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 .
Claim Rejections - 35 USC § 112
The following is a quotation of 35 U.S.C. 112(b):
(b) CONCLUSION.—The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the inventor or a joint inventor regards as the invention.
The following is a quotation of 35 U.S.C. 112 (pre-AIA ), second paragraph:
The specification shall conclude with one or more claims particularly pointing out and distinctly claiming the subject matter which the applicant regards as his invention.
Claims 15-20 rejected under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), second paragraph, as being indefinite for failing to particularly point out and distinctly claim the subject matter which the inventor or a joint inventor (or for applications subject to pre-AIA 35 U.S.C. 112, the applicant), regards as the invention.
Claim 1 recites the limitation "the at least one time subsequent to the current time" in line 23. There is insufficient antecedent basis for this limitation in the claim.
Claims 2-7 are rejected for their dependence on claim 15.
Claim 15 recites the limitation "the memory pool" in lines 4-5. There is insufficient antecedent basis for this limitation in the claim.
Claims 16-20 are rejected for their dependence on claim 15.
Allowable Subject Matter
Claims 8-14 allowed.
The following is an examiner’s statement of reasons for allowance:
The claimed invention is drawn to a computer system with two memory tiers that performs operations; determines an autoregressive integrated moving average of memory use of operations to predict future memory use; inputs the metrics of the forecast into a reinforcement model; uses the feedback of the reinforcement model to adjust the sizes of both a mirrored write segment and a non-mirrored read segment of the memory within the first tier of memory space, and to adjust weightings of forecast results of memory usage; and finally adjust the size of the second tier of memory based on the adjustments to the first tier of memory.
Although using an autoregressive integrated moving average (ARIMA) of workload time series with a machine learning model to predict storage system usage is known (as discussed below in citation of pertinent prior art), none of the prior art of record, as best understood by the Examiner, seems to fairly teach or suggest using the ARIMA results in a reinforcement learning model and producing the output of the reinforcement learning model that comprises first feedback to adjust respective sizes of a mirrored portion of the first tier of memory and a non-mirrored portion of the first tier of memory, and second feedback to adjust a weight of forecasting results associated with the second time series metrics of forecast future performance of the memory pool in a subsequent iteration;
halting, by the system, the performing of the respective iterations where the respective sizes of the mirrored portion and the non-mirrored portion satisfy a defined criterion; and
after the performing of the respective iterations, storing, by the system, specified computer data in the memory pool.
Claims 9-14 are allowed for their dependence on claim 8.
Claims 1-7 and 15-20 would be allowable if rewritten or amended to overcome the rejection(s) under 35 U.S.C. 112(b) or 35 U.S.C. 112 (pre-AIA ), 2nd paragraph, set forth in this Office action.
Similar to claim 8 above, although using an autoregressive integrated moving average (ARIMA) of workload time series with a machine learning model to predict storage system usage is known (as discussed below in citation of pertinent prior art), none of the prior art of record, as best understood by the Examiner, seems to fairly teach or suggest using the ARIMA results in a reinforcement learning model to obtain a first output of the reinforcement learning model that comprises first feedback to adjust a first size of a mirrored write segment portion of the first tier of memory and a second size of a non-mirrored read segment portion of the first tier of memory, wherein an overall size of the first tier of memory remains constant, and
a second output of the reinforcement learning model that comprises second feedback to adjust a weight of forecasting results associated with the second time series metrics of forecast future accesses of the memory pool in a subsequent time of the at least one time subsequent to the current time at which the group of operations is being performed, to then halt the performing of the group of operations at least one time where the first size and the second size satisfy a defined criterion; and
after the performing of the group of operations at least one time, adjust a third size of the second tier of memory based on the first size and the second size, and store computer data in the memory pool.
Any comments considered necessary by applicant must be submitted no later than the payment of the issue fee and, to avoid processing delays, should preferably accompany the issue fee. Such submissions should be clearly labeled “Comments on Statement of Reasons for Allowance.”
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Singh et al. (US 20200125412 A1), hereinafter “Singh”, discloses Dynamic Workload Management Based On Predictive Modeling And Recommendation Engine For Storage Systems. Singh teaches using methods such as ARIMA for workload prediction of operations (Singh [0027] ARIMA (AutoRegressive Integrated Moving Average), ARMA (AutoRegressive Moving Average), and MA (Moving Average) for time-series analysis to predict a future workload, e.g. in terms of TOPS or read+write commands. Such predictions may be used to simulate and determine an optimal hardware configuration for the storage array.). Singh also generally discloses machine learning techniques (Singh [0024]) and dynamic re-allocation of memory based on model results (Singh [0024] As indicated in step 304, the model may be used to dynamically (changing over time during normal operation, possibly in response to an input) re-allocate existing hardware resources to more reliably satisfy a predetermined level of measured performance. For example, and without limitation, the model could be used to re-allocate shared memory capacity, ports, storage engines, and processor cores for servicing IOs rather than host-hidden services at times when resource constraints might otherwise cause measured performance to fall below a target level. Because the model includes temporal dependencies, re-allocation may be performed based on satisfaction of predicted requirements to satisfy a predetermined level of measured performance.). However, Singh, when considered alone and in combination with the remaining prior art of record, does not seem to fairly teach or suggest providing the ARIMA produced second time series metrics as input to a reinforcement learning model, wherein a first output of the reinforcement learning model comprises first feedback to adjust a first size of a mirrored write segment portion of the first tier of memory and a second size of a non-mirrored read segment portion of the first tier of memory, wherein an overall size of the first tier of memory remains constant,
wherein a second output of the reinforcement learning model comprises second feedback to adjust a weight of forecasting results associated with the second time series metrics of forecast future accesses of the memory pool in a subsequent time of the at least one time subsequent to the current time at which the group of operations is being performed, and
determining to halt the performing of the group of operations at least one time where the first size and the second size satisfy a defined criterion; and
after the performing of the group of operations at least one time, adjusting a third size of the second tier of memory based on the first size and the second size, and storing computer data in the memory pool.
Ruan et al. (Workload time series prediction in storage systems: a deep learning based approach. Cluster Comput 26, 25–35. 13 January 2021. https://doi.org/10.1007/s10586-020-03214-y), hereinafter “Ruan”, discloses a practical storage workload prediction method called CrystalLP which includes workload collecting, data preprocessing, time series prediction based on a long short-term memory network(LSTM), and data postprocessing phase. Ruan discloses that an ARIMA model is a known statistical method of workload analysis, but opts to use a deep learning method, and makes no mention of changing the sizes of memory segments.
Bi et al. ("ARIMA-Based and Multiapplication Workload Prediction With Wavelet Decomposition and Savitzky–Golay Filter in Clouds," in IEEE Transactions on Systems, Man, and Cybernetics: Systems, vol. 54, no. 4, pp. 2495-2506, April 2024, doi: 10.1109/TSMC.2023.3343925.), hereinafter “Bi”, although not prior art eligible, discloses an approach that predicts future tasks in the next time interval by adopting a Savitzky–Golay filter, wavelet decomposition (SW), and ARIMA, thus leading to SWARIMA to predict future tasks in CDCs. Bi does not mention adjusting the size of different segments of memory based on the predictions.
Taghavi et al. (Compute Job Memory Recommender System Using Machine Learning. In Proceedings of the 22nd ACM SIGKDD International Conference on Knowledge Discovery and Data Mining (KDD '16). Association for Computing Machinery, New York, NY, USA, 609–616. 13 August 2016. https://doi.org/10.1145/2939672.2939717), hereinafter “Taghavi”, discloses several different models from classic statistical methods to more modern machine learning techniques to develop a memory recommender system based on the historical job execution data on a grid computing platform.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to CHRISTIAN T BRYANT whose telephone number is (571)272-4194. The examiner can normally be reached Monday-Thursday and Alternate Fridays 7:00-4:30.
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, CATHERINE RASTOVSKI can be reached at 571-270-0349. 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.
/CHRISTIAN T BRYANT/Examiner, Art Unit 2857
Prosecution Insights