DETAILED ACTION
This action is in response to application 18/113162, filed on 2/23/2023. Claims 1-20 are pending. 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 § 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. Claims 1 and 11 recite the limitations:
PNG
media_image1.png
279
685
media_image1.png
Greyscale
Claim 16 recites the following limitations:
PNG
media_image2.png
112
638
media_image2.png
Greyscale
PNG
media_image3.png
78
640
media_image3.png
Greyscale
PNG
media_image4.png
79
644
media_image4.png
Greyscale
These limitations, as drafted, describe a process that, under its broadest reasonable interpretation, is capable of being performed solely in the mind. That is, other than reciting "at least one processor" in claims 1 and 16, and the “at least one processing device” of claim 11, nothing in the claim elements precludes the steps from practically being performed in the mind. The claims do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the additional elements when considered both individually and as an ordered combination do not amount to significantly more than the abstract idea. Generic computer components recited as performing generic computer functions that are well-understood, routine and conventional activities amount to no more than implementing the abstract idea with a computerized system. Thus, taken alone, the additional elements do not amount to significantly more than the above-identified judicial exception (the abstract idea).
Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology or effects a transformation or reduction of a particular article to a different state or thing. Their collective functions merely provide conventional computer implementation. Furthermore, the applicant's own specification details the generic nature of the computing components, which also precludes them from presenting anything significantly more (p. 19, In. 11 - p. 20, In. 6).
Claims 2-10, 12-15, and 17-20 do not include additional elements that are sufficient to amount to significantly more than the judicial exception because the additional elements when considered both individually and as an ordered combination do not amount to significantly more than the abstract idea. Generic computer components recited as performing generic computer functions that are well-understood, routine and conventional activities amount to no more than implementing the abstract idea with a computerized system. Thus, taken alone, the additional elements do not amount to significantly more than the above-identified judicial exception (the abstract idea). Looking at the limitations as an ordered combination adds nothing that is not already present when looking at the elements taken individually. There is no indication that the combination of elements improves the functioning of a computer or improves any other technology. Their collective functions merely provide conventional computer implementation.
Claim Rejections - 35 USC § 102
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 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 –
(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.
Claims 1-3, 8-13, and 16-18 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Zhang et al., “CRISP: Critical Path Analysis of Large-Scale Microservice Architectures,” USENIX, 2022, hereinafter “Zhang.”
Regarding claim 1, Zhang anticipates “An apparatus comprising: at least one processing platform comprising at least one processor coupled to at least one memory, the at least one processing platform, when executing program code, is configured to:
obtain an indication of at least one anomalous behavior associated with execution of an application in an information processing system, wherein the application comprises a plurality of services; (see, e.g., Zhang, pg. 656; “Practical deployment of CRISP at Uber over a three-month period working on 40K endpoints while processing ~200GB of traces with ~18 million spans in ~256 hours of CPU time per day”; “An automated anomaly detection system, which detects whether a request is exhibiting abnormal behavior compared with the past history of the endpoint.”; “The system is trained per endpoint using an autoencoder-decoder machine learning technique. This system is set up to expedite problem detection and alert developers. Basing the abnormality detection on the divergence in the critical path as opposed to the full call graph not only makes the training and inference faster but also reduces false alerts.”; para. 664, sec. 7.3; “We collect traces for six microservices in real production over a 14-day period. The training data for each case includes 20,000 traces and the testing data has 500 unseen traces for normal and abnormal data.”)
analyze, across a plurality of time periods, at least one metric associated with the execution of the application to determine at least one critical path associated with the execution of the application, wherein the critical path comprises at least a portion of the plurality of services; (see, e.g., Zhang, pg. 661, sec. 5.3; “While one trace can be compressed into its essential critical path and represented as a CCCT, it may not be representative. Hence, we need to inspect numerous traces to derive a “typical” shape of the critical path. Distinct traces may exhibit different critical paths based on many things, such as calling parameters, scheduling decisions, system load, time of the day, and network delays, to name a few.”; “The aggregate CCCT succinctly summarizes all call paths leading to critical path nodes in all traces; it captures the quantitative aspect by associating higher weights to those call paths that are often on the critical path.”)
analyze the critical path using a set of variance correlation algorithms; (see, e.g., Zhang, pg. 664, sec. 7.3; “We use TraceAnomaly as the baseline against which we compare our results. We adopt the same architecture of the autoencoder and reuse their code. The main difference is that we use CRISP to preprocess the trace before feeding it into the autoencoder so that only paths appearing on the critical path information are included.”) and
identify a set of one or more services in the critical path that are highest in a ranked order determined by the set of variance correlation algorithms, wherein the identified set of one or more services is considered to be associated with the anomalous behavior.” (see, e.g., Zhang, pg. 665, sec. 7.3; “The recall is the part that differentiates the quality of results between TraceAnomaly and CRISP. Recall measures how many of the actual positives the model captures through labeling it as positive . . . When the recall is closer to 1, it indicates that the model makes fewer false-positive predictions (an anomaly in this case). From Table 2, it is clear that CRISP outperforms TraceAnomaly by a noticeable margin. Particularly for Service 3 and 5, half of the positive prediction of the anomaly is false, meaning all normal traces for inference are labeled abnormal by TraceAnomaly.”; “For Service 1 and 2, the performance of CRISP is slightly better than TraceAnomaly, as both models make relatively accurate predictions. CRISP shows more than 5% improvement for Service 4 and 6. CRISP produces superior results on services with a large number of call paths.”).
Regarding claim 2, Zhang anticipates “The apparatus of claim 1, wherein the plurality of time periods comprises a first time period prior to detection of the anomalous behavior, a second time period during detection of the anomalous behavior, and a third time period after activation of a trace on the application.” (see, e.g., Zhang, pg. 662, sec. 6.3; “We also employ CRISP to pinpoint whether a new incoming trace (for a given endpoint) deviates from the normal execution behavior. For this purpose, we have trained a machine learning model and used it for inference. During the offline training, we encode the critical path (CCCT) for each trace of an endpoint into feature vectors, which we call service critical path vectors (SCPV). We feed several SCPVs into an autoencoder to learn the normal execution pattern of the given service. During the online inference, the learned model will infer whether the given new trace is abnormal or not based on an anomaly score.”).
Regarding claim 3, Zhang anticipates “The apparatus of claim 1, wherein analyzing the at least one metric to determine the at least one critical path comprises utilizing a reinforcement learning algorithm.” (see, e.g., Zhang, pg. 670, Appendix B; “We choose the Deep Bayesian Network for anomaly detection given it is capable of learning complex patterns from the trace.”).
Regarding claim 8, Zhang anticipates “The apparatus of claim 1, wherein the plurality of services comprises microservices.” (see, e.g., Zhang, pg. 655, sec. 1; “Uber’s backend is an exemplar of microservice architecture.”).
Regarding claim 9, Zhang anticipates “The apparatus of claim 1, wherein the information processing system comprises a distributed edge system.” (see, e.g., Zhang, pg. 657, sec. 2; “Trusted edge devices (e.g., company mobile app) can validate at the edge improving performance for trusted users and falling back to server validation if the fare has expired”).
Regarding claim 10, Zhang anticipates “The apparatus of claim 9, wherein the distributed edge system is part of a multicloud edge platform.” (see, e.g., Zhang, pg. 656, sec. 2; “Fulfillment1 at Uber is a platform to orchestrate and manage the lifecycle of orders and user sessions with millions of active participants. The Fulfillment platform is a foundational Uber capability that enables the rapid scaling of new verticals. The platform handles more than a million concurrent users and billions of trips per year that span over ten thousand cities. The platform handles billions of database transactions a day. Hundreds of Uber microservices rely on the platform as the source of truth for the accurate state of the trips and driver or delivery sessions. Events generated by the platform are used to build hundreds of offline datasets to make critical business decisions. Over 500 developers extend the platform using APIs, events, and code to build more than 120 unique fulfillment flows.”).
Regarding claims 11-13 and 16-18, the instant claims are equivalents of claims 1-3, differing only by statutory class. Accordingly, the rejections of claims 1-3 apply, mutatis mutandis, to claims 11-13, respectively, and to claims 16-18, respectively.
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.
The factual inquiries for establishing a background for determining obviousness under 35 U.S.C. 103 are summarized as follows:
1. Determining the scope and contents of the prior art.
2. Ascertaining the differences between the prior art and the claims at issue.
3. Resolving the level of ordinary skill in the pertinent art.
4. Considering objective evidence present in the application indicating obviousness or nonobviousness.
Claim 4 is rejected under 35 U.S.C. 103 as being unpatentable over Zhang and Zhu et al., “On Improving Deep Reinforcement Learning for POMDPs,” arXiv, 2018, hereinafter “Zhu.”
Regarding claim 4, Zhang discloses “The apparatus of claim 3, wherein the reinforcement learning algorithm comprises utilizing a Deep Bayesian Network to analyze a graph associated with the execution of the application to extract a critical path for each of the plurality of time periods.” (see, e.g., Zhang, pg. 670, Appendix B; “We choose the Deep Bayesian Network for anomaly detection given it is capable of learning complex patterns from the trace.”; pg. 661, sec. 6.1; pg. 665, sec. 7.3).
Zhang does not appear to disclose the underlined portion of the limitation:
wherein the reinforcement learning algorithm comprises utilizing a Deep Recurrent Q Network (DRQN) to analyze a graph associated with the execution of the application to extract a critical path for each of the plurality of time periods.
However, Zhu discloses (at pg. 2) “a new architecture called Action-based Deep Recurrent Q-Network (ADRQN) to improve learning performance in partially observable domains.” Zhang and Zhu are directed toward machine learning and therefore are analogous art. On or before the effective filing date of the instant application, one of ordinary skill in the art would have deemed it obvious to try to combine the DRQN of Zhu with the machine learning system of Zhu, thereby obtaining the invention of the instant claim. A clear and predictable benefit of so combining would have appeared as the ability to improve learning performance. Accordingly, the instant claim is unpatentable over the combination of Zhang and Zhu.
Claims 5-7, 14-15, and 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over Zhang and Alhowaide et al., “PCA, Random-Forest and Pearson Correlation for Dimensionality Reduction in IoT IDS,” IEEE, 2020, hereinafter “Alhowaide.”
Regarding claim 5, Zhang discloses “The apparatus of claim 1,” but does not appear to disclose the limitation “wherein the set of variance correlation algorithms comprises a random forest classification-based algorithm, wherein a variance correlation result is computed by the random forest classification-based algorithm at the application level.” However, Alhowaide discloses (at sec. III(B)) an Intrusion Detection System that uses a random forest classification-based algorithm to compute a variance correlation result. Alhowaide and Zhang are directed toward monitoring software and therefore are analogous art. On or before the effective filing date of the instant application, one of ordinary skill in the art would have deemed it obvious to combine the random forest classification-based algorithm of Alhowaide with the anomalous activity detection of Zhang, thereby obtaining the invention of the instant claim. A clear and predictable benefit of so combining would have appeared as the ability to “find the best dimensionality reduction method that reduces a dataset to its minimum with the lowest computational requirements.” (Alhowaide, sec. I). Accordingly, the instant claim is unpatentable over the combination of Zhang and Alhowaide.
Regarding claim 6, the combination of Zhang and Alhowaide renders obvious “The apparatus of claim 5, wherein the set of variance correlation algorithms comprises a Pearson correlation coefficient-based algorithm, wherein a variance correlation result is computed by the Pearson correlation coefficient-based algorithm at the service level.” (Alhowaide, sec. III(B); “This method uses the Pearson correlation measure to rank the features. Then, it selects the top n features.”).
Regarding claim 7, the combination of Zhang and Alhowaide renders obvious “The apparatus of claim 6, wherein the ranked order determined by the set of variance correlation algorithms is generated by weighting the respective variance correlation results associated with the random forest classification-based algorithm and the Pearson correlation coefficient-based algorithm.” (Alhowaide, sec. IV; “TABLE 1 shows the obtained percentage of size reduction and the number of chosen features for each dataset”).
Regarding claims 14-15 and 19-20, the instant claims are equivalents of claims 5-6, differing only by statutory class. Accordingly, the rejections of claims 5-6 apply, mutatis mutandis, to claims 14-15, respectively, and to claims 19-20, respectively.
Conclusion
Any inquiry concerning this communication or earlier communications from the examiner should be directed to RYAN D. COYER whose telephone number is (571) 270-5306 and whose fax number is (571) 270-6306. The examiner normally can be reached via phone on Monday-Friday 12pm-10pm Eastern Time. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Wei Mui, can be reached on 571-272-3708. The fax phone number for the organization where this application or proceeding is assigned is 571-273-8300. Information regarding the status of an application may be obtained from the Patent Application Information Retrieval (PAIR) system. Status information for published applications may be obtained from either Private PAIR or Public PAIR. Status information for unpublished applications is available through Private PAIR only. For more information about the PAIR system, see http://pair-direct.uspto.gov. Should you have questions on access to the Private PAIR system, contact the Electronic Business Center (EBC) at 866-217-9197 (toll-free). If you would like assistance from a USPTO Customer Service Representative or access to the automated information system, call 800-786-9199 (IN USA OR CANADA) or 571-272-1000.
/Ryan D. Coyer/Primary Examiner, Art Unit 2191
1 https://www.uber.com/blog/fulfillment-platform-rearchitecture/