CTNF 18/184,712 CTNF 85261 DETAILED ACTION Notice of Pre-AIA or AIA Status 07-03-aia AIA 15-10-aia The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA. Claim Rejections - 35 USC § 102 07-06 AIA 15-10-15 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. 07-07-aia AIA 07-07 The following is a quotation of the appropriate paragraphs of 35 U.S.C. 102 that form the basis for the rejections under this section made in this Office action: A person shall be entitled to a patent unless – 07-08-aia AIA (a)(1) the claimed invention was patented, described in a printed publication, or in public use, on sale, or otherwise available to the public before the effective filing date of the claimed invention. 07-15 AIA Claim s 1-3, 7-14 and 18-20 are rejected under 35 U.S.C. 102( a)(1 ) as being anticipated by Power Modeling of Effective Datacenter Planning and Compute Management (hereafter referred to as Radovanovic et al) . Regarding claim 1, Radovanovic et al teaches a method for estimating energy consumption of a workload in a Cloud computing system, comprising: receiving a request for an estimated energy consumption of the workload ( see pg. 1, Introduction, The models are designed with the following use cases in mind: 1) Real-time estimation of datacenter power consumption and its electricity based carbon footprint) ; obtaining characteristics of the Cloud computing system executing the workload (see pg. 3, The power modeling pipeline, read historical resource and power usage data at 5-minute granularity, as well as hardware characteristics of each machine/PDU) ; identifying and employing a unified power consumption model from a power model database based on the characteristics (see pg. 1 Abstract, The models have been designed (and already deployed in production) based on: 1) individual machine level data, and 2) the entire MW scale datacenter Power Distribution Unit (PDU) level) ; and calculating the estimated energy consumption of the workload based on the unified power consumption model (see pg. 1, Introduction, two types of accurate and light-weight power models are developed to characterize power consumption of a datacenter based on workload CPU usage). Regarding claim 2, Radovanovic et al teaches, wherein the power model database includes a plurality of unified power consumption models, including the unified power consumption model, and wherein each unified power consumption model is created based on collected energy consumption data for different Cloud computing configurations ( see pg. 2, Introduction, usage to its power consumption at 5-minute time granularity. We develop two types of light-weight statistical power models using the vast amount of data across all Google’s datacenter PDUs with varying hardware configurations, machine platforms and workload types). Regarding claim 3, Radovanovic et al teaches, wherein each of the Cloud computing configurations include metadata that identifies hardware and software associated with the Cloud computing configuration (see abstract, an accurate mapping of the datacenter’s compute resources (CPU, RAM, etc.) and hardware types (servers, accelerators, etc.) to power consumption has emerged as an essential requirement for major Web and cloud service providers). Regarding claim 7, Radovanovic et al teaches, wherein the characteristics of the Cloud computing system include a specification of a server of the Cloud computing system, and an identification of a software platform of the Cloud computing system (see pg.2-3, It is demonstrated that the generality and effectiveness in provisioning PDU power is achievable using only basic hardware and resource usage characteristics (such as CPU usage or CPU utilization, i.e., CPU usage divided by the total CPU capacity). It should be mentioned that 3 the models discussed in this paper have been deployed and used for more than a year). Regarding claim 8, Radovanovic et al teaches, wherein the estimated energy consumption of the workload is calculated based by applying a monitored activity of the workload on the Cloud computing system to the unified power consumption model (see pg. 1, Introduction, two types of accurate and light-weight power models are developed to characterize power consumption of a datacenter based on workload CPU usage). Regarding claim 9, Radovanovic et al teaches, wherein one or more parameters of the unified power consumption model are adjusted based on a monitored activity of the workload (see pg. 5, The recurrent, daily, training is done to adapt the model in case of hardware changes (deployment/decommission of servers). The training instances are weighted based on their recency, i.e., higher weights are assigned to more recent measurements to ensure proper adaptation to systematic changes in the fleet (e.g., hardware upgrades and new deployments). In particular, the pipeline uses 1/1+d to weigh data instances from d days ago). Regarding claim 10, Radovanovic et al teaches, further comprising verifying an accuracy of the unified power consumption model by performing autonomous validation of the unified power consumption model using data obtained from hardware vendors corresponding to hardware associated with the unified power consumption model and using one or more power consumption benchmarking tools (see pg. 7, To that end, Mean Absolute Percent Error (MAPE) is chosen to evaluate performance of the power models). Regarding claim 11, Radovanovic et al teaches, where the power model database is separate from the Cloud computing system (see pg.9, Conclusion, The two types of models (already deployed in production) are based on: 1) individual machine level data, and 2) the entire MW-scale datacenter PDU level). Regarding claim 12, Radovanovic et al teaches a computing system having a memory having computer readable instructions and one or more processors for executing the computer readable instructions, the computer readable instructions controlling the one or more processors to perform operations comprising: receiving a request for an estimated energy consumption of a workload ( see pg. 1, Introduction, The models are designed with the following use cases in mind: 1) Real-time estimation of datacenter power consumption and its electricity based carbon footprint) ; obtaining characteristics of a Cloud computing system executing the workload (see pg. 3, The power modeling pipeline, read historical resource and power usage data at 5-minute granularity, as well as hardware characteristics of each machine/PDU) ; identifying and employing a unified power consumption model from a power model database based on the characteristics (see pg. 1 Abstract, The models have been designed (and already deployed in production) based on: 1) individual machine level data, and 2) the entire MW scale datacenter Power Distribution Unit (PDU) level) ; and calculating the estimated energy consumption of the workload based on the unified power consumption model (see pg. 1, Introduction, two types of accurate and light-weight power models are developed to characterize power consumption of a datacenter based on workload CPU usage). Regarding claim 13, Radovanovic et al teaches, wherein the power model database includes a plurality of unified power consumption models, including the unified power consumption model, and wherein each unified power consumption model is created based on collected energy consumption data for different Cloud computing configurations ( see pg. 2, Introduction, usage to its power consumption at 5-minute time granularity. We develop two types of light-weight statistical power models using the vast amount of data across all Google’s datacenter PDUs with varying hardware configurations, machine platforms and workload types). Regarding claim 14, Radovanovic et al teaches, wherein each of the Cloud computing configurations include metadata that identifies hardware and software associated with the Cloud computing configuration (see abstract, an accurate mapping of the datacenter’s compute resources (CPU, RAM, etc.) and hardware types (servers, accelerators, etc.) to power consumption has emerged as an essential requirement for major Web and cloud service providers). Regarding claim 18, Radovanovic et al teaches, wherein the characteristics of the Cloud computing system include a specification of a server of the Cloud computing system, and an identification of a software platform of the Cloud computing system (see pg.2-3, It is demonstrated that the generality and effectiveness in provisioning PDU power is achievable using only basic hardware and resource usage characteristics (such as CPU usage or CPU utilization, i.e., CPU usage divided by the total CPU capacity). It should be mentioned that 3 the models discussed in this paper have been deployed and used for more than a year). Regarding claim 19, Radovanovic et al teaches, wherein the estimated energy consumption of the workload is calculated based by applying a monitored activity of the workload on the Cloud computing system to the unified power consumption model (see pg. 1, Introduction, two types of accurate and light-weight power models are developed to characterize power consumption of a datacenter based on workload CPU usage). Regarding claim 20, Radovanovic et al teaches a computer program product comprising a computer readable storage medium having program instructions embodied therewith, the program instructions executable by a processor to cause the processor to perform operations comprising: receiving a request for an estimated energy consumption of a workload ( see pg. 1, Introduction, The models are designed with the following use cases in mind: 1) Real-time estimation of datacenter power consumption and its electricity based carbon footprint) ; obtaining characteristics of a Cloud computing system executing the workload (see pg. 3, The power modeling pipeline, read historical resource and power usage data at 5-minute granularity, as well as hardware characteristics of each machine/PDU) ; identifying and employing a unified power consumption model from a power model database based on the characteristics (see pg. 1 Abstract, The models have been designed (and already deployed in production) based on: 1) individual machine level data, and 2) the entire MW scale datacenter Power Distribution Unit (PDU) level) ; and calculating the estimated energy consumption of the workload based on the unified power consumption model (see pg. 1, Introduction, two types of accurate and light-weight power models are developed to characterize power consumption of a datacenter based on workload CPU usage) . Claim Rejections - 35 USC § 103 07-06 AIA 15-10-15 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. 07-20-aia AIA 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. 07-21-aia AIA Claim s 4-7 and 15-17 are rejected under 35 U.S.C. 103 as being unpatentable over Radovanovic et al and in further view of “Workload forecasting and energy state estimation in cloud data centres: ML-centric approach” (hereafter referred to as Khan et al) . Regarding claims 4 and 15, Radovanovic et al teaches all the limitations of claim 1 and 12 from which claims 4 and 15 respectively depend on. However, they fail to explicitly teach calculating a similarity score as further recited in the claims. Conversely Khan et al teaches such limitations; wherein identifying the unified power consumption model from the power model database based on the characteristics includes calculating a similarity score between the characteristics and a Cloud computing configuration associated with each power consumption model stored on the power model database (see pg. 327, In detail, this information is used to compute a similarity matrix using pairwise euclidean distance with n number of data points, resulting in a n × n similarity matrix S). Therefore it would have been obvious to combine the teachings of Radovanovic et al with the use of a calculating a similarity score as taught by Khan et al. The motivation for this would have been provide efficient resource management in data centers (see Abstract). Regarding claims 5 and 16, Khan et al further teaches wherein identifying the unified power consumption model from the power model database based on the characteristics further includes selecting a power consumption model associated with having a highest similarity score (see pg. 325, Kmeans [16]: This algorithm aims to group n data points into K classes, with each data point being a neighbour of the cluster centre closest to it). Regarding claims 6 and 17, Khan et al further teaches wherein identifying the unified power consumption model from the power model database based on the characteristics further includes selecting a default power consumption model based on a determination that a highest similarity score is below a threshold value (see pg. 327, To improve accuracy, the preference parameter p can be tweaked with different input values. In our case, p = min(S) / iter × 0.3 provides optimal performance). Conclusion Any inquiry concerning this communication or earlier communications from the examiner should be directed to SHRIPAL K KHAJURIA whose telephone number is (571)270-5662. The examiner can normally be reached Monday - Friday 9:30AM - 6:00PM. 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, Jonathan Horner can be reached at (571)270-7358. 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. /SHRIPAL K KHAJURIA/Primary Examiner, Art Unit 6221 Application/Control Number: 18/184,712 Page 2 Art Unit: 6221 Application/Control Number: 18/184,712 Page 3 Art Unit: 6221 Application/Control Number: 18/184,712 Page 4 Art Unit: 6221 Application/Control Number: 18/184,712 Page 5 Art Unit: 6221 Application/Control Number: 18/184,712 Page 6 Art Unit: 6221 Application/Control Number: 18/184,712 Page 7 Art Unit: 6221 Application/Control Number: 18/184,712 Page 8 Art Unit: 6221 Application/Control Number: 18/184,712 Page 9 Art Unit: 6221
Prosecution Insights