations the rest of the claims’ language is still reciting the determination of additional or omitted assets that are evaluated, measured and compared based on their energy consumption and the energy consumption of a data center to take corrective actions related to environmental, social, and governance (ESG) compliance to manage, optimize such assets and enable an “accurate asset inventory and effective energy monitoring” (see ¶0022 from Applicant disclosure) which encompasses legal interactions related to legal obligations for running assets in accordance to ESG compliance.
As for Applicant arguments directed to mental process identified in the claims in p. 8 from Remarks, these are unpersuasive and the Examiner respectfully disagrees. Because upon review of the claim steps, certain steps directed in part to measuring”, “extrapolating” and “comparing” energy consumption by assets to “determine” total energy consumption as well as the other values previously measured, to identify additional assets that are not in the list and the energy consumption per asset compared to the total consumption of the data center as a threshold difference to “automatically perform” ESG compliance related actions were considered mental processes. Such functions for asset identification and determinations as well as the determination and comparison of energy consumption per asset to automatically perform the actions encompasses observation, evaluation and judgement. Moreover, contrary to Applicant allegations that these specific functions “cannot be performed by the human mind”, such functions can still either be done with the help of physical aid such as pen and paper or can be performed by humans without or with the assistance (e.g. tool) a computer. Thus, the steps do not negate and further still reads in the mental nature of the limitation(s), when evaluating such data to perform automatic actions, as well as the concept is merely claimed to be performed on a generic computer and is merely using a computer as a tool to perform the concept of deciding an ESG compliance related action based on energy consumption and assets information (see MPEP 2106.04(a)(2)(III) (B & C)).
Finally, even after Applicant amendments that eliminated or the claims do not the CO2e determinations (p. 10 from Remarks), the abstract idea of mathematical concepts is now reflected in claim 15 step of “extrapolating the measured sample energy consumption over an intended time period” as it requires mathematical calculations when performing extrapolation calculations for the energy consumption values.
For Step 2A-Prong 2 and Step 2B starting in p. 11: The Applicant alleges that the claims integrate, the judicial exception identified, into a practical application and further alleges that the claims “are directed to a solution of a problem of the data centers consuming more energy than the list of assets for which the data centers are required to provide all necessary facilities including providing power for consumption” and thus they, “provide a practical application of the elements leading to reduction or elimination of wasted energy resource usage and causing improvement in the energy efficiency of the operations of the data center”. However, the Examiner finds these arguments unpersuasive and respectfully disagrees. Because the identified limitations in the claims did not integrate a judicial exception into a practical application since the steps were merely reciting the words "apply it" (or an equivalent) with the judicial exception, or merely including instructions to implement an abstract idea on a computer, or merely using a computer as a tool to perform an abstract idea (see MPEP 2106.05(f) and 2106.04(d)(I)) to achieve the recited end result of taking or automatically performing corrective actions on the identified assets in the data center to comply with ESG requirements as well as to optimize energy consumption and its costs. Specifically, the claims’ limitations are reciting the use of a generic computer that further is “using a platform energy consumption measurement tool” or “one or more energy consumption measurement modules that use vendor provided APIs” that are generally/broadly recited, that further “measure” energy consumption data per asset for further comparison with the energy consumption data of the data center to determine or find any asset (i.e. existing or additional asset from the list of assets) that is not complying to achieve the intended result of automatically performing corrective actions such as decommission, uninstalling and/or powering the asset down. Therefore, the alleged “substantial improvement of operation of a data center” for governing for “how unauthorized power consuming assets are handled” is still the intended result of invoking a computer used as a tool to perform the functions that recite the abstract idea identified (see MPEP 2106.05(f) and 2106.04(d)(I)). Further, “claiming the improved speed or efficiency inherent with applying the abstract idea on a computer" does not integrate a judicial exception into a practical application or provide an inventive concept” (see MPEP 2106.05(f)(2); TLC communications).
Lastly, these claims are improving of the abstract idea itself and further narrows the drafted abstract ideas identified and their specificity in their limitations which does not necessarily equate to eligibility. For instance, In buySAFE, Inc. v. Google, Inc. (Fed. Cir. 2014), the court stated that "abstract ideas, no matter how groundbreaking, innovative, or even brilliant, are outside what the statute means by "new and useful process, machine, manufacture, or composition of matter", and reference is made to Myriad by the court for this position. Also stated in buySAFE is "In defining the excluded categories, the Court has ruled that the exclusion applies if a claim involves a natural law or phenomenon or abstract idea, even if the particular natural law or phenomenon or abstract idea at issue is narrow. Mayo, 132 S. Ct. at 1303. The Court in Mayo rejected the contention that the very narrow scope of the natural law at issue was a reason to find patent eligibility, explaining the point with reference to both natural laws and one kind of abstract idea, namely, mathematical concepts.
As for the arguments regarding to step 2B in p. 13 from Remarks, these claim limitations and their additional elements, individually and in an ordered combination, are not “significantly more” as these are recited in a high level of generality with a computer used as a tool to execute the claimed functions, as previously discussed in step 2A Prong 2 which carries the same analysis that is applied in step 2B. Thus, the claim limitations cannot provide an inventive concept at Step 2B and are not integrating the abstract idea into a practical application (see MPEP 2106.05). Therefore, for all the reasons stated above, the Examiner respectfully disagrees and maintains 35 USC § 101 rejection for these pending claims.
Regarding to Applicant's arguments of rejection under 35 USC § 103 for the pending claims on pages 15 – 18: Applicant’s arguments regarding these amended limitation steps in the pending claim are not persuasive and respectfully disagrees. Because Applicant’s arguments regarding these amended limitation steps in the pending claims are not persuasive and the Examiner respectfully disagrees. Because upon re-evaluation of the previously referenced prior art combination of Kennedy, Ramakrishnan and Katiyar, still reasonably teaches the new amended steps directed in part to “measure…energy consumption by assets…”, “compare” the energy consumption of the assets and the total energy consumption of data center, “determine” additional assets based on the difference between the total energy consumption of the data center and an aggregate of the measured energy consumption by the assets to determine the additional asset that is consuming more energy and is omitted in the list, “automatically perform” corrective actions and “tag the additional assets” with “accurate tagged details” as well as “performing machine learning and data modeling techniques” for energy efficiency strategies refinement and for improving the efficient use of the data center that the Applicant alleges not being taught. Because Applicant is focusing on each prior art teachings, rather than focusing on the actual language claimed as well as the breadth of each claim limitation and how their corresponding limitation steps are different from the prior art teachings. Rather, the steps disclose a broader language that the prior art combination of Kennedy, Ramakrishnan and Katiyar, still reasonably satisfies when evaluated in light of the broadest reasonable interpretation (BRI) of the claim language. Please, refer to the Claim Rejections - 35 USC § 103 section for further details. Therefore, the Examiner respectfully disagrees, and maintains 35 USC § 103 rejection for these pending claims.
Claim Rejections - 35 USC § 101
35 U.S.C. 101 reads as follows:
Whoever invents or discovers any new and useful process, machine, manufacture, or composition of matter, or any new and useful improvement thereof, may obtain a patent therefor, subject to the conditions and requirements of this title.
Claims 1 - 20 are rejected under 35 U.S.C. 101 because the claimed invention is directed to an abstract idea without significantly more. The analysis of this claimed invention recited in the claims begins in view of independent claim 1, the most representative claim of the independent claims set 1, 8 and 15, as follows:
At Step 1: Claims 1 – 7 and claims 15 - 20 falls under statutory category of a machine, while claims 8 - 14 are directed to a process. Claims 15 - 20 recite a computer storage medium. Examiner notes that ordinarily a broadest reasonable interpretation of such claims would include transitory signals, which are not statutory (e.g. “signal per se”) under step 1 of the section 101 analysis (see MPEP § 2106 subsection I and § 2106.03). Nevertheless, Applicant has a special definition for "computer storage medium" in ¶0057 that explicitly states that "a computer storage medium does not include a propagating signal." Thus, claims 15 - 20 are statutory for the purpose of analysis in step 1 of the analysis.
At Step 2A Prong 1: Claims 1, 8 and 15 recites an abstract idea in the following limitations:
For claim 1 (representative of claims 8 and 15):
…obtain a list of assets of a data center;
measure…energy consumption by assets in the list of assets during a time period;
receive total energy consumption of the data center during the time period from an energy provider;
compare the energy consumption by the assets in the list of assets with the total energy consumption of the data center during the time period;
determine, based a difference between the total energy consumption of the data center and an aggregate of the measured energy consumption by the assets in the list of assets, that an additional asset of the data center is consuming more energy and is omitted from the list of assets;
automatically perform a corrective data center control action on the additional asset, the corrective data center control action comprising at least one of powering off, or uninstalling an application corresponding to the additional asset; and
tag the additional asset with energy-consumption or carbon-emission information causing…to include accurate tagged details for the additional asset in a subsequent asset-discovery iteration
Generally, and as disclosed in the specification in ¶0022, this claimed invention “provide a technical effect by offering an integrated approach to enhance environmental, social, and governance (ESG) compliance within data centers through the management and optimization of data center assets” and “enables accurate asset inventory and effective energy monitoring, allowing for comprehensive tracking of energy consumption across all data center components” while facilitating “the identification and optimization of underutilized or redundant assets.” However, the abstract idea(s) of a certain method of organizing human activity (See MPEP 2106.04(a)(2), subsection II) are recited in claims 1, 8 and 15 in the form of “commercial or legal interactions”. Specifically, the abstract idea is recited at least in the steps of “measure… energy consumption by assets in a list of assets during a time period” and “extrapolating the measured sample energy consumption over an intended time period” (see claim 15 only for this step), “receive total energy consumption of the data center” from an “energy provider”, “compare the energy consumption by the assets in the list of assets with the total energy consumption of the data center”, “determine… that an additional asset of the data center is consuming more energy and is omitted from the list of assets” based on energy consumption differences from the data center and the assets list, to further “automatically perform a corrective data center control action on the additional asset” such as actions of “decommissioning the identified asset, virtualizing the identified asset, and configuring the identified asset in balanced mode.” (see claim 15 and see claims 2 and 10 as these are actions are later claimed), “tag the additional asset with energy-consumption or carbon-emission information” causing “accurate tagged details” to be included in a “subsequent asset-discovery iteration” and finally “perform machine learning and data modeling techniques to refine energy efficiency strategies and improving the efficient use of the data center” (see claim 15 only). Because determining additional or omitted assets of a data center based on energy consumption differences of a data center and an aggregate of a list of assets, and compare and tag list of assets (i.e. “servers, storage devices, network equipment, rack power distribution units (PDU), and applications hosted on the servers”; see ¶0001 from Applicant disclosure) with energy-consumption or carbon-emission information to take actions or a corrective data center control action on the additional asset for environmental, social, and governance (ESG) compliance encompasses legal interactions related to legal obligations for running assets in accordance to ESG compliance.
The steps directed in part to “measuring”, “extrapolating” and “comparing” energy consumption by assets to “determine” total energy consumption as well as the other values previously measured, to identify additional assets that are not in the list and the energy consumption per asset compared to the total consumption of the data center as a threshold difference to “automatically perform” ESG compliance related actions fall under the abstract idea of mental processes that can be practically be performed in the human mind or in pen and paper (See MPEP 2106.04(a)(2), subsection III). Because such functions for asset identification and determinations as well as the determination and comparison of energy consumption per asset to automatically perform the actions encompasses observation, evaluation and judgement. Also, these steps can either be done with the help of physical aid such as pen and paper or can be performed by humans without or with the assistance (e.g. tool) a computer. Thus, the steps do not negate and further still reads in the mental nature of the limitation(s), when evaluating such data to perform automatic actions, as well as the concept is merely claimed to be performed on a generic computer and is merely using a computer as a tool to perform the concept of deciding an ESG compliance related action based on energy consumption and assets information (see MPEP 2106.04(a)(2)(III) (B & C)).
As for the step directed in part to “cause a measurement” of energy consumption, specifically “extrapolating the measured sample energy consumption over an intended time period” (claimed at least in claim 15) is falling into mathematical concepts that can be performed mentally or in pen and paper. Because this step require specific mathematical calculations when performing at least the extrapolation calculation.
At Step 2A Prong 2: For independent claims 1, 8 and 15, The judicial exception(s) or abstract idea previously identified is not integrated into a practical application (see MPEP 2106.04 (d)). The claims recite the additional element(s) of a processor; and a memory comprising computer program code, using a platform energy consumption measurement tool, (from claim 1), using one or more energy consumption measurement modules that use vendor provided APIs (from claim 8), a computer storage medium, using a tool, perform machine learning and data modeling techniques to refine energy efficiency strategies and improving the efficient use of the data center (from claim 15) and a discovery module (from claims 1, 8 and 15). These additional elements, individually and in combination, and while considering the claims as a whole, are merely used as a tool to perform the abstract idea (See MPEP 2106.05(f)). Specifically, these steps are recited as being performed by the computer that further uses tools such as “platform energy consumption measurement tool” and perform “machine learning and data modeling techniques” to refine energy efficiency strategies and use. The computer used along with these tools and techniques are recited at a high level of generality that is being used as a tool to perform the generic computer functions for identifying/determining assets as well as comparing energy consumption of assets to automatically perform the ESG compliance related actions. Thus, these steps mentioned above are further describing and applying the abstract idea without placing any limits on how the technological components are being improved, while distinguishing in the claim language, the performing limitations from functions that generic computer components can perform.
Step 2B: For independent claims 1, 8 and 15, these claims do not provide an inventive concept. The recited additional elements of the claim(s) are the following: a processor; and a memory comprising computer program code, using a platform energy consumption measurement tool, (from claim 1), using one or more energy consumption measurement modules that use vendor provided APIs (from claim 8), a computer storage medium, using a tool, perform machine learning and data modeling techniques to refine energy efficiency strategies and improving the efficient use of the data center (from claim 15) and a discovery module (from claims 1, 8 and 15). These additional elements are not sufficient to amount significantly more than the judicial exception or abstract idea (see MPEP 2106.05). Because, as indicated in Step 2A Prong 2, these additional element(s) claimed are merely, instructions to “apply” the abstract ideas, which cannot provide an inventive concept. Thus, even when considered in combination, these additional elements represent mere instructions to implement an abstract idea or other exception on a computer, which do not provide an inventive concept at Step 2B.
For dependent claims 2-7, 9-14 and 16 - 20, the same analysis is incorporated. Due to their dependency to the independent claims analyzed, these claims cover or fall under the same abstract idea(s) of a method of organizing human activity and mental processes. They describe additional limitations steps of:
Claims 2-7, 9-14 and 16 - 20: further describes the abstract idea of the computerized method for ESG compliance of assets in data centers and further defines the actions automatically performed, by whom the total energy consumption is received, by how the energy consumption is measured, determine assets satisfying a criterion that is further defined in terms of thresholds to automate the performing action, determining type of asset, presenting trend analysis of costs and other user inputs received to identify underutilized assets for further automatically performing actions. Thus, being directed to the abstract idea group of commercial interactions related to legal obligations and mental/mathematical processes for ESG compliance based on CO2e determinations.
Step 2A Prong 2 and Step 2B: For dependent claim 11, this claim recites the additional elements of: a user interface (claims 7, 14 and 20). These additional elements recited are invoking computers merely used as a tool to perform or “apply” the abstract idea(s) to the existing process of presenting trend analysis of asset costs data. Thus, amounting to no more than mere instructions to “apply” the exception using a generic computer component (MPEP 2106.05(f) and (f)(2)). Accordingly, for the same reasons stated above, these additional element(s) claimed cannot provide an inventive concept at Step 2B.
Claim Rejections - 35 USC § 103
In the event the determination of the status of the application as subject to AIA 35 U.S.C. 102 and 103 (or as subject to pre-AIA 35 U.S.C. 102 and 103) is incorrect, any correction of the statutory basis (i.e., changing from AIA to pre-AIA ) for the rejection will not be considered a new ground of rejection if the prior art relied upon, and the rationale supporting the rejection, would be the same under either status.
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 8 - 20 are rejected under 35 U.S.C. 103 as being unpatentable over Kennedy (U.S. Pub No. 20200249734 A1) in view of Ramakrishnan (U.S. Patent No. 11996988 B1).
Regarding claim 8:
Kennedy teaches:
obtaining a list of assets of a data center; (In ¶0057; Figs. 1 – 2 (110); Fig. 3 (302): teaches a “power controller 110 monitors variations in power consumption of a plurality of computing devices in an IT stack (302)” and “receives data from various devices, such as power monitors, for each server rack 120”.)
measuring, using one or more energy consumption measurement modules that use vendor provided APIs, energy consumption by the assets in the list of assets, during a time period; (In ¶0057 – 58; Fig. 2 (210 and 110); Fig. 3 (304): teaches “the power consumption can be provided by the fuel cell system 210, or measured by another device”. Further, “power controller 110 determines a power consumption profile for the plurality of computing devices that describes historical statistics of the power consumption (304)”)
receive total energy consumption of the data center during the time period from an energy provider; (In ¶0058; Fig. 3 (304 – 306): teaches that “power controller 110 tracks and records power consumption and power transients to create and maintain a power consumption profile for the data center facility 100 that includes historical statistics” and these “historical statistics of peaks and valleys are used by the power controller 110 to determine a power consumption profile for the plurality of computing devices.” Also, “a power consumption profile for the data center facility 100 may be provided by another entity, such as the fuel cell system provider, or by the utility provider before the fuel system was implemented, if available” (see ¶0059).)
determine, based a difference between the total energy consumption of the data center and an aggregate of the measured energy consumption by the assets in the list of assets, that an additional asset of the data center is consuming more energy and is omitted from the list of assets; and (In ¶0065; Fig. 3 (304 – 306); Fig. 4 (402): teaches that “the power controller 110 can measure the power consumption of the data center facility 100 and/or receive the power consumption level from another device, such as the fuel cell system 210. The power controller 110 can then determine if the current power consumption is above a particular threshold” and in “response to determining that the present consumption level of the plurality of computing devices is above a first threshold, the power controller 110 decreases performance of the plurality of computing devices or decreases an operating level of an ancillary system that draws power from the same power source of the plurality of computing devices (404)”, which is directed to an example of the comparison between the total energy consumption of the data center and an aggregate of the measured energy consumption by the assets in the list of assets to determine the additional asset that is consuming more energy and is omitted in the list. Examiner notes that the omission of the asset in the list is simply descriptive matter that does not hold patentable weight. Further, the “power controller 110” can recognize and instruct operations in “the computer system (e.g., servers 120 and/or additional servers 232)” (see ¶0062) which is directed to identifying/determining additional assets.)
automatically perform the action by, when the identified asset is a physical asset, automatically powering off the identified asset and, (In ¶0041; Fig. 4 (404): teaches this conditional limitation under broadest reasonable interpretation (BRI), in an example for automatically powering off the identified asset wherein “In some implementations, should the power source 102 fail or become unreliable, the power controller 110 can utilize the switches 106 with control line 116 to isolate the data center facility 100 from the power source 102. In particular, the power feeds 103, 105, from the power source 102 and the inverter 206, respectively, can be switched off, and the power can be provided from the utility provider 150 through backup power feed 154”, in accordance with an asset decommissioning example given in ¶0054 from Applicant disclosure. Further, in ¶0051 “the power controller 110 can lower the aggregate load to the one or more devices by activating a switch that cuts the power to the respective device from the fuel cells” and “the power controller 110 can thus reduce the aggregate load powered by the fuel cell system by cutting off the power one or more devices, such as less important devices of the data center facility, by activating a switch”. Examiner notes that having an identified asset as physical asset does not hold patentable weight and is deemed as non-functional descriptive matter.)
when the identified asset is a virtual asset or application, automatically disabling or uninstalling the virtual asset or application from the data center. (In ¶0050; Fig. 4 (404): teaches this conditional limitation under broadest reasonable interpretation (BRI), in an example wherein “the power controller 110 can lower the aggregate load to the one or more devices by sending a control signal to each respective device. The control signal allows the power controller 110 to control and manage the performance of each respective device. For example, the power controller 110 can adjust clock speed management of the servers 120 or additional servers 232 by sending a control signal to reach respective server to reduce power consumption by decreasing the clock speed's on individual servers (e.g., schedule less workload, turn down the clock's, initiate a low-power mode, etc.)” wherein the adjustments that the “power controller” is capable of doing suggests the identification of an application (i.e. a virtual asset) that can be managed to lower power or schedule an operation such as an automatic disablement of the application via a “control signal” inside the remote servers.)
Kennedy does not explicitly teach the ability of determining an asset of the data center not being in the list of assets that is based on the total energy consumption threshold between the assets list and the data center. However, Ramakrishnan further discloses that “various software modules comprising application instructions 187 may be coordinated by an operating system (OS), and/or via an application programming interface (API)” wherein such instructions may be utilized by the system to “power down and reroute workload away from these predicted underutilized components” (see C9; L2 – 25; Ramakrishnan) including software components (i.e. virtual asset or application) such as “managed drives” from “memory devices” to “power down or enter a sleep mode” (see C15; L45 – 55; Ramakrishnan) and teaches the previous limitation of determining the additional asset omitted from the list as:
determine… that an additional asset of the data center…is omitted from in the list of asset; (In C31; L19 – 34; Fig. 2 (200 and 283); Fig. 5 (506 – 508): teaches this negative limitation that does not hold patentable weight under the broadest reasonable interpretation (BRI), as described in “block 506” wherein the system’s “power-throttling learning module 283 may be trained to determine whether any given load-balancing instruction to divert workload away from a given hardware component or combination of hardware components will likely meet the QoS requirements for any JO commands from a host information handling system or hardware performance policies set for all hardware components within the data storage system/data center 210” wherein those hardware components that those not meet the criteria are not (or should not be) in the assets list and thus, may be omitted. Further, in C32; L3 – 18, “the power-throttling learning module 283 receives a notification that memory hardware is predicted to be underutilized during an upcoming time window, the power-throttling learning module 283 may transmit a memory-related load-balancing instruction to alternative memory hardware, meeting the QoS requirements for the data storage system/data center 210, that is also associated with a high reward value (e.g., positive one), instead of other memory-related load-balancing instructions being sent to the power-throttling agent 219 of the data center 210 that may not satisfy customer QoS requirements” which is directed to identifying assets that must be in the requirement lists and separating those assets that do not meet the requirements and therefore are predicted to be underutilized and omitted from the list.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to modify Kennedy to provide the ability of determining an asset of the data center not being in the list of assets that is based on the total energy consumption threshold between the assets list and the data center, as taught by Ramakrishnan in order to “identify an optimal load-balancing technique for rerouting any IO commands directed to the underutilized component(s) to other alternative hardware components that can satisfy customer [quality of service or] QoS requirements” (C29; L39 – 42; Ramakrishnan) as well as to “limit unnecessary power consumption and consequent carbon emissions by these underutilized hardware components by powering down capacity” (C3; L33 – 36; Ramakrishnan), see also MPEP 2143.I.G.
Regarding claim 15:
Kennedy teaches:
obtaining a list of assets of a data center; (In ¶0057; Figs. 1 – 2 (110); Fig. 3 (302): teaches a “power controller 110 monitors variations in power consumption of a plurality of computing devices in an IT stack (302)” and “receives data from various devices, such as power monitors, for each server rack 120”.)
cause a measurement of energy consumption by the assets, using a tool, in the list of assets during a sample time period and extrapolating the measured sample energy consumption over an intended time period; (In ¶0057 – 58; Fig. 1 (110); Fig. 2 (210, 240 and 250); Fig. 3 (304): teaches “the power consumption can be provided by the fuel cell system 210, or measured by another device”. Further, “power controller 110 determines a power consumption profile for the plurality of computing devices that describes historical statistics of the power consumption (304)” which is directed to the system being able to extrapolate measured sample energy consumption over an intended time period using the “plots 240, 250” from Fig. 2.)
receive total energy consumption of the data center during the intended time period; (In ¶0058; Fig. 3 (304 – 306): teaches that “power controller 110 tracks and records power consumption and power transients to create and maintain a power consumption profile for the data center facility 100 that includes historical statistics” and these “historical statistics of peaks and valleys are used by the power controller 110 to determine a power consumption profile for the plurality of computing devices.” Also, “a power consumption profile for the data center facility 100 may be provided by another entity, such as the fuel cell system provider, or by the utility provider before the fuel system was implemented, if available” (see ¶0059).)
compare the energy consumption by the assets in the list of assets with the total energy consumption of the data center during the intended time period; based on the comparison, determine that the total energy consumption of the data center is more than the energy consumption by the assets in the list of assets; In ¶0060; Fig. 3 (304 – 306); Fig. 4 (402): teaches that the “power controller 110 determines that power consumption is changing toward a peak consumption or a valley consumption (306)”. For example, “assume the present power consumption for the data center is X KWH, and based on power consumption profile for the plurality of computing devices (e.g., certain hours of a day experience higher server traffic, thus higher power consumption), the power controller 110 can determine that the power consumption for the data center is going to increase to Y KWH for a particular portion of the day”. See ¶0065 wherein the “power controller 110 determines that a present consumption level of the plurality of computing devices is above a first threshold (402)”. For example, “the power controller 110 can measure the power consumption of the data center facility 100 and/or receive the power consumption level from another device, such as the fuel cell system 210” to then determine the threshold level in terms of power consumption.)
automatically perform an action on the identified asset of the data center, the action comprising causing decommissioning of the identified asset; and (In ¶0062; Fig. 3 (308); Fig. 4 (404): teaches that “the power controller 110 can reduce the variations in power consumed by adjusting which devices in the IT stack are consuming power from the power source.” See ¶0066 for an example wherein “In response to determining that the present consumption level of the plurality of computing devices is above a first threshold, the power controller 110 decreases performance of the plurality of computing devices or decreases an operating level of an ancillary system that draws power from the same power source of the plurality of computing devices (404)”. Further, in ¶0041, discloses an example of decommissioning (i.e. the removal) of an identified asset wherein “In some implementations, should the power source 102 fail or become unreliable, the power controller 110 can utilize the switches 106 with control line 116 to isolate the data center facility 100 from the power source 102. In particular, the power feeds 103, 105, from the power source 102 and the inverter 206, respectively, can be switched off, and the power can be provided from the utility provider 150 through backup power feed 154”, in accordance with an asset decommissioning example given in ¶0054 from Applicant disclosure.)
Kennedy does not explicitly teach the abilities of identifying an asset of the data center not being in the list of assets that is based on the energy consumption threshold and performing machine learning and data modeling techniques for energy efficiency strategies refinement and for improving the efficient use of the data center. However, Ramakrishnan teaches:
based on the determination, identify an asset of the data center not in the list of assets; (In C31; L19 – 34; Fig. 2 (200 and 283); Fig. 5 (506 – 508): teaches this negative limitation that does not hold patentable weight under the broadest reasonable interpretation (BRI), as described in “block 506” wherein the system’s “power-throttling learning module 283 may be trained to determine whether any given load-balancing instruction to divert workload away from a given hardware component or combination of hardware components will likely meet the QoS requirements for any JO commands from a host information handling system or hardware performance policies set for all hardware components within the data storage system/data center 210”. Further, in C32; L3 – 18, “the power-throttling learning module 283 receives a notification that memory hardware is predicted to be underutilized during an upcoming time window, the power-throttling learning module 283 may transmit a memory-related load-balancing instruction to alternative memory hardware, meeting the QoS requirements for the data storage system/data center 210, that is also associated with a high reward value (e.g., positive one), instead of other memory-related load-balancing instructions being sent to the power-throttling agent 219 of the data center 210 that may not satisfy customer QoS requirements” which is directed to identifying assets that must be in the requirement lists and separating those assets that do not meet the requirements and therefore are predicted to be underutilized.)
perform machine learning and data modeling techniques to refine energy efficiency strategies and improving the efficient use of the data center. (In C19; L27 – 46; Fig. 4 (408 – 414): teaches “power-throttling learning module 283 in an embodiment may be trained using training period operational telemetry measurements including the performance metrics and ability to meet QoS requirements by a plurality of data center hardware components (e.g., 232, 240 a, 240 b, 241 a, 241 b, 242 a, 242 b, 243 a, 243 b, 244 a, 244 b, 245 a, 245 b, 211, 212, 250 a, 250 b, 251, 252, 253 or 254) for IO commands, as affected by execution of each of the available load-balancing instructions to develop the probability matrix” and “upon development of that matrix, the power-throttling learning module 283 may be further trained that consequent states (e.g., performance metrics of data center hardware component recorded after execution of a given load-balancing instruction) that meet the QoS requirements for all data center hardware components affected by the executed load-balancing instruction will be “rewarded” (e.g., associated with a value of positive one), and those that do not meet the QoS requirements for IO commands for any one of the data center hardware components affected will be “punished” (e.g., associated with a value of negative one)”. Further in C20; L46 – 53, “The power-throttling learning module 283 in such an embodiment may then identify the potential load-balancing instructions that is most likely to meet all QoS requirements of the host application 284 or data center 210 policy for the data storage system/data center 210, based on the reward outputs from the execution of code instructions of the reinforcement learning algorithm described herein, as the recommended load-balancing instruction” which reflects the improvement of efficient use of the data center. See Fig 2 and C11; L9 – 22 for general details of “reinforcement learning (RL) based data center power consumption minimizing system 280 executing on a unified endpoint management (UEM) platform 200 information handling system for identifying data storage system/data center 210 hardware components that are predicted to be underutilized during an upcoming time window, powering down those components during the upcoming time window, and using an RL-based method to identify predicted load-balancing instructions for rerouting any input/output (IO) commands from being directed to the underutilized component(s) to other alternative optimal hardware components that will meet 10 command QoS requirements of a host information handling system”. See C27; L57 – 67 wherein the refinement of energy efficiency strategies via the model training is taught at “block 410, the processor or processors executing code instructions of the RL based data center power consumption minimizing system at the UEM platform information handling system in an embodiment may receive monitoring period operational telemetry including updated information of the same type identified within the training period operational telemetry received at block 402. For example, following training of the utilization forecast engine 286 and the power-throttling learning module 283 in an embodiment, the power-throttling agent 219 at the data storage system/data center 210 may transmit updated operational telemetry, including performance metrics, and utilization rates to the RL based data center power consumption minimizing system 280.”)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to modify Kennedy to provide the abilities of identifying an asset of the data center not being in the list of assets that is based on the energy consumption threshold and performing machine learning and data modeling techniques for energy efficiency strategies refinement and for improving the efficient use of the data center, as taught by Ramakrishnan in order to “identify an optimal load-balancing technique for rerouting any IO commands directed to the underutilized component(s) to other alternative hardware components that can satisfy customer [quality of service or] QoS requirements” (C29; L39 – 42; Ramakrishnan) as well as to “limit unnecessary power consumption and consequent carbon emissions by these underutilized hardware components by powering down capacity” (C3; L33 – 36; Ramakrishnan), see also MPEP 2143.I.G.
Regarding claim 9:
The combination of Kennedy and Ramakrishnan, as shown in the rejection above, discloses the limitations of claim 8.
Kennedy further teaches:
wherein the energy consumption by each of the assets is measured via one of the following: an actual energy consumption, calculated energy consumption, estimated energy consumption, specification sheet of the asset, or an average of energy consumption of assets other than the asset. (In ¶0048; Fig. 2 (240 and 250): teaches the “Plot 250 includes a dotted line that depicts power consumption over time for the data center facility after the power controller 110 is implemented with the power controller 110 controlling the loads. The plot 250 also includes a solid line that represents the same power consumption of plot 240 for purposes of reference. The dotted line depicts the resulting power consumption as controlled by the power controller 110”. See ¶0065 and ¶0068 for more examples of energy consumption determinations.)
Regarding claim 10:
The combination of Kennedy and Ramakrishnan, as shown in the rejection above, discloses the limitations of claim 8.
Kennedy further teaches:
wherein the action comprises one or more of the following: decommissioning the identified asset, virtualizing the identified asset, and configuring the identified asset in balanced mode. (In ¶0066; Fig. 4 (404): teaches that “In response to determining that the present consumption level of the plurality of computing devices is above a first threshold, the power controller 110 decreases performance of the plurality of computing devices or decreases an operating level of an ancillary system that draws power from the same power source of the plurality of computing devices (404)” which is directed to at least the action of configuring the identified asset in balanced mode. Further, in ¶0041, discloses an example of decommissioning (i.e. the removal) of an identified asset wherein “In some implementations, should the power source 102 fail or become unreliable, the power controller 110 can utilize the switches 106 with control line 116 to isolate the data center facility 100 from the power source 102. In particular, the power feeds 103, 105, from the power source 102 and the inverter 206, respectively, can be switched off, and the power can be provided from the utility provider 150 through backup power feed 154”, in accordance to an asset decommissioning example given in ¶0054 from Applicant disclosure.)
Regarding claim 11:
The combination of Kennedy and Ramakrishnan, as shown in the rejection above, discloses the limitations of claim 8.
Kennedy does not explicitly teach the ability of presenting a trending analysis of each assets cost including historical, current and future costs. However, Ramakrishnan teaches:
presenting a trend analysis of the cost associated with each asset in a user interface, the trend analysis comprising historical cost, current cost, and future cost associated with each asset. (See Figs. 3A – 3C: in Fig. 3A shows a graphical diagram of “memory utilization rate for a data storage system/data center determined using a utilization forecasting engine” per “plurality of data center hardware components by group (e.g., memory hardware, processors, fabric network paths, PCIe cards, or other ASIC cards), based on previously recorded utilization rates for each of the plurality of component groups” and further illustrates “a predicted combined utilization rate for all memory hardware components (e.g., storage arrays 250 a and 250 b, and managed drives 251, 252, 253, and 254 as shown in FIG. 2 ) within a data storage system/data center (e.g., 210 of FIG. 2 ) over a monitoring period (e.g., 7 days)” (see C23; L7 – 39). Similarly, in Fig. 3B, shows “a predicted combined utilization rate for all peripheral component interconnect express (PCIe) cards or other application-specific integrated circuit (ASIC) cards, and subscriber line interface cards (SLIC) network paths within a data storage system/data center (e.g., 210 of FIG. 2 ) over a monitoring period (e.g., 7 days)” (see C23; L40 – 55) and in Fig. 3C displays a “a predicted combined utilization rate for all processing hardware components (e.g., computing nodes 240 a and 240 b, storage engine 232, or processors 242 a or 242 b) within a data storage system/data center (e.g., 210 of FIG. 2 ) over a monitoring period (e.g., 7 days)” (see C23; L56 – 67), in accordance to the “cost” or consumption in percentage example given in ¶0046 from Applicant’s disclosure. Refer to C16; L9 – 37 for more details of the GUI employed for the managing user at “a management terminal 217”.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to modify Kennedy to provide the ability of presenting a trending analysis of each assets cost including historical, current and future costs, as taught by Ramakrishnan in order to visualize and “identify an optimal load-balancing technique for rerouting any IO commands directed to the underutilized component(s) to other alternative hardware components that can satisfy customer [quality of service or] QoS requirements” (C29; L39 – 42; Ramakrishnan) as well as to “limit unnecessary power consumption and consequent carbon emissions by these underutilized hardware components by powering down capacity” (C3; L33 – 36; Ramakrishnan), see also MPEP 2143.I.G.
Regarding claim 12:
The combination of Kennedy and Ramakrishnan, as shown in the rejection above, discloses the limitations of claim 11.
Kennedy does not explicitly teach the abilities of receiving user input to identify underutilize asset and automatically perform an action for the underutilized asset identified. However, Ramakrishnan further teaches:
further comprising: receiving an input from a user in the user interface to identify an underutilized asset in the data center; (In C16; L9 – 37; Fig. 2 (215 and 219): teaches that “the managing user in an embodiment may employ a graphical user interface (GUI) (e.g., 215) at a management terminal 217 accessible by a managing user for the data storage system/data center(s) 210 to specify, for example, the low-utility threshold value (e.g., 0.1%, 1%, 2%, 5% of total utilization capacity). A Quality of Service (QoS) requirement such as a maximum allowable response time for processing an IO command (e.g., less than 1 ms, less than 2 ms) for one or more hardware components” and then the “UEM platform 200 in an embodiment may gather this operational telemetry, like that described directly above, routinely from a plurality of hardware components” in order to “execute an RL based data center power consumption minimizing system 280 to identify one or more groups of hardware components (e.g., memory components, processors, PCIe cards, other ASIC cards, fabric network paths) predicted to be underutilized at one or more data centers 210 during an upcoming time window”.)
identifying the underutilized asset; and (In C33; L48 – 53; Fig. 6 (602): teaches that “the power-throttling agent executing at a data storage system/data center in an embodiment may receive instructions to power down an underutilized hardware component and to shift workload away from the underutilized hardware component during an upcoming time window”.)
automatically performing the action on the identified underutilized asset. (In C35; L11 – 15; Fig. 6 (606 – 608): teaches that the “power-throttling agent in an embodiment may terminate or limit power supply to the predicted under-utilized hardware component during the upcoming time window at block 608”.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to modify Kennedy to provide the abilities of receiving user input to identify underutilize asset and automatically perform an action for the underutilized asset identified, as taught by Ramakrishnan in order to “identify an optimal load-balancing technique for rerouting any IO commands directed to the underutilized component(s) to other alternative hardware components that can satisfy customer [quality of service or] QoS requirements” (C29; L39 – 42; Ramakrishnan) as well as to “limit unnecessary power consumption and consequent carbon emissions by these underutilized hardware components by powering down capacity” (C3; L33 – 36; Ramakrishnan), see also MPEP 2143.I.G.
Regarding claim 13:
The combination of Kennedy and Ramakrishnan, as shown in the rejection above, discloses the limitations of claim 8.
Kennedy does not explicitly teach the abilities of determining that the identified asset(s) satisfies a criterion and automatically perform an action the identified asset. However, Ramakrishnan further teaches:
wherein the memory and computer program code are configured to further cause the processor to: determine that the identified asset satisfies a criterion; and (In C34; L43 – 49; Fig. 6 (606): teaches “at block 606 may execute the recommended or optimal load-balancing instruction to reroute incoming IO requests from a host computer directed to the predicted under-utilized hardware component(s) to alternative hardware components set to remain powered on and which may satisfy the IO requests QoS requirements”.)
upon determining that the identified asset satisfies the criterion, automatically perform the action on the identified asset. (In C35; L11 – 14; Fig. 6 (608): teaches “power-throttling agent in an embodiment may terminate or limit power supply to the predicted under-utilized hardware component during the upcoming time window at block 608”.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to modify Kennedy to provide the abilities of determining that the identified asset(s) satisfies a criterion and automatically perform an action the identified asset, as taught by Ramakrishnan in order to “identify an optimal load-balancing technique for rerouting any IO commands directed to the underutilized component(s) to other alternative hardware components that can satisfy customer [quality of service or] QoS requirements” (C29; L39 – 42; Ramakrishnan) as well as to “limit unnecessary power consumption and consequent carbon emissions by these underutilized hardware components by powering down capacity” (C3; L33 – 36; Ramakrishnan), see also MPEP 2143.I.G.
Regarding claim 14:
The combination of Kennedy and Ramakrishnan, as shown in the rejection above, discloses the limitations of claim 13.
Kennedy does not explicitly teach the ability of having a criterion that specifically determines identified asset(s) as a virtual machine, if it’s not running a database or both. However, Ramakrishnan further teaches:
wherein the criterion comprises determining that the identified asset is a virtual machine, determining that the identified asset is not running a database, or both. (In C11; L9 – 21; Fig. 2 (270 and 210): teaches that “a reinforcement learning (RL) based data center power consumption minimizing system 280 executing on a unified endpoint management (UEM) platform 200 information handling system for identifying data storage system/data center 210 hardware components that are predicted to be underutilized during an upcoming time window” wherein the “host computer 270 may be implemented as a virtual machine within storage system 210” (see C11; L53 – 56) which such determination claimed is further defining descriptive material.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to modify Kennedy to provide the ability of having a criterion that specifically determines identified asset(s) as a virtual machine, if it’s not running a database or both, as taught by Ramakrishnan in order to “identify an optimal load-balancing technique for rerouting any IO commands directed to the underutilized component(s) to other alternative hardware components that can satisfy customer [quality of service or] QoS requirements” (C29; L39 – 42; Ramakrishnan) as well as to “limit unnecessary power consumption and consequent carbon emissions by these underutilized hardware components by powering down capacity” (C3; L33 – 36; Ramakrishnan), see also MPEP 2143.I.G.
Regarding claim 16:
The combination of Kennedy and Ramakrishnan, as shown in the rejection above, discloses the limitations of claim 15, respectively.
Kennedy further teaches:
wherein the total energy consumption is received from an energy provider. (In ¶0059: teaches “a power consumption profile for the data center facility 100 may be provided by another entity, such as the fuel cell system provider, or by the utility provider before the fuel system was implemented, if available”.)
Regarding claim 17:
The combination of Kennedy and Ramakrishnan, as shown in the rejection above, discloses the limitations of claim 15, respectively.
Kennedy further teaches:
wherein the energy consumption by each of the assets is measured via one of the following: an actual energy consumption, calculated energy consumption, estimated energy consumption, specification sheet of the asset, or an average of energy consumption of assets other than the asset. (In ¶0048; Fig. 2 (240 and 250): teaches the “Plot 250 includes a dotted line that depicts power consumption over time for the data center facility after the power controller 110 is implemented with the power controller 110 controlling the loads. The plot 250 also includes a solid line that represents the same power consumption of plot 240 for purposes of reference. The dotted line depicts the resulting power consumption as controlled by the power controller 110”. See ¶0065 and ¶0068 for more examples of energy consumption determinations.)
Regarding claim 18:
The combination of Kennedy and Ramakrishnan, as shown in the rejection above, discloses the limitations of claim 15, respectively.
Kennedy does not explicitly teach the abilities of determining that the identified asset(s) satisfies a criterion and automatically perform an action the identified asset. However, Ramakrishnan further teaches:
wherein the memory and computer program code are configured to further cause the processor to: determine that the identified asset satisfies a criterion; and (In C34; L43 – 49; Fig. 6 (606): teaches “at block 606 may execute the recommended or optimal load-balancing instruction to reroute incoming IO requests from a host computer directed to the predicted under-utilized hardware component(s) to alternative hardware components set to remain powered on and which may satisfy the IO requests QoS requirements”.)
upon determining that the identified asset satisfies the criterion, automatically perform the action on the identified asset. (In C35; L11 – 14; Fig. 6 (608): teaches “power-throttling agent in an embodiment may terminate or limit power supply to the predicted under-utilized hardware component during the upcoming time window at block 608”.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to modify Kennedy to provide the abilities of determining that the identified asset(s) satisfies a criterion and automatically perform an action the identified asset, as taught by Ramakrishnan in order to “identify an optimal load-balancing technique for rerouting any IO commands directed to the underutilized component(s) to other alternative hardware components that can satisfy customer [quality of service or] QoS requirements” (C29; L39 – 42; Ramakrishnan) as well as to “limit unnecessary power consumption and consequent carbon emissions by these underutilized hardware components by powering down capacity” (C3; L33 – 36; Ramakrishnan), see also MPEP 2143.I.G.
Regarding claim 19:
The combination of Kennedy and Ramakrishnan, as shown in the rejection above, discloses the limitations of claim 18, respectively.
Kennedy further teaches:
wherein the criterion comprises one of determining that the energy consumption by the identified asset is more than a threshold value or determining that the energy consumption by the identified asset is less than a threshold value. (In ¶0065: teaches “power controller 110 determines that a present consumption level of the plurality of computing devices is above a first threshold (402)”. See ¶0068 for more examples of energy consumption determinations when determining that the energy consumption by the identified asset(s) is less than a threshold value.)
Regarding claim 20:
The combination of Kennedy and Ramakrishnan, as shown in the rejection above, discloses the limitations of claim 18, respectively.
Kennedy does not explicitly teach the ability of having a criterion that specifically determines identified asset(s) as a virtual machine, if it’s not running a database or both. However, Ramakrishnan further teaches:
wherein the criterion comprises determining that the identified asset is a virtual machine, determining that the identified asset is not running a database, or both. (In C11; L9 – 21; Fig. 2 (270 and 210): teaches that “a reinforcement learning (RL) based data center power consumption minimizing system 280 executing on a unified endpoint management (UEM) platform 200 information handling system for identifying data storage system/data center 210 hardware components that are predicted to be underutilized during an upcoming time window” wherein the “host computer 270 may be implemented as a virtual machine within storage system 210” (see C11; L53 – 56) which such determination claimed is further defining descriptive material.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to modify Kennedy to provide the ability of having a criterion that specifically determines identified asset(s) as a virtual machine, if it’s not running a database or both, as taught by Ramakrishnan in order to “identify an optimal load-balancing technique for rerouting any IO commands directed to the underutilized component(s) to other alternative hardware components that can satisfy customer [quality of service or] QoS requirements” (C29; L39 – 42; Ramakrishnan) as well as to “limit unnecessary power consumption and consequent carbon emissions by these underutilized hardware components by powering down capacity” (C3; L33 – 36; Ramakrishnan), see also MPEP 2143.I.G.
Claims 1 - 7 are rejected under 35 U.S.C. 103 as being unpatentable over Kennedy (U.S. Pub No. 20200249734 A1) in view of Ramakrishnan (U.S. Patent No. 11996988 B1) and in further view of Katiyar (U.S. Pub No. 20240127260 A1).
Regarding claim 1:
Kennedy teaches:
a processor; and a memory comprising computer program code, the memory and the computer program code configured to cause the processor to: (See Fig. 1 (100 and 120); Fig. 2 (110, 120 and 220): Refer to ¶0048 for more details about the “a processor-based controller, such as power controller 110” and ¶0044 wherein “the power controller 110 is shown being operatively connected to a historical database 220 to store historical data”.)
obtain a list of assets of a data center; (In ¶0057; Figs. 1 – 2 (110); Fig. 3 (302): teaches a “power controller 110 monitors variations in power consumption of a plurality of computing devices in an IT stack (302)” and “receives data from various devices, such as power monitors, for each server rack 120”.)
measure, using a platform energy consumption measurement tool, energy consumption by assets in a list of assets during a time period; (In ¶0057 – 58; Fig. 1 (110); Fig. 2 (210, 240 and 250); Fig. 3 (304): teaches “the power consumption can be provided by the fuel cell system 210, or measured by another device”. Further, “power controller 110 determines a power consumption profile for the plurality of computing devices that describes historical statistics of the power consumption (304)”.)
receive total energy consumption of the data center during the time period from an energy provider; (In ¶0058; Fig. 3 (304 – 306): teaches that “power controller 110 tracks and records power consumption and power transients to create and maintain a power consumption profile for the data center facility 100 that includes historical statistics” and these “historical statistics of peaks and valleys are used by the power controller 110 to determine a power consumption profile for the plurality of computing devices.” Also, “a power consumption profile for the data center facility 100 may be provided by another entity, such as the fuel cell system provider, or by the utility provider before the fuel system was implemented, if available” (see ¶0059).)
compare the energy consumption by the assets in the list of assets with the total energy consumption of the data center during the [intended] time period; In ¶0060; Fig. 3 (304 – 306); Fig. 4 (402): teaches that the “power controller 110 determines that power consumption is changing toward a peak consumption or a valley consumption (306)”. For example, “assume the present power consumption for the data center is X KWH, and based on power consumption profile for the plurality of computing devices (e.g., certain hours of a day experience higher server traffic, thus higher power consumption), the power controller 110 can determine that the power consumption for the data center is going to increase to Y KWH for a particular portion of the day”. See ¶0065 wherein the “power controller 110 determines that a present consumption level of the plurality of computing devices is above a first threshold (402)”. For example, “the power controller 110 can measure the power consumption of the data center facility 100 and/or receive the power consumption level from another device, such as the fuel cell system 210” to then determine the threshold level in terms of power consumption.)
determine, based a difference between the total energy consumption of the data center and an aggregate of the measured energy consumption by the assets in the list of assets, that an additional asset of the data center is consuming more energy and is omitted from in the list of assets; (In ¶0065; Fig. 3 (304 – 306); Fig. 4 (402): teaches that “the power controller 110 can measure the power consumption of the data center facility 100 and/or receive the power consumption level from another device, such as the fuel cell system 210. The power controller 110 can then determine if the current power consumption is above a particular threshold” and in “response to determining that the present consumption level of the plurality of computing devices is above a first threshold, the power controller 110 decreases performance of the plurality of computing devices or decreases an operating level of an ancillary system that draws power from the same power source of the plurality of computing devices (404)”, which is directed to an example of the comparison between the total energy consumption of the data center and an aggregate of the measured energy consumption by the assets in the list of assets to determine the additional asset that is consuming more energy and is omitted in the list. Examiner notes that the omission of the asset in the list is simply descriptive matter that does not hold patentable weight. Further, the “power controller 110” can recognize and instruct operations in “the computer system (e.g., servers 120 and/or additional servers 232)” (see ¶0062) which is directed to identifying/determining additional assets.)
automatically perform a corrective data center control action on the additional asset, the corrective data center control action comprising at least one of powering off, or uninstalling an application corresponding to the additional asset; and (In ¶0066; Fig. 4 (404): teaches that “In response to determining that the present consumption level of the plurality of computing devices is above a first threshold, the power controller 110 decreases performance of the plurality of computing devices or decreases an operating level of an ancillary system that draws power from the same power source of the plurality of computing devices (404)” which is directed to at least the action of configuring the identified asset in balanced mode. Further, see ¶0041, “In some implementations, should the power source 102 fail or become unreliable, the power controller 110 can utilize the switches 106 with control line 116 to isolate the data center facility 100 from the power source 102. In particular, the power feeds 103, 105, from the power source 102 and the inverter 206, respectively, can be switched off, and the power can be provided from the utility provider 150 through backup power feed 154”.)
Kennedy generally teaches the tagging of assets since the “the power controller can monitor the power drawn by the computing devices to create a power consumption profile that is representative of the power consumption over time” (see ¶0028; Kennedy) wherein the profile includes (i.e. or is tagged with) “information as identification (e.g., name, location, and/or other identification information) identifying a given device to which the power consumption profile pertains” and the system “can store a power consumption profile at the component and subcomponent level” (see ¶0045 – 46; Kennedy) as well as it “records power consumption and power transients to create and maintain a power consumption profile for the data center facility 100 that includes historical statistics” (see ¶0058; Kennedy). Kennedy does not explicitly teach the ability of determining an asset of the data center not being in the list of assets that is based on the total energy consumption threshold between the assets list and the data center. However, Ramakrishnan teaches:
determine… that an additional asset of the data center…is omitted from in the list of asset; (In C31; L19 – 34; Fig. 2 (200 and 283); Fig. 5 (506 – 508): teaches this negative limitation that does not hold patentable weight under the broadest reasonable interpretation (BRI), as described in “block 506” wherein the system’s “power-throttling learning module 283 may be trained to determine whether any given load-balancing instruction to divert workload away from a given hardware component or combination of hardware components will likely meet the QoS requirements for any JO commands from a host information handling system or hardware performance policies set for all hardware components within the data storage system/data center 210” wherein those hardware components that those not meet the criteria are not (or should not be) in the assets list and thus, may be omitted. Further, in C32; L3 – 18, “the power-throttling learning module 283 receives a notification that memory hardware is predicted to be underutilized during an upcoming time window, the power-throttling learning module 283 may transmit a memory-related load-balancing instruction to alternative memory hardware, meeting the QoS requirements for the data storage system/data center 210, that is also associated with a high reward value (e.g., positive one), instead of other memory-related load-balancing instructions being sent to the power-throttling agent 219 of the data center 210 that may not satisfy customer QoS requirements” which is directed to identifying assets that must be in the requirement lists and separating those assets that do not meet the requirements and therefore are predicted to be underutilized and omitted from the list.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to modify Kennedy to provide the ability of determining an asset of the data center not being in the list of assets that is based on the total energy consumption threshold between the assets list and the data center, as taught by Ramakrishnan in order to “identify an optimal load-balancing technique for rerouting any IO commands directed to the underutilized component(s) to other alternative hardware components that can satisfy customer [quality of service or] QoS requirements” (C29; L39 – 42; Ramakrishnan) as well as to “limit unnecessary power consumption and consequent carbon emissions by these underutilized hardware components by powering down capacity” (C3; L33 – 36; Ramakrishnan), see also MPEP 2143.I.G.
Ramakrishnan at least teaches the discovery module including new information (i.e. accurate tagged details for the additional asset) in a subsequent asset-discovery iteration when the system “receive monitoring period operational telemetry including updated information of the same type identified within the training period operational telemetry received at block 402” and upon receiving it and following training of the “the utilization forecast engine 286 and the power-throttling learning module 283”, “the power-throttling agent 219 at the data storage system/data center 210 may transmit updated operational telemetry, including performance metrics, and utilization rates to the RL based data center power consumption minimizing system 280” (see C27; L57 – 67; Ramakrishnan). However, neither Kennedy or Ramakrishnan explicitly teaches the ability of specifically include additional tags to the additional asset identified for accurate tagged details subsequent asset-discovery iterations. However, Katiyar teaches:
tag the additional asset with energy-consumption or carbon-emission information causing a discovery module to include accurate tagged details for the additional asset in a subsequent asset-discovery iteration (In ¶0019; Fig. 4: teaches that “the power management software 120 can be configured to receive emission factors from sources, such as public sources/websites or entered by a user. Implementations provide for a user interface, in the form of power management console 122 to interface power management software 120, allowing users to enter emission factor information, in particular when operating as a dark site or when the data center does not have access to sources/websites that provide emission factor information. The power management console 122 allows the ability to read and report energy consumption data of devices in a data center” which is directed to tagging energy-consumption or carbon-emission information as additional information provided by user input for an additional asset identified from the devices in a data center, in accordance with an asset tagging upon user input examples given in ¶0031 – 32 and ¶0037 from Applicant disclosure. Further, another additional asset tagging example is given in ¶0029, wherein “An owner/administrator of data center 202 may desire to use emission factors that they are aware of, instead of receiving of receiving emission factors from resources and databases, such as energy site(s) 214. In such instances, an override capability can be provided through the power management console 122” and enter “custom emission factors”, the emission factors desired values are provided by owner/administrator of data center 202”. See ¶0039 for Fig 4 details.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to modify Kennedy and Ramakrishnan to provide the ability of specifically include additional tags to the additional asset identified for accurate tagged details subsequent asset-discovery iterations, as taught by Katiyar in order to provide “Measured emissions [that] can be reported based on a timeseries stream for comparison and other purposes related to tracking progress towards improved GHG emissions or sustainability goals” as well as to provide the emission factors per device in case of a “dark site” that is “defined as a site/location, such as the data center, which does not have access to external resource information, such as lack of connectivity to the Internet” for the administrator or user to be able to manually include accurate tag details of their devices inventory in the data center (¶0013; Katiyar), see also MPEP 2143.I.G.
Regarding claim 2:
The combination of Kennedy, Ramakrishnan and Katiyar, as shown in the rejection above, discloses the limitations of claim 1.
Kennedy further teaches:
wherein the action comprises one or more of the following: decommissioning the identified asset, virtualizing the identified asset, and configuring the identified asset in balanced mode. (In ¶0066; Fig. 4 (404): teaches that “In response to determining that the present consumption level of the plurality of computing devices is above a first threshold, the power controller 110 decreases performance of the plurality of computing devices or decreases an operating level of an ancillary system that draws power from the same power source of the plurality of computing devices (404)” which is directed to at least the action of configuring the identified asset in balanced mode. Further, in ¶0041, discloses an example of decommissioning (i.e. the removal) of an identified asset wherein “In some implementations, should the power source 102 fail or become unreliable, the power controller 110 can utilize the switches 106 with control line 116 to isolate the data center facility 100 from the power source 102. In particular, the power feeds 103, 105, from the power source 102 and the inverter 206, respectively, can be switched off, and the power can be provided from the utility provider 150 through backup power feed 154”, in accordance to an asset decommissioning example given in ¶0054 from Applicant disclosure.)
Regarding claim 3:
The combination of Kennedy, Ramakrishnan and Katiyar, as shown in the rejection above, discloses the limitations of claim 1.
Kennedy further teaches:
wherein the total energy consumption is received from an energy provider. (In ¶0059: teaches “a power consumption profile for the data center facility 100 may be provided by another entity, such as the fuel cell system provider, or by the utility provider before the fuel system was implemented, if available”.)
Regarding claim 4:
The combination of Kennedy, Ramakrishnan and Katiyar, as shown in the rejection above, discloses the limitations of claim 1.
Kennedy further teaches:
wherein the energy consumption by each of the assets is measured via one of the following: an actual energy consumption, calculated energy consumption, estimated energy consumption, specification sheet of the asset, or an average of energy consumption of assets other than the asset. (In ¶0048; Fig. 2 (240 and 250): teaches the “Plot 250 includes a dotted line that depicts power consumption over time for the data center facility after the power controller 110 is implemented with the power controller 110 controlling the loads. The plot 250 also includes a solid line that represents the same power consumption of plot 240 for purposes of reference. The dotted line depicts the resulting power consumption as controlled by the power controller 110”. See ¶0065 and ¶0068 for more examples of energy consumption determinations.)
Regarding claim 5:
The combination of Kennedy, Ramakrishnan and Katiyar, as shown in the rejection above, discloses the limitations of claim 1.
Kennedy does not explicitly teach the abilities of determining that the identified asset(s) satisfies a criterion and automatically perform an action the identified asset. However, Ramakrishnan further teaches:
wherein the memory and computer program code are configured to further cause the processor to: determine that the identified asset satisfies a criterion; and (In C34; L43 – 49; Fig. 6 (606): teaches “at block 606 may execute the recommended or optimal load-balancing instruction to reroute incoming IO requests from a host computer directed to the predicted under-utilized hardware component(s) to alternative hardware components set to remain powered on and which may satisfy the IO requests QoS requirements”.)
upon determining that the identified asset satisfies the criterion, automatically perform the action on the identified asset. (In C35; L11 – 14; Fig. 6 (608): teaches “power-throttling agent in an embodiment may terminate or limit power supply to the predicted under-utilized hardware component during the upcoming time window at block 608”.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to modify Kennedy to provide the abilities of determining that the identified asset(s) satisfies a criterion and automatically perform an action the identified asset, as taught by Ramakrishnan in order to “identify an optimal load-balancing technique for rerouting any IO commands directed to the underutilized component(s) to other alternative hardware components that can satisfy customer [quality of service or] QoS requirements” (C29; L39 – 42; Ramakrishnan) as well as to “limit unnecessary power consumption and consequent carbon emissions by these underutilized hardware components by powering down capacity” (C3; L33 – 36; Ramakrishnan), see also MPEP 2143.I.G.
Regarding claim 6:
The combination of Kennedy, Ramakrishnan and Katiyar, as shown in the rejection above, discloses the limitations of claim 4.
Kennedy further teaches:
wherein the criterion comprises one of determining that the energy consumption by the identified asset is more than a threshold value or determining that the energy consumption by the identified asset is less than a threshold value. (In ¶0065: teaches “power controller 110 determines that a present consumption level of the plurality of computing devices is above a first threshold (402)”. See ¶0068 for more examples of energy consumption determinations when determining that the energy consumption by the identified asset(s) is less than a threshold value.)
Regarding claim 7:
The combination of Kennedy, Ramakrishnan and Katiyar, as shown in the rejection above, discloses the limitations of claim 4.
Kennedy does not explicitly teach the ability of having a criterion that specifically determines identified asset(s) as a virtual machine, if it’s not running a database or both. However, Ramakrishnan further teaches:
wherein the criterion comprises determining that the identified asset is a virtual machine, determining that the identified asset is not running a database, or both. (In C11; L9 – 21; Fig. 2 (270 and 210): teaches that “a reinforcement learning (RL) based data center power consumption minimizing system 280 executing on a unified endpoint management (UEM) platform 200 information handling system for identifying data storage system/data center 210 hardware components that are predicted to be underutilized during an upcoming time window” wherein the “host computer 270 may be implemented as a virtual machine within storage system 210” (see C11; L53 – 56) which such determination claimed is further defining descriptive material.)
It would have been obvious to one of ordinary skill in the art before the earliest effective filing date of the claimed invention to modify Kennedy to provide the ability of having a criterion that specifically determines identified asset(s) as a virtual machine, if it’s not running a database or both, as taught by Ramakrishnan in order to “identify an optimal load-balancing technique for rerouting any IO commands directed to the underutilized component(s) to other alternative hardware components that can satisfy customer [quality of service or] QoS requirements” (C29; L39 – 42; Ramakrishnan) as well as to “limit unnecessary power consumption and consequent carbon emissions by these underutilized hardware components by powering down capacity” (C3; L33 – 36; Ramakrishnan), see also MPEP 2143.I.G.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Aurongzeb (U.S. Pub No. 20240036999 A1) is pertinent because it “relates to a method of predicting future occurrence of hardware failures based on presence of problematic adjustable system configurations and adjusting those problematic adjustable system configurations prior to such hardware failure to avoid such future failures.”
Lange (U.S. Pub No. 20250045772 A1) is pertinent because it “relate to optimizing the carbon efficiency of a data center, and more specifically, monitoring a carbon efficiency metric associated with the data center as the data center is used (e.g., operated) and providing a recommendation of a change for the data center, such as substituting a component of the data center with alternative hardware, to improve the carbon efficiency metric.”
Moore (U.S. Pub No. 20100191998 A1) is pertinent because it “relate to, among other things, calculating and apportioning an environmental impact associated with the operation of a data center. One or more data centers are identified and the environmental impacts attributable to the data centers are determined. By way of example and not limitation, the carbon dioxide emissions can be and are apportioned on the basis of a data center application. Accordingly, the present invention permits apportioning an environmental impact on a per application basis.”
Mallet (U.S. Pub No. 20240281826 A1) is pertinent because it “relates to CO2e and power footprint tracking systems.”
Rattihalli (U.S. Pub No. 20250265527 A1) is pertinent because it is a “method includes generating a first data center graph including a plurality of host graphs, the host graphs representing resources of hosts of a data center, combining the first data center graph with an application signature graph to produce a second data center graph, the second data center graph representing the resource utilization of an application when running on the data center, predicting an energy consumption of the application when running on the data center by processing the second data center graph using a graph neural network (GNN), and scheduling the application on the hosts of the data center based on the predicted energy consumption of the application.”
Karia (U.S. Pub No. 20250139643 A1) is pertinent because it “relates to the field of data processing. More specifically, the present disclosure relates to methods and systems for facilitating managing clouds.”
Roper (U.S. Patent No. 12373851 B2) is pertinent because it “relates to the secure provision and usage of tools for digital engineering, e.g., including modeling and simulation tools, and certification of digitally engineered products.”
Reynolds (U.S. Pub No. 20250293542 A1) is pertinent because it “relate to a control system that utilizes a machine learning model to protect a data center during an environmental failure.”
Magcale (U.S. Pub No. 20190033945 A1) is pertinent because it “generally relates to data centers, in particular to methods and systems to evaluate data center performance, assess data center efficiency, data center sustainability, data center availability, compute performance, storage performance and to provide data center customers with an overall data center performance rating.”
Mohanbabu (U.S. Pub No. 20230259859 A1) is pertinent because it is “directed to various innovative technologies for managing, processing, and generating display information relating to sustainability footprint calculations.”
Applicant's amendment necessitated the new ground(s) of rejection presented in this Office action. Accordingly, THIS ACTION IS MADE FINAL. See MPEP § 706.07(a). Applicant is reminded of the extension of time policy as set forth in 37 CFR 1.136(a).
A shortened statutory period for reply to this final action is set to expire THREE MONTHS from the mailing date of this action. In the event a first reply is filed within TWO MONTHS of the mailing date of this final action and the advisory action is not mailed until after the end of the THREE-MONTH shortened statutory period, then the shortened statutory period will expire on the date the advisory action is mailed, and any nonprovisional extension fee (37 CFR 1.17(a)) pursuant to 37 CFR 1.136(a) will be calculated from the mailing date of the advisory action. In no event, however, will the statutory period for reply expire later than SIX MONTHS from the mailing date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Ivonnemary Rivera Gonzalez whose telephone number is (571)272-6158. The examiner can normally be reached Mon - Fri 9:00AM - 5:30PM.
Examiner interviews are available via telephone, in-person, and video conferencing using a USPTO supplied web-based collaboration tool. To schedule an interview, applicant is encouraged to use the USPTO Automated Interview Request (AIR) at http://www.uspto.gov/interviewpractice.
If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Jessica Lemieux can be reached at (571) 270-3445. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300.
Information regarding the status of published or unpublished applications may be obtained from Patent Center. Unpublished application information in Patent Center is available to registered users. To file and manage patent submissions in Patent Center, visit: https://patentcenter.uspto.gov. Visit https://www.uspto.gov/patents/apply/patent-center for more information about Patent Center and https://www.uspto.gov/patents/docx for information about filing in DOCX format. For additional questions, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/IVONNEMARY RIVERA GONZALEZ/Examiner, Art Unit 3626 /DENNIS W RUHL/Primary Examiner, Art Unit 3626
Prosecution Insights