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 .
Claims 1-20 are pending.
This is in response to communications filed on 5/14/26.
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.
Claim(s) 1-4, 9-13, 18-20 is/are rejected under 35 U.S.C. 103 as being unpatentable over Stabler (US Patent 20240045698; cited in IDS), in view of Sutera (US Patent 7721011).
For claim 1, Stabler et al teach the following limitations: An apparatus, comprising: a memory ([0098][0099]); and a processing device (Fig 8; processor 816; [0098]-[0099]), operatively coupled to the memory, configured to: receiving, from a storage system (Fig 1 is the storage system; Fig 8 processor 816 and system 814; API mentioned in [0040[]-[0041] obtains information; [0033][0039][0099][0089][0108]), one or more indications that a plurality of input/output (I/O) operations have been performed by the storage system (information obtained by API indicates IOPS performed by the storage system; [0041]-[0042] determined metrics is IOPS/watt and metrics is determined from information; [0033][0054][0061][0073][0083]; thus number of input/output operations information is received by management agent 148 and other components [0039]; [0005] – collected information from storage devices); determining an amount of energy consumed by the storage system to perform the plurality of I/O operations (calculate power consumption – [0039]; [0049][0054][0055][0062]-[0069]; power (i.e., energy in unit time) is calculated using IOPS; thus the calculation determines the energy consumed to perform the IO operations); and generating an energy usage effectiveness of the storage system ([0042]-[0046][0055][0064][0065][0066][0070] - least efficient workload/aggregate is identified; GUI provides recommendations/suggestion to increase efficiency by moving/relocating workload) based on a number of the plurality of I/O operations performed by the storage system ([0042][0054] [0061]-[0064] mention about IOPS) and the amount of energy consumed by the storage system to perform the plurality of I/O operations ([0054]-[0055] [0062]-[0064] power consumed for the IOPS to calculate IOPS/watt; Fig 2 and [0058][0060][0061] mentions how to compute power); automatically modify scheduling ([0067][0075][0095][0105][0114] mention that workload schedule can be changed; schedule can be changed for a time when node is less busy [0067]) of one or more subsequent I/O operations (workloads are associated with I/O operations as mentioned in col 5, [0064] –[0066]; [0066]-[0067] mention workload with high IOPS can be moved to improve workload IOPS or workload can be scheduled for a different time; thus the scheduling of the subsequent I/O operations can be modified) based on the energy usage effectiveness ([0064]-[0071] mention that the inefficient workloads are identified and efficiency is increased by relocating and scheduling another time).
Stabler does not explicitly mention the reordering the I/O operation within a queue of pending I/O operations. Thus, Stabler does not explicitly teach the limitations automatically modify scheduling of one or more subsequent I/O operations by reordering or allocating the one or more subsequent I/O operations within a queue of pending I/O operations of the storage
Sutera teach the following limitations: automatically modify scheduling of one or more subsequent I/O operations by reordering or allocating the one or more subsequent I/O operations (lines 5-13 of col 6 mention that memory access commands are received by the queue 250 and 250 improves the scheduling by reordering the commands; lines 52-55 of col 10 – 250 stores/reorders read and write commands; thus the pending I/O operations are received and reordered to modify the scheduling) within a queue of pending I/O operations of the storage (Fig 2 shows the reordering command queue 250; lines 27-63 of col 7 mentions that reordering logic reorders based on whether the commands are in desired order) based on the energy usage effectiveness (reordering is performed by improving the scheduling of memory accesses with a pattern that is optimized for power – lines 35-45 of col 4; lines 27-63 of col 7 mentions that reordering is based on whether the commands are in desired order; lines 35-45 of col 10 mention improved power consumption; thus the desired order provides the basis of the energy usage effectiveness as it optimizes the power).
It would have been obvious for one ordinary skill in the art before the effective filing date of the invention to combine the teachings of Stabler and Sutera to use a queue to reorder the I/O operations to modify the scheduling, since the queue reordering provides benefits of optimization, which further provides improved scheduling and power savings. Stabler uses workload scheduling at different time by moving the workload to different schedule/location ([0065]-[0067]), which can employ the teachings of Sutera to use a reorder queue to provide improve the scheduling of memory accesses. That way, the workload efficiency can be improved to further enhance the system’s performance.
For claims 2, 11, 19, Stabler teaches the following limitations: wherein the processing device is further configured to: receive, from the storage system, a throughput of the storage system when performing the plurality of I/O operations ([0043][0055] IOPS/watt metric can provide insight into power consumption in terms of throughput; metrics associated with busiest workloads based on throughput; thus the throughput is received from the storage system), wherein generating the energy usage effectiveness of the storage system is further based on the throughput of the storage system when performing the plurality of I/O requests ([0043]-[0045][0055][0064]-[0067][0070]-[0071] the metric is in terms of the throughput and the metric is used to compare the workloads/aggregates to determine least/most efficient).
For claims 3, 12, 20, Stabler teaches wherein the apparatus/processing device is a storage system controller of the storage system (Fig 1 and Fig 8; [0035][0039][0099] – processor and management agent 148 is the storage system controller of the storage system shown in Fig 2).
For claim 4, Stabler teach the following limitations: wherein the processing device is further configured to: modify scheduling of a subsequent plurality of (I/O) operations to be performed by the storage system to increase the energy usage effectiveness of the storage system ([0067][0075] – changing schedules for operations [0095][0105][0114]), wherein the modifying the scheduling comprising delaying the I/O operations ([0067]). Sutera further teaches modifying the scheduling comprising delaying the I/O operations (lines 24-45 of col 11).
For claim 9, Stabler teaches the following limitations: wherein the processing device is further configured to: generate a graphical user interface (GUI) for providing the energy usage effectiveness of the storage system ([0101] graphical user interface with various dashboards and recommendation).
For claim 10, Stabler et al teach the following limitations: A method, comprising: receiving, by a processing device from a storage system (Fig 1 is the storage system; Fig 8 processor 816 and system 814; API mentioned in [0040[]-[0041] obtains information; [0033][0039][0099][0089][0108]), one or more indications that a plurality of input/output (I/O) operations have been performed by the storage system (information obtained by API indicates IOPS performed by the storage system; [0041]-[0042] determined metrics is IOPS/watt and metrics is determined from information; [0033][0054][0061][0073][0083]; thus number of input/output operations information is received by management agent 148 and other components [0039]; [0005] – collected information from storage devices); determining an amount of energy consumed by the storage system to perform the plurality of I/O operations (calculate power consumption – [0039]; [0049][0054][0055][0062]-[0069]; power (i.e., energy in unit time) is calculated using IOPS; thus the calculation determines the energy consumed to perform the IO operations); and generating an energy usage effectiveness of the storage system ([0042]-[0046][0055][0064][0065][0066][0070] - least efficient workload/aggregate is identified; GUI provides recommendations/suggestion to increase efficiency by moving/relocating workload) based on a number of the plurality of I/O operations performed by the storage system ([0042][0054] [0061]-[0064] mention about IOPS) and the amount of energy consumed by the storage system to perform the plurality of I/O operations ([0054]-[0055] [0062]-[0064] power consumed for the IOPS to calculate IOPS/watt; Fig 2 and [0058][0060][0061] mentions how to compute power); automatically modify scheduling ([0067][0075][0095][0105][0114] mention that workload schedule can be changed; schedule can be changed for a time when node is less busy [0067]) of one or more subsequent I/O operations (workloads are associated with I/O operations as mentioned in col 5, [0064] –[0066]; [0066]-[0067] mention workload with high IOPS can be moved to improve workload IOPS or workload can be scheduled for a different time; thus the scheduling of the subsequent I/O operations can be modified) based on the energy usage effectiveness ([0064]-[0071] mention that the inefficient workloads are identified and efficiency is increased by relocating and scheduling another time).
Stabler does not explicitly mention the reordering the I/O operation within a queue of pending I/O operations. Thus, Stabler does not explicitly teach the limitations automatically modify scheduling of one or more subsequent I/O operations by reordering or allocating the one or more subsequent I/O operations within a queue of pending I/O operations of the storage
Sutera teach the following limitations: automatically modify scheduling of one or more subsequent I/O operations by reordering or allocating the one or more subsequent I/O operations (lines 5-13 of col 6 mention that memory access commands are received by the queue 250 and 250 improves the scheduling by reordering the commands; lines 52-55 of col 10 – 250 stores/reorders read and write commands; thus the pending I/O operations are received and reordered to modify the scheduling) within a queue of pending I/O operations of the storage (Fig 2 shows the reordering command queue 250; lines 27-63 of col 7 mentions that reordering logic reorders based on whether the commands are in desired order) based on the energy usage effectiveness (reordering is performed by improving the scheduling of memory accesses with a pattern that is optimized for power – lines 35-45 of col 4; lines 27-63 of col 7 mentions that reordering is based on whether the commands are in desired order; lines 35-45 of col 10 mention improved power consumption; thus the desired order provides the basis of the energy usage effectiveness as it optimizes the power).
It would have been obvious for one ordinary skill in the art before the effective filing date of the invention to combine the teachings of Stabler and Sutera to use a queue to reorder the I/O operations to modify the scheduling, since the queue reordering provides benefits of optimization, which further provides improved scheduling and power savings. Stabler uses workload scheduling at different time by moving ([0065]-[0067]), which can employ the teachings of Sutera to use a reorder queue to provide improve the scheduling of memory accesses. That way, the workload efficiency can be improved.
For claim 13, Stabler teach the following limitations: wherein the method comprises: modify scheduling of a subsequent plurality of (I/O) operations to be performed by the storage system to increase the energy usage effectiveness of the storage system ([0067][0075] – changing schedules for operations [0095][0105][0114]), wherein the modifying the scheduling comprising delaying the I/O operations ([0067]).
For claim 18, Stabler et al teach the following limitations: A non-transitory computer readable storage medium storing instructions ([0012] [0147]) which, when executed, cause a processing device (Fig 8; processor 816; [0098]-[0099]), to: receiving, from a storage system (Fig 1 is the storage system; Fig 8 processor 816 and system 814; API mentioned in [0040[]-[0041] obtains information; [0033][0039][0099][0089][0108]), one or more indications that a plurality of input/output (I/O) operations have been performed by the storage system (information obtained by API indicates IOPS performed by the storage system; [0041]-[0042] determined metrics is IOPS/watt and metrics is determined from information; [0033][0054][0061][0073][0083]; thus number of input/output operations information is received by management agent 148 and other components [0039]; [0005] – collected information from storage devices); determining an amount of energy consumed by the storage system to perform the plurality of I/O operations (calculate power consumption – [0039]; [0049][0054][0055][0062]-[0069]; power (i.e., energy in unit time) is calculated using IOPS; thus the calculation determines the energy consumed to perform the IO operations); and generating an energy usage effectiveness of the storage system ([0042]-[0046][0055][0064][0065][0066][0070] - least efficient workload/aggregate is identified; GUI provides recommendations/suggestion to increase efficiency by moving/relocating workload) based on a number of the plurality of I/O operations performed by the storage system ([0042][0054] [0061]-[0064] mention about IOPS) and the amount of energy consumed by the storage system to perform the plurality of I/O operations ([0054]-[0055] [0062]-[0064] power consumed for the IOPS to calculate IOPS/watt; Fig 2 and [0058][0060][0061] mentions how to compute power); automatically modify scheduling ([0067][0075][0095][0105][0114] mention that workload schedule can be changed; schedule can be changed for a time when node is less busy [0067]) of one or more subsequent I/O operations (workloads are associated with I/O operations as mentioned in col 5, [0064] –[0066]; [0066]-[0067] mention workload with high IOPS can be moved to improve workload IOPS or workload can be scheduled for a different time; thus the scheduling of the subsequent I/O operations can be modified) based on the energy usage effectiveness ([0064]-[0071] mention that the inefficient workloads are identified and efficiency is increased by relocating and scheduling another time).
Stabler does not explicitly mention the reordering the I/O operation within a queue of pending I/O operations. Thus, Stabler does not explicitly teach the limitations automatically modify scheduling of one or more subsequent I/O operations by reordering or allocating the one or more subsequent I/O operations within a queue of pending I/O operations of the storage
Sutera teach the following limitations: automatically modify scheduling of one or more subsequent I/O operations by reordering or allocating the one or more subsequent I/O operations (lines 5-13 of col 6 mention that memory access commands are received by the queue 250 and 250 improves the scheduling by reordering the commands; lines 52-55 of col 10 – 250 stores/reorders read and write commands; thus the pending I/O operations are received and reordered to modify the scheduling) within a queue of pending I/O operations of the storage (Fig 2 shows the reordering command queue 250; lines 27-63 of col 7 mentions that reordering logic reorders based on whether the commands are in desired order) based on the energy usage effectiveness (reordering is performed by improving the scheduling of memory accesses with a pattern that is optimized for power – lines 35-45 of col 4; lines 27-63 of col 7 mentions that reordering is based on whether the commands are in desired order; lines 35-45 of col 10 mention improved power consumption; thus the desired order provides the basis of the energy usage effectiveness as it optimizes the power).
It would have been obvious for one ordinary skill in the art before the effective filing date of the invention to combine the teachings of Stabler and Sutera to use a queue to reorder the I/O operations to modify the scheduling, since the queue reordering provides benefits of optimization, which further provides improved scheduling and power savings. Stabler uses workload scheduling at different time by moving ([0065]-[0067]), which can employ the teachings of Sutera to use a reorder queue to provide improve the scheduling of memory accesses. That way, the workload efficiency can be improved to further improve the power efficiency in the system.
Claim(s) 5-7, 14-16, is/are rejected under 35 U.S.C. 103 as being unpatentable over Stabler (US Patent 20240045698; cited in IDS), in view of Sutera (US Patent 7721011) and further in view of Honnavara-Prasad (US Patent Application Publication 2020/0183476).
For claim 5 and claim 14, Stabler et al teach executing the machine learning algorithms ([0045] [0091]) and generate modifications to scheduling ([0091] recommendations, [0075] recommendations include schedule changes). Stabler, in view of Sutera does not explicitly mention using the AI model. Honnavara-Prasad teaches the AI model to modify scheduling of IO operations ([0025] Fig 1Fig 5). It would have been obvious for one ordinary skill in the art before the effective filing date of the invention to include AI model to determine the scheduling modifications, since AI can optimize power and performance in the system ([0003] Honnavara-Prasad), which further increases the performance power scenario of the system.
For claim 6 and claim 15, Stabler teaches wherein the storage system is part of a data center comprising a plurality of storage systems (Fig 12). Stabler further teaches selective scheduling to reduce power consumption based on respective energy usage effectiveness values of the plurality of storage systems ([0066][0067] – nodes have different IOPS/watt workloads can be moved to increase efficiency … to reduce power consumption). Sutera further teaches modifying scheduling to control power consumption (lines 10-45 of col 10). Stabler, in view of Sutera does not mention receiving a maximum power available to the plurality of storage systems of the data center; and modify scheduling of I/O operations to be performed by the plurality of storage systems to cause an amount of power consumed by the plurality of storage systems to not exceed the maximum amount of power. Honnavara-Prasad teaches receiving a maximum power available to the plurality of storage systems ([0035] power is maintained below a threshold level; the threshold level power is the maximum available power for the system); and modify scheduling of I/O operations to be performed by the plurality of storage systems to cause an amount of power consumed by the plurality of storage systems to not exceed the maximum amount of power ([0035] selective scheduling to throttle power below threshold). It would have been obvious for one ordinary skill in the art before the effective filing date of the invention to receive the maximum power and modifying scheduling so as not to cause the exceeding of the maximum power (as taught in Honnavara-Prasad) in the system of Stabler in view of Sutera. That way power consumption can be optimized to not exceed the maximum value, which ensures the safety of the system.
For claim 7 and claim 16, Stabler, in view of Sutera, in view of Honnavara-Prasad teaches slow processing of one or more of the I/O operations to reduce the amount of power consumed by the plurality of storage systems (putting nodes to sleep [0075] Stabler; [0035] Honnavara-Prasad – NOOP operation to reduce power consumption).
Claims 8 and 17, is/are rejected under 35 U.S.C. 103 as being unpatentable over Stabler (US Patent 20240045698; cited in IDS), in view of Sutera (US Patent 7721011) and further in view of Honnavara-Prasad (US Patent Application Publication 2020/0183476), further in view of Cheng (US Patent Application Publication 20210026688).
For claims 8 and claim 17, Stabler, in view of Sutera, further in view of Honnavara-Prasad teaches wherein the processing device is further configured to: modify scheduling of a plurality operations across the plurality of storage systems to cause power consumption associated with the plurality operations and the subsequent plurality of I/O operations to not exceed the maximum power available (putting nodes to sleep [0075] [0066]-0067] – migration to other nodes [0122]-[0124] Stabler; [0035] Honnavara-Prasad – NOOP operation to reduce power consumption below threshold). Stabler, Sutera and Honnavara-Prasad do not explicitly mention background operations. These background operations are well known in the art (Cheng [0073] modifying scheduling for background operation). It would have been obvious for one ordinary skill in the art before the effective filing date of the invention to have the background operations and modify scheduling for the background operations to exceed the power available, since background operations can be switched to lower performance system and the system can be optimized while staying within power boundaries.
Response to Arguments
Applicant’s arguments have been considered but are moot because of the new ground of rejection. However, rejection is still relied upon Stabler, Honnavara-Prasad and Cheng. Thus, Examiner is addressing the arguments regarding these references.
Applicants argue that Stabler’s recommendations require user action and do not constitute automatic control of I/O execution as required by the claim (page 8).
Examiner disagrees. Claim does not mention “user action” and there is no requirement that automatic control must exclude user action. Instead, applicant’s specification Fig 7 mention “user interface” in step 708. Stabler’s system can perform the relocation of workload including modifying the scheduling ([0066][0067]).
Applicant further argues that cited art does not disclose feedback control mechanism in which an energy usage effectiveness metric derived from I/O operations is used to automatically reorder pending I/O operations within a queue of a storage system (Page 9).
Examiner disagrees. Claim does not require any such feedback control mechanism. Thus. The argument is not relevant.
Applicant further argues that cited art does not teach why an ordinary skill would have been motivated to implement Stabler’s workload level recommendations to queue level automatic control of I/O scheduling based on such a metric.
Examiner disagrees. Stabler teaches workload’s scheduling modification based on the metric ([0066]-[0067]) as explained above in detail. Stabler does not explicitly mention queue implementation and the newly cited art Sutera teaches queue to improve the scheduling. Therefore, the workload’s IO operations in Stabler can be reordered to improve scheduling by including queue of Sutera. Sutera mentions that such improving benefits power usage (lines 34-45 of col 4; Sutera).
Applicant further argues that Stabler does not teach throughput based effectiveness as a control input for modifying scheduling (Page 10).
Examiner disagrees. According to [0043] Stabler, IOPS/Watt metric is the indication of throughput, [0055] mentions metrics are provided based on throughput. The other section [0044] [0064] mention how IOPS/Watt is used to modify scheduling of workload (ie. I/O operations).
Applicant further argues that cited art does not teach delaying operations (Page 10). Examiner disagrees as Stabler teaches delaying workload ([0067], which delays the I/O operations. The newly cited art Sutera teaches delaying the I/O operations (lines 24-45 of col 11).
Applicant further argues that cited art does not teach claimed coordination across multiple storage systems using energy usage effectiveness values to maintain a maximum power value (page 11).
Examiner disagrees. Honnavara Prasad’s AI processor circuit includes the plurality of storage systems ([0034] – memory and other circuits using multiple cells/blocks) and is configured to receive the maximum power and not to exceed the maximum power ([0035] – power below threshold) by using selective scheduling ([0035]). Setura further teaches power conservation using I/O reordering (lines 5-20 of col 6). Therefore, with the teachings of Setura and Honnavara Prasad, the plural storage system’s power can be managed not to exceed the maximum power by using selective and modified scheduling.
Applicant further argues that Cheng does not disclose coordinating background operation with foreground operation based on a shared energy usage effectiveness metric or modifying both types of operations to collectively satisfy a power constraint (Page 11).
Examiner disagrees. Claim does not require coordinating background operation with foreground operation and therefore, the argument is not relevant.
Applicant further argues that cited art does not teach the combination of AI with specific queue level control mechanism as the Honnava-prasad does not teach AI model to control scheduling of IO within a storage system queue.
Examiner disagrees. Hannavan-Prasad teaches AI to control scheduling of IO (Fig 1) and Sutera teaches reorder queue to modify the scheduling (Fig 2). Stabler teaches workload scheduling modification based on energy usage effectiveness ([0066]-0067]). Thus, the combination of Stabler, Sutera and Hannava-Prasad provides sufficient teaching for the claim language as explained above.
Conclusion
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 FAHMIDA RAHMAN whose telephone number is (571)272-8159. The examiner can normally be reached Monday - Friday 10 AM - 7 PM. 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, Andrew Jung can be reached at 571-270-3779. 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.
/FAHMIDA RAHMAN/Primary Examiner, Art Unit 2175
Prosecution Insights