DETAILED ACTION
Notice of Pre-AIA or AIA Status
1. 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
2. Claims 1-17 are presented for examination.
Claim Rejections - 35 USC § 112
3. The rejection of claims 1-17 under 35 U.S.C. § 112, second paragraph, is withdrawn in view of applicant's amendments/remarks.
Request for Continued Examination
4. A request for continued examination under 37 CFR 1.114, including the fee set forth in 37 CFR 1.17(e), was filed in this application after final rejection. Since this application is eligible for continued examination under 37 CFR 1.114, and the fee set forth in 37 CFR 1.17(e) has been timely paid, the finality of the previous Office action has been withdrawn pursuant to 37 CFR 1.114. Applicant’s submission filed on 05/20/2026 has been entered.
Response to Arguments
5. Applicant’s argument filed on 04/20/2026 with respect claims 1-17 have been fully considered but they are not persuasive.
The applicant contends that the office action fails to teach or suggest the limitation of “determining a subset of the test data from the test data, the subset including specific test data based on which the SSD has been determined to have the anomaly.” As recited in claim 1 and similar feature in claim 9. The Examiner respectfully disagrees and asserts the reference of Gaber et al. (US 10,216,558 B1) in column 3, lines 48-67, column 6, lines 38-57, and Fig. 6. For example, FIG. 1 is a block diagram illustrating an overview 100 of a system to predict drive failures in which different types of disk drives, such as Hard Disk Drives (HDD) 102 and Solid State Drives (SSD) 106 are equipped with SMART monitors 104/108 to self-monitor and report SMART attributes. In one embodiment, a raw sample SMART data collector module 110 collects samples of the SMART attributes reported by drives 102/106 and relays them to a first application of a feature selection machine learning model 114. The feature selection machine learning model 114 is applied to the last collected sample of SMART attributes to generate a snapshot of drive features at a single point in time. In one embodiment, the SMART data collector collects consecutive samples, or historical samples, of SMART attributes over a specific time period, and relays them to a second application of a machine learning model, this time a feature engineering machine learning model 116. The feature engineering machine learning model 116 generates one or more drive behavior history features, also referred to as an engineered, or new feature, that are added to the drive features generated in the first application of the feature selection machine learning model 114. See column 3, lines 48-67.In one embodiment, at 605, for drives predicted to fail, the drive is identified as predicted to fail, such as by flagging the drive by its drive identifier as belonging to the classification of predicted to fail in the drive behavior features data sets repository 206. At 606, the process 600 is repeated until all drives have been classified. With reference to the embodiment of the process 700 illustrated in FIG. 7, once the drive's behavior has been modeled with the latest collected samples of drive health indicators and the drive classified as predicted to fail as illustrated in FIGS. 5 and 6, additional processing and analysis of the modeling results are performed in a separate process. For example, at process 700 at 702 obtains the identification of those drives that are currently predicted to fail. At 703 process 700 optionally generates a prediction that specifies the average predicted time before failure. As with the decision at decision block 604 in FIG. 6, the time before failure prediction can be based on one or more features in combination with one another, or on the aggregated predictions as described in FIG. 2. column 6, lines 38-57. Also see Fig. 6 printed below for your convenience.
PNG
media_image1.png
707
519
media_image1.png
Greyscale
Also, the applicant contends that the office action fails to teach or suggest the limitation of “identifying an anomaly cause of the SSD based on the subset of the test data.” As recited in claim 1 and similar feature in claim 9. The Examiner notes that the applicant’s arguments regarding to the above limitations have been considered but are moot in view of the new ground(s) of rejection. In addition to, the Examiner maintained the reference of Gaber et al. (US 10,216,558 B1) since there is no further argument/s regarding to this reference.
Claim Rejections - 35 USC § 103
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.
6. Claims 1-5, 9-13, and 17 are rejected under 35 U.S.C. 103 (a) as being unpatentable over Gaber et al. (US 10,216,558 B1) "herein after as Gaber" in view of Guo et al. (US 2013/0318392 A1) “herein after as Guo.”
As per claims 1 and 9:
Gaber substantially teaches or discloses an anomaly detection processing method for solid-state drive (SSD), comprising (see abstract): collecting test data of an SSD (see column 3, lines 53-56, herein a raw sample SMART data collector module 110 collects samples of the SMART attributes reported by drives 102/106 and relays them to a first application of a feature selection machine learning model 114; and column 6, lines 25-30, herein the process 600 at 602 obtains the feature dataset for the drive from the drive behavior feature datasets repository 206. At 603, process 600 analyzes the drive behavior based on a combination of the raw features and the historical features stored in the drive's features dataset), the test data including at least one of self-monitoring, analysis and reporting technology (S.M.A.R.T.) data, NAND flash cell threshold voltage distribution data, and bit error rate eye diagram data (see column 3, lines 53-56, herein a raw sample SMART data collector module 110 collects samples of the SMART attributes reported by drives 102/106 and relays them to a first application of a feature selection machine learning model 114); determining whether the SSD has an anomaly based on the test data (see column 6, lines 30-34, herein at decision block 604, the process 600 classifies the drive as predicted to fail when the individual drive failure probabilities predicted by the model exceed a certain threshold or meet other criteria for classifying failed and healthy drives); determining a subset of the test data from the test data, the subset including specific test data based on which the SSD has been determined to have the anomaly (see column 3, lines 48-59, herein a raw sample SMART data collector module 110 collects samples of the SMART attributes reported by drives 102/106 and relays them to a first application of a feature selection machine learning model 114. The feature selection machine learning model 114 is applied to the last collected sample of SMART attributes to generate a snapshot of drive features at a single point in time; and column 6, lines 39-43, herein for drives predicted to fail, the drive is identified as predicted to fail, such as by flagging the drive by its drive identifier as belonging to the classification of predicted to fail in the drive behavior features data sets repository 206, and Fig. 6 step 605). Gaber does not explicitly teach identifying an anomaly cause of the SSD based on the subset of the test data. However, Guo in the same the field of endeavor teaches identifying an anomaly cause of the SSD based on the subset of the test data (see paragraph [0072], the monitoring unit (sign detection unit) 112 detects a sign of a fault in the SSD 20. Specifically, the monitoring unit 112 detects a sign of a fault in the SSD 20 using at least some of the S.M.A.R.T. values (self-diagnostic history information) notified by the controller 21 of the SSD 20) . Therefore, it would have been obvious to one of ordinary skill in the art, before the effective filing date of the invention, to modify the system of Gaber with the teachings of Guo by identifying an anomaly cause of the SSD based on the subset of the test data. This modification would have been obvious to one of ordinary skill in the art, before the effective filing date of the invention, because one of ordinary skill in the art would have recognized the identifying an anomaly cause of the SSD based on the subset of the test data would have improved reliability of data (see paragraph [0166] of Guo) .
As per claims 2 and 10: Guo teaches that wherein the determining whether the SSD has an the anomaly comprises: determining whether the SSD has the anomaly, based on the S.M.A.R.T. data by using a trained first anomaly detection model, based on the NAND flash cell threshold voltage distribution data by using a second anomaly detection model, or based on the bit error rate eye diagram data by using a trained third anomaly detection model (see paragraph [0073], herein when detecting that the current value or the worst value is an abnormal value in at least some of the attributes in the S.M.A.R.T. values, the monitoring unit 112 detects a sign of a fault of the SSD 20); or determining whether the SSD has the anomaly based on the S.M.A.R.T. data, the NAND flash cell threshold voltage distribution data or the bit error rate eye diagram data, by using a trained anomaly detection model (see paragraph [0074], herein The abnormal value in the S.M.A.R.T. values as used here means that the value is more than or less than a THRESHOLD value (a threshold) meant in the S.M.A.R.T. values, for example. That is, when any value of the S.M.A.R.T. values exceeds the threshold, it is determined that the S.M.A.R.T. value is an abnormal value. Hereinafter, a state where any of the S.M.A.R.T. values exceeds the THRESHOLD value may be called an S.M.A.R.T. abnormality).
As per claims 3 and 11: Guo teaches that wherein the determining the anomaly cause of the SSD comprises: identifying, by using a trained anomaly cause analysis model, the anomaly cause of the SSD based on the subset of the test data (see paragraph [0076], herein the S.M.A.R.T. abnormality may be detected using an original threshold based on empirical rules by statistically analyzing the S.M.A.R.T. values at the time of occurrence of a fault; and paragraph [0121).
As per claims 4 and 12: Guo teaches that before the determining whether the SSD has the anomaly, performing feature extraction on the test data to obtain features of the test data (see paragraph [0056], herein the controller 21 controls the writing and reading of data to and from the storage medium 22. When the CPU 11 performs data access to the storage medium 22 of the SSD 20, reading and writing are requested using an ATA command, for example. The controller 21 performs various data accesses to the storage medium 22 in conformity with a command transmitted from the CPU 11; and paragraphs [0057], [0070] - [0073] ).
As per claims 5 and 13: Gaber teaches that wherein for the S.M.A.R.T. data, the collecting test data of an SSD comprises: collecting a S.M.A.R.T. data set of the SSD, the S.M.A.R.T. data set including S.M.A.R.T. data (see column 3, lines 53-56, herein a raw sample SMART data collector module 110 collects samples of the SMART attributes reported by drives 102/106 and relays them to a first application of a feature selection machine learning model 114); determining a correlation between each S.M.A.R.T. data in the S.M.A.R.T. data set and whether the SSD has the anomaly; and taking a number of S.M.A.R.T. data with high correlation as the test data for determining whether the SSD has the anomaly (see column 2, lines 40-54, herein Another limitation of rule-based learning of disk drive failure patterns is the difficulty in taking into account the numerous correlations of multiple SMART attributes that can be part of the failure pattern. Some of the SMART attributes display high correlation in their measured values. For example, SMART feature 241, total Logical Block Addresses (LBA) written, is highly correlated with SMART feature 9, power on hours. Correlation between SMART attributes can affect the accuracy of rule-based learning if it is not taken into account by, for example, extracting the correlation level and incorporating it as a feature for the model. In the case of manually-set thresholds for a large number of attributes (e.g. if 80 SMART metrics are collected) most rule-based models can't handle that amount of attributes and their combinations effectively).
As per claim 17: Gaber teaches that a non-transitory computer-readable medium storing computer-executable instructions thereon, which when executed by at least one processor, cause an electronic apparatus to perform the method of claim 1 (see column 7, lines 56-65, herein Processor 801 may communicate with memory 803, which in one embodiment can be implemented via multiple memory devices to provide for a given amount of system memory. Memory 803 may include one or more volatile storage (or memory) devices such as random access memory (RAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), static RAM (SRAM), or other types of storage devices. Memory 803 may store information including sequences of instructions that are executed by processor 801, or any other device, and Fig. 8).
Allowable Subject Matter
7. Claim 6, 8, 14, and 16 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims. Dependent claims 7 and 15 depend from on claims 6 and 14 respectively, and inherently include limitations therein and therefore are allowed as well.
Examiner Notes
8. When amending the claims, applicants are respectfully requested to indicate the portion(s) of the specification which dictate(s) the structure relied on for proper interpretation and also to verify and ascertain the metes and bounds of the claimed invention.
Prior Art
9. The prior art of record, considered pertinent to the applicant’s disclosure, is listed in the attached PTO-892 form.
Conclusion
10. Any inquiry concerning this communication or earlier communications from the examiner should be directed to OSMAN ALSHACK whose telephone number is (571)272-2069. The examiner can normally be reached on MON-FRI 8:30 AM-5:00 PM EST, also please fax interview request to (571) 273- 2069. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, ALBERT DECADY can be reached on 5712723819. 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.
/OSMAN M ALSHACK/Examiner, Art Unit 2112
Prosecution Insights