DETAILED ACTION
This is in response to the amendment filed on June 15th 2026.
Response to Arguments
Applicant’s statement, pg. 8-12, regarding the interview is noted.
Applicant’s arguments, see pg. 12, filed 6/15/26, with respect to the claim objection have been fully considered and are persuasive. In view of the amendment, the objection of claim 1 has been withdrawn.
Applicant's arguments, pg. 13, regarding the drawing objection have been fully considered but they are not persuasive. Applicant states Fig. 1 clearly shows the “plurality of endpoints 162” and the “plurality of endpoints 164”. There is no dispute about this. As applicant correctly points out, Fig. 1 uses two different numbers “162” and “164” to refer to the same thing “plurality of endpoints”. Thus, the drawing objection is maintained.
Applicant’s arguments, see pg. 13-15, with respect to the rejection(s) of claim(s) 1-23 under 102 and 103 have been fully considered and are persuasive. Therefore, the rejection has been withdrawn. However, upon further consideration, a new ground(s) of rejection is made in view of Cao et al. US 2021/0357256 A1.
Drawings
The drawings are objected to as failing to comply with 37 CFR 1.84(p)(4) because reference characters "162" and "164" have both been used to designate “Endpoint” (Fig. 1). Corrected drawing sheets in compliance with 37 CFR 1.121(d) are required in reply to the Office action to avoid abandonment of the application. Any amended replacement drawing sheet should include all of the figures appearing on the immediate prior version of the sheet, even if only one figure is being amended. Each drawing sheet submitted after the filing date of an application must be labeled in the top margin as either “Replacement Sheet” or “New Sheet” pursuant to 37 CFR 1.121(d). If the changes are not accepted by the examiner, the applicant will be notified and informed of any required corrective action in the next Office action. The objection to the drawings will not be held in abeyance.
Claim Rejections - 35 USC § 103
The text of those sections of Title 35, U.S. Code not included in this action can be found in a prior Office action.
Claim(s) 1-5, 7-8, 11-13, 15, 17 and 19-23 are rejected under 35 U.S.C. 103 as being unpatentable over Bastug et al. US 2022/0272035 A1 in view of Cao et al. US 2021/0357256 A1.
Regarding claim 1, Bastug discloses:
a network configuration controller (network controller – paragraph 30, Fig. 1 item 120) comprising one or more processors (network controller provides control functions such as provisioning, monitoring etc. – paragraph 30; Bastug discloses an apparatus/computer which comprises a processor – paragraph 3, Fig. 18, paragraph 153; Bastug teaches the computer/controller perform the described functions – see paragraphs 30, 156; thus the network controller comprises a processor) configured to:
monitor a plurality of node-related flow information sets based on flow information corresponding to a plurality of data flows between a first plurality of endpoints and a second plurality of endpoints via a network connecting between a plurality of processor (network has plurality of flows, and plurality of endpoints – see paragraph 3, Figs. 1-2; network controller performs monitoring function – paragraph 30; endpoints have processors), the plurality of node- related flow information sets corresponding to a plurality of networking nodes connecting between a plurality of network inputs of the network and a plurality of network outputs of the network, wherein a node-related flow information set corresponding to a networking node of the plurality of networking nodes comprises information corresponding to one or more data flows communicated via the networking node (nodes have inputs and outputs – Figs. 1-2; flow information corresponds to data flows – see paragraph 3 which teaches state information for set of nodes and corresponding set of flows; the state information is an “information set” that comprises information); and
determine a network-configuration setting to configure the network based on the plurality of node-related flow information sets and at least one target End to End (E2E) performance parameter (state information includes requirements such as latency, performance, etc. – see paragraph 3; configure network resources based on the node/flow state information in order to support deterministic connections which have performance guarantees – see paragraphs 30-32; also see Fig. 4 and paragraphs 54-66 which teaches a controller determining configuration/provisioning information based on flow information), the at least one target E2E performance parameter corresponding to an E2E performance of the plurality of data flows between the first plurality of endpoints and the second plurality of endpoints … wherein the network configuration controller is configured to monitor the flow information in real time, and to update the network configuration setting based on a detected real-time change in the plurality of node-related flow information sets (performance requirement is for the data flows – paragraphs 3, 31-32; network controller is configured to perform network monitoring – paragraph 30, and update configuration settings dynamically in response, i.e. “adaptive” – see abstract, paragraph 3; system responds to “real-time” performance metrics such as latency, jitter, etc. – paragraph 116); and
provide output information based on the network-configuration setting (promote provision/configuration determination for implementation at local nodes/controllers– Figs. 4-5, paragraphs 3, 57, 81).
Bastug does not explicitly disclose an AI training cluster or the at least one target E2E performance parameter comprising a job completion time (JCT) of the AI training cluster to complete an AI training procedure. But this is taught by Cao. Cao discloses an AI training cluster (abstract, paragraph 12, Figs. 1-2, 5), optimized for job completion time performance (abstract, paragraph 13). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Bastug with the AI training cluster and job completion time parameter taught by Cao for the purpose of configuring network resources. Cao suggests optimization of resource configuration is determined so as to achieve the shortest possible JCT (paragraph 39, Fig. 4). One of ordinary skill in the art would recognize the benefit of faster training.
Regarding claim 2, Bastug discloses the node-related flow information set … identifies a network input and a network output corresponding to the flow … and next-hop information to identify a next-hop node corresponding to the flow (deterministic path is from source to destination including each hop along the path – paragraph 32; includes interfaces for a route – paragraph 113). Bastug does not explicitly disclose source/destination address information or source/destination port information, or data length information. But such information is extremely well-known routine and conventional in the art. So 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 information set of Bastug to include address, port and length/size data. This is merely the incorporation of a well-known technique (e.g. all IP packets include this information) in order to yield a predictable result.
Regarding claim 3, Bastug discloses the network configuration controller is configured to update the network-configuration setting … based on the detected real-time change in the plurality of node-related flow information sets (support adaptive deterministic routing – abstract; system is dynamic – paragraph 91; real-time performance metrics include latency, jitter, etc. – see paragraphs 31-32). Bastug does not explicitly disclose a criteria relating to a predicted JCT for the AI training cluster but this is taught by Cao as explained above. The motivation to combine is the same.
Regarding claim 4, Bastug discloses the detected real-time change in the plurality of node-related flow information sets comprises a change indicative of an expected degradation in the at least one target E2E performance parameter (indicate expected flow metrics – paragraph 3).
Regarding claim 5, Bastug discloses the network configuration controller is configured to determine a first network-configuration setting based on a first plurality of node-related flow information sets corresponding to first flow information related to a first time frame, and to determine a second network-configuration setting, different from the first network- configuration setting, based on a second plurality of node-related flow information sets, different from the first plurality of node-related flow information sets (multiple flows, each has state information; thus there are multiple configurations/allocations determined – paragraphs 3, 30-32), corresponding to second flow information related to a second time frame subsequent to the first time frame (repeat process for each request – Figs. 4-5).
Regarding claim 7, Bastug discloses the network configuration controller comprises a ML engine trained to generate ML output information based on an ML input, which is based on the plurality of node-related flow information sets, wherein the network configuration setting is based on the ML output information (agent determine paths using neural network that takes input and generates output – paragraphs 3, 103-107, Figs. 7-8 and 10).
Regarding claim 8, Bastug discloses the network configuration controller is configured to determine network topography information corresponding to a network topology of the network based on the plurality of node-related flow information sets (support representation of network topology including discovery of link states for routing – paragraph 146; also see paragraph 96 which teaches the network controller queries are performed based on the topology); wherein the ML input information is based on the network topography information (agent determine paths using neural network that takes input and generates output – paragraphs 3, 103-107, Figs. 7-8 and 10; input is state information – Fig. 8, paragraph 107).
Regarding claim 11, Bastug discloses the ML engine comprises a Deep Reinforcement Learning (DRL) engine (use deep reinforcement learning techniques – paragraph 91) configured to generate the ML output information comprising action information based on the ML input comprising observation information and reward information (rewards/reward function – paragraphs 3 and 106), wherein the observation information is based on the plurality of node-related flow information sets (state information – paragraph 3), the reward information is based on the at least one target E2E performance parameter (performance requirements – paragraphs 30-32), the network-configuration setting is based on the action information (use deep reinforcement learning agents to provision the deterministic flow – paragraph 93-94, Fig. 7).
Regarding claim 12, Bastug discloses the network configuration controller is configured to determine network topography information corresponding to a network topology of the network based on the plurality of node-related flow information sets (support representation of network topology including discovery of link states for routing – paragraph 146; also see paragraph 96 which teaches the network controller queries are performed based on the topology); and to determine the network-configuration setting based on the network topography information (deterministic paths are based on state information and network topology – see paragraphs 3, 28 and Figs. 1-2).
Regarding claim 13, Bastug discloses the network topography information comprises a topography map to map [flows] to a plurality of ingress-egress [pairs] (map topology including nodes, flows and pairs – see Figs. 1-3 and paragraph 83), the plurality of ingress-egress pairs corresponding to a plurality of ingress … of the plurality of network nodes and a plurality of egress … of the plurality of networking nodes (this is just the definition of a pair). Bastug does not explicitly disclose data sizes or ports. But such information is extremely well-known routine and conventional in the art. So 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 topology of Bastug to include port and length/size data. This is merely the incorporation of a well-known technique (e.g. all IP packets include this information) in order to yield a predictable result.
Regarding claim 15, Bastug discloses the network configuration controller is configured to determine network topology information corresponding to a network topology of the network based on the plurality of node-related flow information sets (support representation of network topology including discovery of link states for routing – paragraph 146; also see paragraph 96 which teaches the network controller queries are performed based on the topology), and to determine the network topography information based on the network topology information, wherein the network topology information comprises routing information corresponding to active data flow routes between the plurality of networking nodes (represent topology including active data flows between nodes – Figs. 1-3 and paragraph 83).
Regarding claim 17, Bastug discloses the network-configuration setting comprises one or more node specific parameter settings corresponding to one or more networking nodes of the plurality of networking nodes (nodes receive individual configuration settings – paragraph 57, Fig. 4).
Regarding claim 19, Bastug discloses the network configuration controller is configured to update the network-configuration setting in real-time to update an active data flow route of an active data flow in the network (support adaptive deterministic routing – abstract; system is dynamic – paragraph 91; routes are adaptive based on monitoring and paths are determined dynamically – paragraph 116).
Regarding claim 20, Bastug discloses the target E2E performance parameter comprises at least one of a usage efficiency, quality of experience, quality of service, delay, bandwidth, or power consumption of the network (performance measures include latency, jitter, and reliability requirements – paragraphs 3, 31; thus they comprise at least quality and delay).
Regarding claim 21, Bastug discloses the network configuration controller is configured to update the network-configuration setting in real time such that a predicted time … is not to exceed a target time (in a score based reward system, an expected completion time is compared to the actual completion time – see Fig. 12, paragraph 135). Bastug does not explicitly disclose JCT of the AI training cluster but this is taught by Cao as discussed above. The motivation to combine is the same.
Regarding claim 22, it is a system that corresponds to the apparatus of claim 1. The corresponding limitations are rejected for the same reasons. The motivation to combine Cao is the same as that given above. Bastug also discloses a network comprising a plurality of networking nodes connecting between a plurality of network inputs of the network and a plurality of network outputs of the network (paragraph 3, Figs. 1-2); and a network controller configured to control the network based on the network configuration setting (network controller controls the network (i.e. provisions/allocates) based on the determined settings – paragraph 30; also see Fig. 4, paragraphs 57-59).
Regarding claim 23, it is a system claim that corresponds to the apparatus of claim 7; thus it is rejected for the same reasons.
Claim(s) 9-10 are rejected under 35 U.S.C. 103 as being unpatentable over Bastug and Cao in view of Sanchez Charles et al. US 2020/0136957 A1.
Regarding claim 9, Bastug does not explicitly disclose the network configuration controller is configured to determine size-reduced network topography information by reducing a size of the network topography information based on a size of the ML input, wherein the ML input is based on the size-reduced network topology information. But this is taught by Sanchez as an autonomous network controller that uses AI/ML to optimize network routing by analyzing topology and reducing a search space for the ML model by reducing/eliminating some network nodes (i.e. “reduced representation”, see abstract, Figs. 1, 3 and paragraphs 14 and 28; paragraph 29 discloses the reduced size can be based on the resources for the ML model). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Bastug with the topology size reduction taught by Sanchez for the purpose of determining network routes. Sanchez teaches this optimizes the network automatically and lightens the load on the AI/ML model (paragraph 10).
Regarding claim 10, Bastug does not disclose the size-reduced network topography information comprises network topography information corresponding to a subset of networking nodes, the ML output information corresponding to the subset of networking nodes. But this is taught by Sanchez as the size reduced topology has nodes removed/eliminated and thus is a subset of nodes. Furthermore, the ML output corresponds to the subset because the eliminated nodes cannot be selected (see abstract, paragraphs 28-29 and Figs. 1-3). The motivation to combine is the same as that given above.
Claim(s) 14 is rejected under 35 U.S.C. 103 as being unpatentable over Bastug and Cao in view of Gray et al. US 2011/0305143 A1.
Regarding claim 14, Bastug does not explicitly disclose the network topography information comprises a plurality of statistical ingress data sizes corresponding to the plurality of ingress ports, and a plurality of statistical egress data sizes corresponding to the plurality of egress ports, wherein a statistical ingress data size corresponding to an ingress port is based on statistical data flow sizes mapped to ingress-egress port pairs comprising the ingress port, wherein a statistical egress data size corresponding to an egress port is based on statistical data flow sizes mapped to ingress-egress port pairs comprising the egress port. But this is taught by Gray as a network topology discovery mechanism (Fig. 1, abstract) that discovers and records the statistical data sizes corresponding to ingress/egress ports (Figs. 2-4, paragraphs 42 and 62-63). It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Bastug with the statistical ingress/egress data size corresponding to the ports as taught by Gray for the purpose of improving network communication. Gray suggests knowing data size is important to prevent loops (paragraphs 25-27) and improve performance (paragraphs 28-29).
Claim(s) 6, 16 and 18 are rejected under 35 U.S.C. 103 as being unpatentable over Bastug and Cao in view of Lee et al. US 2020/0280518 A1.
Regarding claim 6, Bastug discloses the network configuration controller is configured to determine a … state of the plurality of data flows via the network based on the plurality of node related flow information sets and determine the network configuration setting based on the … state of the plurality of data flows via the network and the at least one target E2E performance parameter (network controller uses state information and performance requirements to determine network configuration settings as explained above – see abstract, paragraphs 3, 30-32, Figs. 1-2 and 4-5). Bastug does not explicitly disclose “predicting” a state of the flow or determine configuration settings based on “the predicted” state. But this is clearly taught by Lee as using network/node state information in a probabilistic function to predict congestion and then make configuration adjustments based on the prediction (paragraphs 25 and 29-30).
It would have been obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to modify Bastug to make predictions as taught by Lee for the purpose of avoiding congestion. Lee teaches that by predicting congestion, proactive actions can be taken to reduce congestion and delay (paragraphs 24-27).
Regarding claim 16, Bastug does not explicitly disclose determining at least one ingress-port setting, the ingress-port setting comprising one of ECN or maximum buffer queue size setting, wherein the at least one ingress-port setting is configured such that any Priority Flow Control event based on a PFC setting is not to occur before an ingress-port event based on the at least one ingress-port setting. But this is taught by Lee as using ECN (ECN - paragraph 16, Fig. 1). Lee further teaches improving pause times with the use of ECN to avoid PFC which causes issues such as head-of-line blocking; thus the port is configured such that ECN occurs before any PFC event (see paragraphs 23 and 25-28). It would have been obvious to one of ordinary skill in the art to modify Bastug to use ECN as taught by Lee. ECN is very well-known in the art and yields predictable results. Also, Lee explicitly suggests this reduces congestion and delay (paragraph 27).
Regrading claim 18, Bastug does not explicitly disclose a node-specific parameter setting comprises at least one of PFC, ECN, or maximal buffer queue size. But this is taught by Lee as discussed above. The motivation to combine is the same.
Conclusion
The prior art made of record and not relied upon is considered pertinent to applicant's disclosure.
Francini et al. US 2021/0250301 A1 discloses configuring networks to support delay guarantees, the network including a plurality of endpoints and flows (abstract, paragraph 19, Fig. 1).
Clemm et al. US 2020/0052979 A1 discloses performing network SLO and KPI validation for network flows between nodes (abstract, paragraph 36, Fig. 1).
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 JASON D RECEK whose telephone number is (571)270-1975. The examiner can normally be reached Flex M-F 9-5.
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, Umar Cheema can be reached at 571-270-3037. 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.
/JASON D RECEK/Primary Examiner, Art Unit 2458
Prosecution Insights