DETAILED ACTION
Introduction
Claims 21-40 are pending. Claims 1-20 are cancelled. Claims 21 and 31 are amended. No new claims are added. This Office action is in response to Applicant’s request for reconsideration after non-final rejection filed on 3/5/2026.
Other Prior Art
Colgrove (US 2012/0066449) teaches splitting a read request into a reconstruction request (i.e., comprising a plurality of requests each directed to a different storage device) in response to predicting that a storage system is in a long latency response (i.e., congested) state. See par. 69.
Huang teaches generating a single transfer request corresponding to a requested chunk when a node storing the chunk is not congested, and generating multiple transfer requests corresponding to portions of the requested chunk to be sent to the other nodes storing the portions when the node storing the chunk is congested, thereby allowing the system to reconstruct the chunk from the portions without having to obtain the chunk directly from the congested node. See Huang et al., “Erasure Coding in Windows Azure Storage.”
Response to Arguments
Examiner discusses the arguments of Applicant’s representative below.
Rejection of claims 21 and 31 under 31 U.S.C. 103
Applicant’s representative has amended claims 21 and 31 to recite new features and now argues that the combination of Liang and Ellis/Hasegawa does not teach the system of claims 21 and 31, as amended. Examiner agrees. Nonetheless, the combination of Liang, Ellis/Hasegawa, Peters, and Povzner teach the system of amended claims 21 and 31, as discussed in the rejection below.
Claim Objections
Claims 21 and 31 recite a step to “transform the plurality of file system requests into DESS network requests for servicing by one or more other nodes of the DESS,” but Examiner suspects that it is the DESS requests that are transformed into DESS network requests, not the file system requests. The closest support for this step is provided in the abstract, which recites that “the DESS interface circuitry is operable to… generate a plurality of DESS file system requests for the first file system request; and transmit the plurality of DESS file system requests onto the one or more network links.” This support suggests that the claimed “DESS requests” are transformed into the claimed “DESS network requests” because it is the DESS requests that are transmitted over one or more network links, not the file system requests.
Claims 21 and 31 recite a step to “transmit the DESS requests via a network interface using congestion mitigation techniques,” but Examiner suspects that it is the DESS network requests that are transmitted via the network interface, not the DESS requests.” Currently, the DESS network requests serve no function in claims 21 and 31 after they are generated via the transformation step.
Claim Rejections: 35 U.S.C. 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.
Claims 21-29 and 31-39 are rejected under 35 U.S.C. 103 because they are unpatentable over Liang (US 2015/0309874) in view of either Ellis (US 2006/0031600) or Hasegawa (US 2013/0073795), Peters (US 6,760,808), and Povzner (US 9,665,294).1
Regarding claims 21 and 31, Liang teaches a system, the system comprising: a first device operable in a distributed electronic storage system (DESS), wherein the first device comprises: a receive buffer operable to receive and store a plurality of file system requests (A key-value store client 302 includes a request queue 300 for buffering file system requests from applications 301. See par. 46; fig. 3); a transmit buffer operable to store and transmit a plurality of DESS requests (The key-value store client includes a task queue 440 that stores and transmits a plurality of tasks to storage devices of a distributed key-value store 303. See par. 51); and a DESS processor operable to: dynamically split the plurality of file system requests into DESS requests according to predicted congestion states of the DESS (They key-value store obtains historical delay statistics for a request of a given type and object size, uses these category specific delay statistics to estimate the delay distribution that an incoming request of that category is likely to experience, and then uses that estimate to determine the number of tasks to generate for the incoming request. See par. 71. Similarly, the key-value receives backlog statistics, which it uses as an estimate of system utilization that a new request is likely to encounter when determining the number of tasks to generate for the new request. See par. 23), transform the DESS requests into DESS network requests for servicing by one or more other nodes of the DESS (The key-value store client sends the plurality of tasks in parallel over a plurality of network connections to the distributed key-value store, which requires packaging the tasks according to at least one network protocol for transmission over the network connections. See par. 118-119, 121), and transmit the DESS network requests via a network interface (As indicated above, the key-value store client sends the plurality of tasks in parallel over a plurality of network connections to the distributed key-value store, which requires a network interface. See par. 118-119, 121).
However, Lian does not teach that the DESS processor is operable to: generate a single DESS request corresponding to each of the plurality of file system requests, if a total size of all file system requests of the plurality of file system requests combined fails to exceed a threshold; and generate the plurality of DESS requests corresponding to a first file system request of the plurality of file system requests, if the total size of all file system requests of the plurality of file system requests combined exceeds the threshold. Nonetheless, Ellis teaches a storage system whereby the system generates a single transfer instruction for a newly received DMA request if the size of the newly received DMA request plus the size of the previously received DMA requests is less than the capacity of a transfer buffer, and whereby the system generates multiple transfer instructions for the newly received DMA request if the size of the newly received DMA request plus the size of the previously received DMA requests exceeds the capacity of the transfer buffer. See par. 7. Alternatively, Hasegawa teaches a storage system whereby the system generates a single data transfer operation for a newly received write operation if the size of the newly received write operation plus the size of previously received write operations is less than a threshold, and whereby the system generates multiple data transfer operations for the newly received write operation if the size of the newly received write operation plus the size of the previously received write operations exceeds the threshold. See par. 76-84; fig. 14.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Liang so that the system generates a single task for a newly received request if the size of the newly received request plus the size of other requests fails to exceed a threshold, and the system generates multiple tasks for the newly received request if the size of the newly received request plus the size of other requests exceeds the threshold, because doing so enables the system to optimize the number of tasks generated for each request based on the state of the system.
Lastly, Liang and Ellis/Hasegawa do not teach that the DESS processor is operable to schedule the plurality of DESS requests according to resource availability across nodes of the DESS. However, Peters teaches a system for reading a file that is distributed across a plurality of network-accessible storage units (i.e., storage servers. See col. 6, ln. 27-40) whereby the system receives a request to read the file, generates a segment request for each segment of the file, and schedules each segment request at one of a plurality of storage units storing the segment based on based on resource availability across the storage units. See col 4, ln. 18-29; col. 6, ln. 13-16; col. 18, ln. 16-31.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Liang and Ellis/Hasegawa so that the key-value store client schedules tasks at the distributed key-value store based on resource availability across storage servers of the distributed key-value store, because doing so ensures that tasks are load-balanced across storage servers of the distributed key-value store.
Lastly, Liang, Ellis/Hasegawa, and Peters do not teach that the DESS network requests are transmitted using a congestion mitigation technique. Nonetheless, Povzner teaches a system for performing a traffic mitigate technique whereby a rate controller receives a plurality of I/O requests, selects a token bucket from a plurality of token buckets based on a service class of each I/O request, sends the I/O requests to the storage system at a control rate, and then dynamically adjusts tat control rate based on observed throughout. See col. 1, ln. 13-28.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Liang, Ellis/Hasegawa, and Peters so that the tasks are transmitted using a congestion mitigation technique like the one described by Povzner because doing so allows the system to control the rate at which tasks are injected into the distributed key-value store, thereby ensuring that the distributed key-value store does not become overloaded.
Regarding claims 22 and 32, Liang, Ellis/Hasegawa, Peters, and Povzner teach the system of claim 21, wherein: the plurality of file system requests are for access to a storage resource on a second device of the DESS, and the plurality of DESS requests are transmitted to the second device via a network link (Liang teaches that the key-value store client receives read/write requests from applications, generates tasks corresponding to the read/write requests, and sends the tasks to storage devices of a distributed key-value store over a plurality of network links. See par. 68-73).
Regarding claims 23 and 33, Liang, Ellis/Hasegawa, Peters, and Povzner teach the system of claim 21, wherein: a state of the DESS is predicted according to resources required for servicing the first file system request and characteristics of a second file system request, and the quantity of the plurality of DESS requests corresponding to the first file system request is determined according to the predicted state of the DESS (Liang teaches that for each request, the system determines a number of tasks based on a predicted state of the system, which in turn is based on characteristics of the request (i.e., type and size of request. See par. 32), and backlog information which includes characteristics of other requests, such as the size and type of the other requests. See par. 80, 115).
Regarding claims 24 and 34, Liang, Ellis/Hasegawa, Peters, and Povzner teach the system of claim 23, wherein the resources required for servicing the first file system request are determined according to an amount of information to be read during servicing of the first file system request (Liang teaches that for each request, the system determines a number of tasks based on a predicted state of the system, which in turn is based on characteristics of the request, including whether the request is a read or write request, and the size of the read or write request. See par. 32).
Regarding claims 25 and 35, Liang, Ellis/Hasegawa, Peters, and Povzner teach the system of claim 23,wherein the resources required for servicing the first file system request are determined according to an amount of information to be written during servicing of the first file system request (Liang teaches that for each request, the system determines the number of tasks based on a predicted state of the system, which in turn is based on characteristics of the request, including whether the request is a read or write request, and the size of the read or write request. See par. 32).
Regarding claims 26 and 36, Liang, Ellis/Hasegawa, Peters, and Povzner teach the system of claim 21, wherein the DESS processor is operable to: determine a level of congestion of the DESS; and determine the quantity of the plurality of DESS requests corresponding to the first file system request according to the level of congestion of the DESS (Liang teaches that the key-value store client determines backlog information (i.e., utilization information) and uses the backlog information to determine the number of parallel read/write tasks to generate. See par. 70).
Regarding claims 27 and 37, Liang, Ellis/Hasegawa, Peters, and Povzner teach the system of claim 26, wherein the determination of the level of congestion of the DESS comprises a determination of a load on one or more resources of the DESS (Liang teaches that the backlog information represents a level of load or utilization of the resources of the key-value store. See par. 23).
Regarding claims 28 and 38, Liang, Ellis/Hasegawa, Peters, and Povzner teach the system of claim 27, wherein the one or more resources comprises one or more of: processor resources, memory resources, storage resources, and networking resources (Liang teaches that the utilized resources of the key-value store must include at least one of the resources enumerated in claim 27).
Regarding claims 29 and 39, Liang, Ellis/Hasegawa, Peters, and Povzner teach the system of claim 21, wherein the quantity of the plurality of DESS requests corresponding to the first file system request is determined according to whether a file system request, in the plurality of file system requests, is a data request or metadata request (Liang teaches that the metadata requests do not require writing/reading a data object to/from memory and are therefore relatively small in size (if not the smallest type of request). The system may take into the number of pending read or write requests of different sizes in the queue, and will therefore take into account the number of small metadata requests. See par. 80).
Claims 30 and 40 are rejected under 35 U.S.C. 103 because they are unpatentable over Liang, Ellis/Hasegawa, Peters, and Povzner, as applied to claims 21 and 31 above, in further view of Pillai (US 8,090,801).
Regarding claims 30 and 40, Liang, Ellis/Hasegawa, Peters, and Povzner do not teach the system of claim 21, wherein the DESS processor is operable to: determine a DESS usage metric, and determine the threshold according to the DESS usage metric. However, Pillai teaches a system for performing remote access commands between nodes, whereby the system adjusts available resources based on a usage metric, and whereby the usage metric comprises an average request size over a period of time. See col. 15, ln. 32-41.
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify the system of Liang, Ellis/Hasegawa, Peters, and Povzner so that the threshold is dynamically determined based on an average task size over a determined period of time, because doing so allows the threshold to change over time to account for changing system conditions.
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 extension fee 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 date of this final action.
Any inquiry concerning this communication or earlier communications from the examiner should be directed to Andrew Georgandellis whose telephone number is 571-270-3991. The examiner can normally be reached on Monday through Friday, 7:30-5:00 PM EST. If attempts to reach the examiner by telephone are unsuccessful, the examiner’s supervisor, Tonia Dollinger, can be reached on 571-272-4170. 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.
/ANDREW C GEORGANDELLIS/Primary Examiner, Art Unit 2459
1 Examiner collectively refers to Ellis and Hasegawa as “Ellis/Hasegawa.”
Prosecution Insights