FINAL REJECTION
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 § 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 and 11-13 are rejected under 35 U.S.C. 102(a)(1) as being anticipated by Yamamoto et al. (2021/0200639).
Regarding claim 1:
Yamamoto teaches:
A method comprising:
detecting an event requiring a rebuild of a persistent storage device of a storage array [par 53, 54 – failure of a physical segment of storage device that is of a storage array requiring rebuild of the data from the failed segment to be rebuilt is detected];
identifying each back-end slice associated with the persistent storage device [par 52, 53, 54, 57 – determining the physical segments that need to be rebuilt. These physical segments provide the physical storage for regions of a virtual volume and are thus back-end segments as the term back-end is commonly understood. A slice, as claimed, is interpreted as meaning a portion of the back-end storage as is equated with the segment or region discussed in the reference]; and
rebuilding the persistent storage device in an order corresponding to a priority rebuild score of each back-end slice associated with the persistent storage device [fig10; par 53, 137, 138 – the regions to rebuild are place in a queue based on a rebuild priority assigned to the regions. The priority is based on access frequency, which equates to the rebuild score of the claim.].
Regarding claim 2:
Yamamoto teaches:
The method of claim 1, further comprising: identifying each front-end logical track corresponding to each back-end slice associated with the persistent storage device [fig 1; par 52-54 – access is done via virtual volumes, which are front-end representations of the back-end storage. Virtual volumes are organized/managed with regard to regions as shown in figure 1 and described in paragraph 52, inter alia. The broadest reasonable interpretation of a front-end logical track is a virtual or logical representation of a portion of physical backing storage presented to a host, client, or the like to access the physical storage. The virtual volume management and regions within meet the BRI of the claim term].
Regarding claim 3:
The method of claim 2, further comprising: monitoring one or more input/output (IO) workloads received by the storage array [par 53, 56, 122-130 – collecting and updating access statistics for the regions/physical segments of the storage pool]; and
collecting IO statistics corresponding to each IO operation targeting each front-end logical track [par 53, 56, 122-130].
Regarding claims 11-13:
See the teachings outlined above with respect to claims 1-3.
Yamamoto further teaches a memory and a processor [par 363].
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.
Claims 4 and 14 are rejected under 35 U.S.C. 103 as being unpatentable over Yamamoto in view of Storer et al. (2012/0266011).
Regarding claims 4 and 14:
See the teachings of Yamamoto outlined above.
Yamamoto does not explicitly teach determining a service level objective (SLO) corresponding to each front-end logical track.
Storer teaches determining a service level objective (SLO) corresponding to each front-end logical track [par 6, 8, and 9 – determines reliability values of logical volumes based on reliability SLOs and uses them for prioritizing data recovery].
It would have been obvious to one of ordinary skill in the art prior to the effective filing date to combine the SLOs of Storer with the storage system rebuild prioritization of Yamamoto.
One of ordinary skill in the art prior to the effective filing date would have been motivated to make the combination because Storer teaches that prioritizing based on reliability SLOs results in significant performance and reliability improvements, minimizes windows of vulnerability to data-loss and also provides lower interference with foreground workload [par 9].
Allowable Subject Matter
Claims 5-10 and 15-20 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.
Response to Arguments
Applicant argues, on pages 6-8, that “Applicant respectfully submits that Yamamoto does not anticipate Claims 1 and 11 because Yamamoto fails to disclose the claim limitation "rebuilding the persistent storage device in an order corresponding to a priority rebuild score of each back-end slice associated with the persistent storage device," as recited in both claims.
The NFOA alleges that Yamamoto teaches this limitation because "the regions to rebuild are place in a queue based on a rebuild priority assigned to the regions" where "the priority is based on access frequency, which equates to the rebuild score of the claim." The NFOA's position rests on an overly broad and specification-inconsistent interpretation that improperly equates "access frequency" with "priority rebuild score." This interpretation fails under the proper broadest reasonable interpretation standard and ignores critical structural and functional distinctions in the claimed subject matter.
The claim term "priority rebuild score" must be interpreted in light of the Specification, which provides the ordinary and customary meaning to one of ordinary skill in the art. See Phillips v. A WHCorp., 415 F.3d 1303, 1313 (Fed. Cir. 2005) (en banc) (claims must be read in view of the specification); In re Translogic Tech., Inc., 504 F.3d 1249, 1257 (Fed. Cir. 2007) (broadest reasonable interpretation must be consistent with specification). The Specification unambiguously and consistently describes the priority rebuild score as calculated based on both IO statistics and Service Level Objectives. The Specification teaches at paragraph 6: "In embodiments, a priority score for each back-end slice can be calculated based on the IO statistics and the SLO of each front-end logical track corresponding to each back-end slice." (Applicant's Original Specification; 16). This dual-factor calculation is further detailed at paragraphs 59 and 60, which explain that the controller generates a priority score for each back-end slice using the IO statistics and SLO information corresponding to the back-end slice's associated TDEV track, provides weights to each back-end slice based on the priority of their corresponding IO statistics and SLO information, and computes a combined score based on the weighted SLO and IO statistics. (Applicant's Original Specification; 59-60).
In marked contrast, Yamamoto teaches a rebuild ordering system based exclusively on access frequency without any SLO component. Yamamoto discloses that the storage system monitors the frequency at which regions have been accessed and reorders entries in a queue in descending order of the access frequency. (Yamamoto; 53, 130, 137). Yamamoto's priority level is determined by access frequency, as explicitly stated at paragraph 138: "Priority level 1021 is information indicating priority of rebuilding. Note that priority level 1021 is determined by access frequency with the addition of a target region specified by user." (Yamamoto; 1138). Yamamoto calculates fixed values of access frequency using formulas that divide read/write counts by time periods. (Yamamoto; 125, 128). The rebuild order queue is then organized in descending order of these fixed values of access frequency. (Yamamoto; 279, Fig. 17).
Yamamoto contains no disclosure of Service Level Objectives, SLO categories, SLO tiers, or any concept of predetermined contractual performance requirements. Yamamoto does not teach assigning service levels such as Platinum, Diamond, Gold, Silver, or Bronze to data. Yamamoto does not teach weighting priority based on data importance categories. Yamamoto does not teach combining SLO factors with IO statistics to generate a unified priority score. The NFOA does not identify any paragraph, figure, or disclosure in Yamamoto that teaches or suggests these missing elements because no such disclosure exists.
The distinction between access frequency and priority rebuild score calculated from SLO plus IO statistics is not merely semantic but represents a fundamental technical difference. Access frequency measures only how often data has been accessed historically. It cannot distinguish between critical business data that happens to be accessed infrequently and non- critical temporary data that is accessed frequently. The claimed priority rebuild score solves this problem by incorporating Service Level Objective information representing the contractual importance and performance requirements for the data. (Applicant's Original Specification; 24, 27). As taught in the Specification, this dual-factor approach enables storage arrays to recover more effectively from disk errors, prioritizing reconstructing mission-critical data and minimizing the impact on system performance, offering an end-to-end solution for customers utilizing Quality of Service requirements. (Applicant's Original Specification; 27).
Under established precedent, anticipation requires that every element of the claimed subject matter be disclosed in a single prior art reference, either expressly or inherently. Verdegaal Bros. Inc. v. Union Oil Co. of Cal., 814 F.2d 628, 631 (Fed. Cir. 1987); see also MPEP § 2131. The Federal Circuit has held that "the identical invention must be shown in as complete detail as is contained in the claim." Richardson v. Suzuki Motor Co., 868 F.2d 1226, 1236 (Fed. Cir. 1989). Yamamoto is missing the SLO component of the priority rebuild score calculation entirely. This is not a minor omission but a fundamental structural element that defines how priority scores are calculated and what information they represent. The NFOA has not established, and cannot establish, that Yamamoto discloses this missing element.
The NFOA's attempt to equate Yamamoto's access frequency with Applicant's priority rebuild score contradicts the well-established principle that the Patent Office must give claims their broadest reasonable interpretation consistent with the specification. In re Morris, 127 F.3d 1048, 1054 (Fed. Cir. 1997); see also MPEP § 2111. An interpretation that reads out the SLO component from "priority rebuild score" is unreasonable and inconsistent with the Specification's clear and repeated teaching that the score is calculated from both SLO and IO statistics.
For these reasons, Yamamoto does not anticipate Claim 1. Claim 11 recites limitations corresponding to those in Claim 1 and is patentable for at least the same reasons as Claim 1.”
The examiner respectfully disagrees. Applicant’s argument rests, generally, on the allegation that the examiner has not afforded the term “priority rebuild score” a proper broadest reasonable interpretation in light of the specification. Specifically, applicant alleges that interpreting the access frequency of Yamamoto used to establish rebuild priority as equating to the “priority rebuild score” of the claim is unreasonable and inconsistent with the specification. Applicant points to various disclosures of the specification throughout the argument and alleges that such disclosures must necessarily be read into the broad language of the claim in order for the interpretation to be reasonable.
The examiner respectfully disagrees. The specification does not provide an explicit, special definition of the term “priority rebuild score” that would require such a narrow definition of the term. The specification discloses that the priority rebuild score is used to set the order of storage device rebuild (par 0002) and that the score “can be calculated based on the IO statistics and the SLO of each front-end logical track.” (par 0006). It is reasonable, therefore, in light of the specification to interpret the claim term broadly to encompass an scoring or ranking mechanism calculated based on at least one element of IO statistics and/or SLO, as the specification merely states that it “can be calculated” based on those broad groupings of metrics. Yamamoto calculates rebuild priority based on access frequency, which is an IO statistic, and thus meets the BRI of a score that can be calculated using IO statistics and SLO.
The examiner notes that features upon which applicant relies (i.e., dual-factor calculation, combined score based on weighted SLO and IO statistics, assigned service levels, etc.) are not recited in rejected claims 1 and 11. Although the claims are interpreted in light of the specification, limitations from the specification are not read into the claims. See In re Van Geuns, 988 F.2d 1181, 26 USPQ2d 1057 (Fed. Cir. 1993).
The examiner further notes that claim 5 explicitly recites calculating the priority score based on the IO statistics and the SLO of each front-end logical track. If the recitation of claim 1 required that such limitations be read into the broad language of claim 1, claim 5 would not be further limiting.
Applicant argues, on page 9, that “Claim 2 recites: "identifying each front-end logical track corresponding to each back- end slice associated with the persistent storage device."
The NFOA alleges that Yamamoto teaches this limitation because "access is done via virtual volumes, which are front-end representations of the back-end storage" and "virtual volumes are organized/managed with regard to regions." (Yamamoto; Fig. 1, 52-54). The NFOA interprets "front-end logical track" under the broadest reasonable interpretation as "a virtual or logical representation of a portion of physical backing storage presented to a host, client, or the like to access the physical storage."
This interpretation is overly broad and fails to account for the Specification's teaching of the specific front-end to back-end mapping architecture. The Specification teaches that front-end logical tracks are tracks of thin devices (TDEVs) that are mapped to back-end slices. (Applicant's Original Specification; 49, 55). TDEVs are logical storage units representing front- end tracks, and each TDEV track can correspond to a portion of a back-end slice. (Applicant's Original Specification; 55). The controller maintains a mapping table that maps front-end tracks to back-end tracks and uses the mapping table to map each IO operation's target TDEV track to its corresponding back-end slice. (Applicant's Original Specification; 58).
Yamamoto does not teach this specific mapping architecture. Yamamoto teaches virtual volumes with pages and regions where pages are mapped to logical chunks. (Yamamoto; 98, 99). Yamamoto's pages are units in which a storage program accesses a virtual volume. (Yamamoto; 99). This is fundamentally different from Applicant's teaching of front-end TDEV tracks that are specifically mapped to back-end RAID slices for the purpose of propagating SLO and IO statistics from the front-end to the back-end for priority score calculation. (Applicant's Original Specification; 55, 58, 59).”
The examiner respectfully disagrees. Applicant is again relying on features from the specification. i.e. the specific architecture cited from paragraphs 49 and 55 and the controller and mapping table of paragraph 58. These features are not recited in the claims at issue.
Although the claims are interpreted in light of the specification, limitations from the specification are not read into the claims. See In re Van Geuns, 988 F.2d 1181, 26 USPQ2d 1057 (Fed. Cir. 1993).
Claim 3 recites: "monitoring one or more input/output (IO) workloads received by the storage array; and collecting IO statistics corresponding to each IO operation targeting each front-end logical track."
The NFOA alleges that Yamamoto teaches "collecting and updating access statistics for the regions/physical segments of the storage pool." (Yamamoto; 53, 56, 122-130). However, Yamamoto monitors and collects statistics for pages in virtual volumes, not for front-end logical tracks as claimed. (Yamamoto; 122-130). Moreover, Yamamoto collects only access frequency data (read counts and write counts over time periods), not the comprehensive IO statistics taught in Applicant's Specification. (Yamamoto; 124-128).
Applicant's specification teaches that IO statistics include read and write IO activities for each front-end track, IO size, IO frequency, IO burst patterns, IO trends, IO rate, and the like. (Applicant's Original Specification; 57). The controller collects this information during the LSWD process corresponding to each IO operation and their corresponding TDEV tracks. (Applicant's Original Specification; 57). This comprehensive collection of diverse IO statistics for use in weighted priority score calculation is not taught by Yamamoto's simple read/write counters for access frequency calculation.
The examiner respectfully disagrees. The argument regarding the interpretation of the “front-end logical tracks” was addressed above with respect to claim 2. Applicant further argues that Yamamoto does not collect “the comprehensive IO statistics taught in Applicant’s Specification.” Applicant cites a listing of various statistics collected from specification paragraph 57. Again, these features are not recited in the claim and such features are not read into the broad term “IO statistics” recited in the claim.
Although the claims are interpreted in light of the specification, limitations from the specification are not read into the claims. See In re Van Geuns, 988 F.2d 1181, 26 USPQ2d 1057 (Fed. Cir. 1993).
Applicant argues, on pages 10-11, that “Applicant respectfully submits that the NFOA has not established a prima facie case of obviousness because it lacks articulated reasoning with rational underpinning to support the legal conclusion of obviousness as required by KSR Int'l Co. v. Teleflex Inc., 550 U.S. 398, 418 (2007). The NFOA provides only conclusory statements regarding the combination and fails to explain why one of ordinary skill in the art would modify Yamamoto's access-frequency-based system to incorporate Storer's volume-level reliability assignment system to achieve the claimed track-level SLO-based priority scoring.
The NFOA states that one of ordinary skill would have been motivated to make the combination "because Storer teaches that prioritizing based on reliability SLOs results in significant performance and reliability improvements, minimizes windows of vulnerability to data-loss and also provides lower interference with foreground workload." (Storer; 9). This statement is insufficient under controlling case law. The Federal Circuit has repeatedly held that conclusory statements about the benefits of a reference are inadequate to establish motivation to combine. In re Nuvasive, Inc., 842 F.3d 1376, 1381-82 (Fed. Cir. 2016) ("Our precedent has consistently required the PTO to provide a reason why a person of ordinary skill would have combined the prior art references... Conclusory statements regarding the general benefits of a reference are insufficient."). The NFOA must provide specific findings explaining why one of ordinary skill would modify Yamamoto by incorporating elements from Storer to arrive at the claimed subject matter. MPEP § 2143.
The Supreme Court in KSR emphasized that while common sense can support obviousness findings, the analysis still requires articulated reasoning and cannot rely on mere conclusory statements. KSR, 550 U.S. at 418. The Court explained that "rejections on obviousness grounds cannot be sustained by mere conclusory statements; instead, there must be some articulated reasoning with some rational underpinning to support the legal conclusion of obviousness." Id. The NFOA's statement that Storer teaches benefits does not explain why one would apply Storer's volume-level approach to Yamamoto's region-level system, how one would extract track-level SLO from Storer 's volume-level teachings, or why one would complicate Yamamoto's working system with Storer's allocation methodology.”
The examiner respectfully disagrees. In response to applicant’s argument that there is no teaching, suggestion, or motivation to combine the references, the examiner recognizes that obviousness may be established by combining or modifying the teachings of the prior art to produce the claimed invention where there is some teaching, suggestion, or motivation to do so found either in the references themselves or in the knowledge generally available to one of ordinary skill in the art. See In re Fine, 837 F.2d 1071, 5 USPQ2d 1596 (Fed. Cir. 1988), In re Jones, 958 F.2d 347, 21 USPQ2d 1941 (Fed. Cir. 1992), and KSR International Co. v. Teleflex, Inc., 550 U.S. 398, 82 USPQ2d 1385 (2007).
In this case, Yamamoto explicitly seeks to avoid performance decreases while rebuilding by utilizing priority rebuild ordering. Storer directly addresses this concept by utilizing SLO metrics of particular regions for rebuild and notes that doing so results in significant performance and reliability improvement, which are clearly advantages of using such an metric for rebuild prioritization. The expectation of some advantage is the strongest rationale for combining references. MPEP 2144(II).
Applicant argues, on pages 11-12, The references address fundamentally different technical problems and employ incompatible solutions, which weighs against obviousness. In re Kahn, 441 F.3d 977, 987-88 (Fed. Cir. 2006) (references addressing different problems less likely to be combined).
Yamamoto addresses the problem of how to dynamically reorder rebuild priorities during failure recovery when access patterns change unexpectedly. (Yamamoto; 7, 8). Yamamoto's stated objective is to "rebuild a storage device involved in failure properly" by adapting to changing IO patterns during the rebuild process. (Yamamoto; 8). Yamamoto achieves this through dual-period monitoring (long-period and short-period access frequency tracking) that allows real-time reordering of the rebuild queue. (Yamamoto; 9, 10, 53, 137).
Storer, in contrast, addresses the entirely different problem of how to initially allocate newly written data to appropriate parity groups based on the reliability importance of the data. (Storer; 4, 83). Storer's background explains that "the underlying physical layer of a storage system does not have a semantic understanding of the stored data" and does not distinguish important business information from scratch-space information. (Storer; 4). Storer solves this by assigning reliability values to volumes and selecting parity groups for data storage based on those values. (Storer; 6, 7, 83, 84). Storer's system operates during data allocation, not during rebuild operations.
The NFOA provides no explanation of why one of ordinary skill would look to Storer's data allocation system to improve Yamamoto's rebuild ordering system. The references operate at different stages of the storage lifecycle, address different problems, and employ architecturally distinct solutions. Storer assigns static reliability values to volumes for allocation decisions. Yamamoto monitors dynamic access frequencies for rebuild reordering decisions. These are not merely different implementations of the same concept but distinct approaches to distinct problems.
Moreover, the NFOA does not explain how one would combine these incompatible systems. Storer teaches volume-level reliability assignment. (Storer; 6, 89, 92). The NFOA asserts that this teaches "determining a service level objective (SLO) corresponding to each front-end logical track" as recited in Claims 4 and 14. This assertion requires significant modification of Storer's volume-level system to operate at the much more granular track level. The NFOA provides no articulated reasoning explaining why one would make this modification or how one would implement it. MPEP § 2143.01(VII) provides that the Examiner must explain the reasoning why the prior art references would be combined to produce the claimed subject matter.
The examiner respectfully disagrees. Yamamoto and Storer are both concerned with prioritizing rebuild of particular regions of storage based on metrics (IO statistics of Yamamoto, SLO of Storer) in order to avoid performance decrease due to rebuilding (Yamamoto). Storer is not directed solely to assignation of parity groups, instead explicitly stating that during recovery areas with high reliability values are prioritized, resulting in significant performance increases (Storer par 9). Storer thus directly addresses a desire of Yamamoto and is directed to the same goal of increasing recovery performance. The system of Yamamoto would not require significant modification in order to assign SLO values to regions of storage managed by Yamamoto. The Yamamoto reference would merely be modified by assigning particular SLO values to areas of storage utilized for storing data and, when rebuilding, referencing those values to increase rebuild performance as disclosed by Storer.
Applicant argues, on pages 12-13, that “The NFOA alleges that Storer teaches "determining a service level objective (SLO) corresponding to each front-end logical track" as recited in Claims 4 and 14 based on Storer's paragraphs 6, 8, and 9, which allegedly teach "determines reliability values of logical volumes based on reliability SLOs and uses them for prioritizing data recovery."
This allegation is factually incorrect. Storer does not determine SLOs corresponding to individual tracks. Storer assigns reliability values to logical containers of data such as volumes, files, and directories. (Storer; 6, 83). The system assigns a reliability value to each logical container of data located at an upper logical layer of the storage system. (Storer; 6, 27). Storer then identifies a particular parity group from the underlying physical storage layer for storing data corresponding to the logical container based on the container-level reliability value. (Storer; 7, 84). Storer's architecture operates at the container level, not at the individual track level within containers.
Storer states: "The system, in at least one embodiment, assigns a reliability value to each logical container of data (e.g., a volume) located at an upper logical layer of the storage system. In one embodiment, the reliability value is assigned according to objectives dictated by reliability SLOs." (Storer; 6). Paragraphs 8 and 9 describe using these volume-level reliability values to prioritize parity groups during recovery operations. (Storer; 8, 9). Neither paragraph teaches determining an SLO for each individual front-end logical track as required by Claims 4 and 14.
The claimed subject matter requires granular track-level SLO determination where each front-end logical track has a corresponding SLO. (Applicant's Original Specification; 57). The Specification teaches that the controller collects SLO information corresponding to each IO operation and their target TDEV tracks, where the SLO information can correspond to a service level such as Diamond, Silver, Bronze, and the like, representing different data priority levels. (Applicant's Original Specification; 57). This track-level granularity is not taught by Storer 's container-level approach.
Furthermore, even if Storer could somehow be construed to suggest track-level SLO (which it does not), Storer uses reliability values for data allocation decisions during write operations, not for rebuild priority calculations during failure recovery. (Storer; 83, 84, 97, 98). The NFOA does not explain why one would take Storer 's allocation-time reliability concept and apply it to rebuild-time priority scoring, particularly given that these occur at different times and serve different purposes.”
The examiner respectfully disagrees. The rejection is based on a combination of references. It has been previously established that Yamamoto teaches track level metric values and related priority management. Storer teaches assignment of SLO values to logical containers of data. Those values are used, when recovering, to prioritize one region over another for rebuild in order to manage recovery performance, and explicitly desire of Yamamoto. The containers of Storer are not defined explicitly as a volume as applicant appears to allege. In fact, a container, as disclosed by Storer, is a reference to any logical container of data and could be a volume, LUN, directory, file, etc. (Storer par 30). Thus, the tracks of Yamamoto, which are a logical representation of physical backing storage as established with regard to claim 2, would qualify as logical containers of Storer. Thus the SLO values of Storer which are assigned to logical containers can readily be assigned to the logical containers of Yamamoto. Those assigned values are then used when rebuilding to manage priority and increase performance.
Applicant again recites unclaimed features from the specification to argue against the Storer reference. Although the claims are interpreted in light of the specification, limitations from the specification are not read into the claims. See In re Van Geuns, 988 F.2d 1181, 26 USPQ2d 1057 (Fed. Cir. 1993).
Applicant argues, on pages 13-14, that “The proposed combination of Yamamoto and Storer would require significant redesign of both systems and appears to be based on impermissible hindsight knowledge of Applicant's claimed subject matter. The Federal Circuit has cautioned that "the mere fact that the prior art may be modified in the manner suggested by the Examiner does not make the modification obvious unless the prior art suggested the desirability of the modification." In re Fritch, 972 F.2d 1260, 1266 n.14 (Fed. Cir. 1992). Hindsight reconstruction using the applicant's disclosure as a roadmap is improper. Grain Processing Corp. v. Am. Maize-Prods. Co., 840 F.2d 902, 907 (Fed. Cir. 1988).
The proposed combination would require taking Yamamoto's access-frequency-based system and modifying it to add an SLO component, despite no teaching or suggestion in Yamamoto that SLO would improve its system; taking Storer's volume-level reliability assignment system and drastically modifying it to work at the track level, despite Storer providing no teaching or suggestion for track-level implementation; combining Yamamoto's IO frequency calculations with Storer 's reliability values through a weighted scoring algorithm, despite neither reference teaching or suggesting such an algorithm; and implementing the combination during LSWD processes in real-time, despite neither reference teaching or suggesting this implementation.
The NFOA provides no teaching, suggestion, or motivation in the prior art for any of these modifications. The combination is only obvious in hindsight after reviewing Applicant's disclosure. See MPEP § 2145 ("[R]ejections on obviousness grounds cannot be sustained with mere conclusory statements; instead, there must be some articulated reasoning with some rational underpinning to support the legal conclusion of obviousness.").”
The examiner respectfully disagrees.
Regarding applicant’s statement that there is no teaching or suggestion in Yamamoto that SLO would improve its system, the suggestion of improvement is from the Storer reference, as covered above.
Regarding applicant’s argument regarding Storer’s “volume-level reliability,” the examiner previous explained that Storer discusses reliability values of logical containers and is not so limited as to define a logical container as only a “volume.” As noted above, Yamamoto teaches track-level metrics wherein the track is a logical construct similar to the disclosed logical containers of Storer. The rejection is based on a combination of references and nonobviousness cannot by shown by attacking references singly when the rejection is based on a combination.
Regarding applicant’s argument regarding a weight scoring algorithm and implementing the combination during LWSD process in real-time, there is absolutely not recitation in the claim regarding such features and whether or not the references teach those features is irrelevant. The examiner suggests amending the claims to recites such features if applicant intends to rely on them.
Applicant argues, on pages 14-15, that “Storer teaches away from implementing SLO at the track level. A reference teaches away when a person of ordinary skill, upon reading the reference, would be discouraged from following the path set out in the reference, or would be led in a direction divergent from the path that was taken by the applicant. In re Gurley, 27 F.3d 551, 553 (Fed. Cir. 1994).
Storer's entire architectural approach is premised on assigning reliability at the logical container level (volumes, files, directories) and propagating those values down to the physical layer. (Storer; 6, 27, 83, 94). Storer teaches that reliability values are first assigned to volumes, then translated to regions, then to slabs, and finally to parity groups. (Storer; 94). This top-down container-based approach is fundamentally inconsistent with the claimed bottom-up track-level approach where SLO is determined for each individual front-end logical track.
Moreover, Storer provides explicit reasoning for its container-level approach: it enables the system to make efficient data operations by attaching reliability information to parity groups and physical storage units. (Storer; 7, 85). If Storer intended track-level implementation, it would have taught doing so. Instead, Storer consistently operates at the container level throughout its disclosure. A person of ordinary skill reviewing Storer would be led away from track-level SLO implementation and toward container-level reliability assignment.”
The examiner respectfully disagrees. Applicant again, as above, that the container level reliability is not applicable to the tracks of Yamamoto. Again, Storer teaches that reliability values are assigned to a logical container, which can be a volume, directory, file, etc. This does not teach away from the logical container being the regions of Yamamoto, which are logical representations of physical storage. Storer does teach, in one embodiment, that the SLO can be propagated down. Storer also teaches that SLO can simply be assigned to containers of simply to logical data stored in volumes (par 89). Storer’s teaching of propagation in one embodiment in paragraph 94 thus does not teach away from the combination.
Applicant argues, on page 15, that “Even if there were motivation to combine (which there is not), the NFOA provides no analysis of whether one of ordinary skill would have had a reasonable expectation of success in making the combination. See PAR Pharm., Inc. v. TWIPharms., Inc., 773 F.3d 1186, 1193 (Fed. Cir. 2014) (reasonable expectation of success is required for obviousness).
The proposed combination would require modifying Storer's volume-level SLO system to work at the track level, potentially requiring tracking millions of individual SLO assignments; integrating Storer's static reliability values with Yamamoto's dynamic access frequency monitoring; creating a weighted scoring algorithm to combine SLO and IO statistics (not taught by either reference); and ensuring the combination works during real-time LSWD operations without degrading system performance.
Neither reference provides any teaching or suggestion that such a combination would work or would be technically feasible. The NFOA provides no analysis of whether success would be expected. The burden is on the Examiner to establish a prima facie case, which includes demonstrating that one of ordinary skill would have reasonably expected success. In re Rinehart, 531 F.2d 1048, 1052 (CCPA 1976). The NFOA has not met this burden.”
The examiner respectfully disagrees. The combination would not require modifying Storer’s volume level SLO system to work at track level. Rather, it would require modifying Yamamoto’s track-level system to include an assigned SLO value for the Yamamoto’s regions and referencing those values when organizing a rebuild priority queue. Yamamoto stores data in and access data from logical regions. When rebuilding logical regions, Yamamoto accesses a metric that represents access frequency of those regions. To access an additional piece of data associated with those regions, i.e. an assigned SLO value, would be trivial and would not be technically infeasible. The combination would not require creating a weighted scoring algorithm or ensuring the combination works during real-time LSWD operations as applicant alleges, because those features are not recited in the claims.
Regarding applicant’s arguments concerning claims 5 and 15, the examiner has fully considered the arguments and finds them persuasive. The 35 USC 103 rejection of claims 5 and 15 is withdrawn.
Regarding applicant’s arguments concerning claims 6-10 and 16-20, the arguments are moot. Claims 6 and 16 were not rejected under art and after the amendment to claims 4 and 14 to obviate the 35 USC 112 rejections, the claims are no longer rejected. After the amendment to claims 7 and 17 to now depend from claims 6 and 16, respectively, claims 7-10 and 17-20 are now considered to recite allowable matter at least by virtue of their dependencies.
Conclusion
THIS ACTION IS MADE FINAL. 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 MARC M DUNCAN whose telephone number is (571)272-3646. The examiner can normally be reached M-F: 730am-9am, 10am-4:30pm.
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, Bryce Bonzo can be reached at 571-272-3655. 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.
/MARC DUNCAN/ Primary Examiner, Art Unit 2113
Prosecution Insights