DETAILED ACTION
Notice of Pre-AIA or AIA Status
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
Response to Remarks
This communication is considered fully responsive to the Amendment filed on 6/16/26.
Claim objection (claim 4) withdrawn since amended accordingly.
112 rejections of claims withdrawn since amended accordingly.
Response to Arguments
Applicant's arguments filed 6/16/26 have been fully considered but they are not persuasive.
1] applicant argues: (emphasis added)
Patel is generally directed to distributed control plane management of distributed systems. See Patel, [0001]. Pabon is generally directed to using hierarchical control planes to manage distributed computing environments. See Pabon, [0024]. Patel and Pabon, however, at least fail to disclose the above-emphasized claim language. For example, Patel and Pabon fail to describe determining a route for migrating a workload. Regarding previous claim 5, the Office Action asserts that Banerjee discloses "determining, based on the data associated with the decentralized hierarchy, a route to the target". See OA, page 14. Applicant respectfully traverses this assertion. Banerjee is generally directed to nodes for handling a prediction of an event. See Banerjee, [0001]. More specifically, Banerjee describes the amount of mutual information between nodes corresponding to a strength of a connection between the nodes. See OA, page 14 (citing Banerjee, [0107]). Further, Banerjee describes creating an adjacency matrix from a
network graph based on a geographic location of nodes. See Banerjee, [0 107] - [0 11 0]. Banerjee, however, is at least silent as to the above-emphasized claim language.
The examiner respectfully disagrees.
Specifically, as to claim 5, Patel, Pabon and Banerjee further disclose determining, based on the data associated with the decentralized hierarchy, a route to the target (Banerjee: fig 1-14, [0017-335]: ... actions enable to perform a hierarchical, decentralized learning method, which may align, for example, with a fog layered architecture [0025] ... availability of large amounts of data, such as those collected for example, from IoT devices, may be understood to enable the possibility of analysing such data to make predictions on events, with a high predictive power and to make predictions on events (determining, based on the data associated with the decentralized hierarchy ...) may be understood to refer to building mathematical models that may fit those data, which mathematical models may then be used to make predictions for such events, e.g., what is the fastest route to reach a destination (determining, based on the data associated with the decentralized hierarchy, a route to the target) when there may be most traffic at a particular junction, etc [0007] ... Federated Learning (FL)-based, decentralized approach is used where the goal is to train a high-quality centralized model, while training data may remain distributed over a large number of clients in a network and each client may independently compute an update to the current model based on its local data (see with [0007] above - determining, based on the data associated with the decentralized hierarchy, a route to the target) [0008] ... in a distributed computing paradigm, a single processor or computing resource may need to be identified as an organizer, or leader, to distribute some task among other computing resource, which may be referred to as followers ... leader nodes may need to take this charge of coordinating the collaborative process among a group of similar nodes by assigning the modular task, gathering the results and communicating the aggregated result to outside of the group (see with [0007-8] above - determining, based on the data associated with the decentralized hierarchy, a route to the target) [0014]),
wherein causing the workload to be migrated to the target comprises causing the workload to be migrated to the target via the route (Banerjee: fig 1-14, [0017-335]: ... determining is configured to be based on: i) the first pattern of arrival of service requests received over the first period of time by the first plurality of nodes (see with [0007-8;14] above - determining, based on the data associated with the decentralized hierarchy, a route to the target) ... vote is for one of the nodes configured to be comprised in the first set of nodes to be candidate for leader node of the first set of nodes in the handling of the prediction of the event (see with [0007-8;14] above -wherein causing the workload to be migrated to the target ...) ... and the vote is configured to be based on at least one of the following with respect to the candidate: i) the available energy resources, ii) the average uptime, iii) the available computing power, iv) the computing latency, v) the communication constraint, the communication constraint being configured to be based on the energy cost and the number of connections of the candidate with the other nodes in the first set of nodes (see with [0007-8;14] above - ... comprises causing the workload to be migrated to the target via the route) [0022] ... different vertical application use cases may use a fog hierarchy differently ... in most fog deployments, there may usually be several tiers (N-tiers) of nodes and the particular example depicted in fig 1, the lower level devices provide information to respective roadside traffic fog devices 16, which may in turn feed information to the next level fog nodes, neighbourhood traffic fog devices 17, which may then provide information to the next level fog nodes, regional traffic fog devices 18, which may in turn provide information to an Enterprise Mobility Suite (EMS) cloud 19 or a Metropolitan Traffic Services Cloud ... tiers may be created in order to deal efficiently with the amount of data that may need to be processed and provide better operational and system intelligence (see with [0007-8,14; 22] above - wherein causing the workload to be migrated to the target comprises causing the workload to be migrated to the target via the route) [0051] ... in contrast to CFL, generates clusters of fog nodes considering the context awareness and may then execute decentralized learning with the leader nodes as coordinators ... to allow fog-based analytics to act fast for dynamic events and be helpful to the decision making of nodes connected for information related to specific locations and this information may be extended to use in the development of the entire fog networking for specific tasks (see with [0007-8,14; 22; 51] above - wherein causing the workload to be migrated to the target comprises causing the workload to be migrated to the target via the route) [0056-57;220]).
Applicant’s other arguments with respect to claims have been considered but are moot in view of new ground(s) of rejection.
Claim Rejections - 35 USC § 103
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 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 1-11, 16-17 and 19-20 are rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Publication No. 2024/0236109 to Patel et al. (“Patel”) in view of U.S. Patent Publication No. 2025/0077295 to Pabon et al. (“Pabon”) in view of U.S. Patent Publication No. 2023/0169356 to Banerjee et al. (“Banerjee”) and further in view of U.S. Patent Publication No. 2017/0201423 to Crowe et al. (“Crowe”).
As to claim 1, Patel discloses a method (Patel: fig 1-4), comprising:
obtaining an indication of a workload associated with a decentralized hierarchical control plane (Patel: fig 1-4, [0004-129]: fig 1 & 2A-F ... facilitate performance of workloads e.g., computer-implemented workloads performed by executing computing instructions (obtaining an indication of a workload ... ) [0026] ... ... the distributed control plane may vest decision making authority at different levels of a hierarchy used to manage data processing systems (... a decentralized hierarchical control plane) [0012] ... after providing desired use of the data processing systems to a client using the subscribed data processing systems, making, by the local control plane, a second determination that the local control plane is communicatively connected to other control plane elements and performing, by the leader, a validation of the temporary deployment plan (see with [0012] above - obtaining an indication of a workload ... ) with the other control plane elements (... associated with a decentralized hierarchical control plane) ... in a first instance of the validation where the temporary deployment plan is accepted by the other control plane elements: finalizing the temporary deployment plan; and in a second instance of the validation where the temporary deployment plan is not accepted by the other control plane elements (see with [0012] above - obtaining an indication of a workload associated with a decentralized hierarchical control plane) [0020] ... deployment manager 106 may (i) obtain information e.g. intents regarding services to be provided to clients 100, (ii) generate intermediate representations based on the obtained information, (iii) distribute the intermediate representations to deployment 110 for implementation ... deployment manager 106 may operate as a member of distributed control plane that extends down to data processing systems 111 see with [0012;20] above - obtaining an indication of a workload associated with a decentralized hierarchical control plane) [0044]),
wherein the decentralized hierarchical control plane comprises a plurality of control nodes in a decentralized hierarchy (Patel: fig 1-4, [0004-129]: fig 1 & 2A-F ... distributed control plane may vest decision making authority, at least in part, at different levels (hierarchical) within the distributed control plane and modify that decision making authority responsive to changes in operable connectivity between the different levels (hierarchical) of the distributed control plane (wherein the decentralized hierarchical control plane comprises a plurality of control nodes in a decentralized hierarchy) [0045] ... deployment 110 may implement a portion (a plurality of control nodes) of the distributed control plane and the portion of the distributed control plane implemented by deployment 110 may be responsible for providing services by implementing intermediate representations provided by deployment manager 106 and/or other members (a plurality of control nodes) of other portions of the distributed control plane ...the portion of the distributed control plane implemented by deployment 110 may (i) obtain information based on the intermediate representation, (ii) identify members (a plurality of control nodes) of deployment 110 that may take on roles based on the information, (iii) cooperatively decide which member(s) of deployment 110 will take on roles in scenarios in which a role may require multiple data processing systems (wherein the decentralized hierarchical control plane comprises a plurality of control nodes in a decentralized hierarchy) [0049]).
Patel did not explicitly disclose determining, by a processing device and based on data associated with the decentralized hierarchy, a target for migration of the workload.
Pabon discloses determining, by a device comprising a processor and based on data associated with the decentralized hierarchy, a target for migration of the workload (Pabon: fig 1-7, [0002-236]: title “USING HIERARCHICAL CONTROL PLANES TO MANAGE DISTRIBUTED COMPUTING ENVIRONMENTS” ... when executed, a containerized application may provide one or more containerized workloads and/or services [0220] ... a control plane of the container system 380 may implement services that include: deploying applications via a controller 386 (determining, by a device comprising a processor ...), monitoring applications via the controller 386, providing an interface via an API server 387, and scheduling deployments via scheduler 388 (determining, by a device comprising a processor ...) ... for example, for resiliency, control plane(s) may be implemented by multiple, redundant nodes, where if a node that is providing management services for the container system 380 fails (see with title above - determining, by a device comprising a processor and based on data associated with the decentralized hierarchy ...) then another redundant node (see with [0220] above - a target for migration of the workload) may provide management services for the cluster 384 [0225]).
Patel and Pabon are analogous art because they are from the same field of endeavor with respect to hierarchical control planes.
Before the effective filing date, for AIA , it would have been obvious to a person of ordinary skill in the art to incorporate the strategies by Pabon into the method by Patel. The suggestion/motivation would have been to provide hierarchical control planes to manage distributed computing environments (Pabon: [0024]).
Patel did not explicitly disclose determining a route for migrating the workload to the target based on the data associated with the decentralized hierarchy.
Banerjee discloses determining a route for migrating the workload to the target based on the data associated with the decentralized hierarchy (Banerjee: fig 1-14, [0017-335]: ... actions enable to perform a hierarchical, decentralized learning method, which may align, for example, with a fog layered architecture [0025] ... availability of large amounts of data, such as those collected for example, from IoT devices, may be understood to enable the possibility of analysing such data to make predictions on events, with a high predictive power and to make predictions on events (determining ... for migrating the workload to the target based on the data associated with the decentralized hierarchy) may be understood to refer to building mathematical models that may fit those data, which mathematical models may then be used to make predictions for such events, e.g., what is the fastest route to reach a destination (determining a route for migrating the workload to the target based on the data associated with the decentralized hierarchy) when there may be most traffic at a particular junction, etc [0007] ... Federated Learning (FL)-based, decentralized approach is used where the goal is to train a high-quality centralized model, while training data may remain distributed over a large number of clients in a network and each client may independently compute an update to the current model based on its local data (see with [0007] above - determining a route for migrating the workload to the target based on the data associated with the decentralized hierarchy) [0008] ... in a distributed computing paradigm, a single processor or computing resource may need to be identified as an organizer, or leader, to distribute some task among other computing resource, which may be referred to as followers ... leader nodes may need to take this charge of coordinating the collaborative process among a group of similar nodes by assigning the modular task, gathering the results and communicating the aggregated result to outside of the group (see with [0007-8] above - determining a route for migrating the workload to the target based on the data associated with the decentralized hierarchy) [0014]).
Patel, Pabon and Banerjee are analogous art because they are from the same field of endeavor with respect to hierarchical fog architecture.
Before the effective filing date, for AIA , it would have been obvious to a person of ordinary skill in the art to incorporate the strategies by Banerjee into the method by Patel and Pabon. The suggestion/motivation would have been to provide for finding leader fog nodes in fog networking to smoothen the flow of communication and execution of special tasks will (Banerjee: [0043]).
Patel did not explicitly disclose transmitting migration data to the target via the route to cause the workload to be migrated to the target via the route.
Crowe discloses transmitting migration data to the target via the route to cause the workload to be migrated to the target via the route (Crowe: fig 1-8, [0006-104]: ... mapping (302) for each of a plurality of workloads operating on source devices in a networked computing environment, a corresponding target device within said networked computing environment; accessing (304) topological information (352) defining the networked computing environment and reducing (316) unbalanced utilization of infrastructure in the networked computing environment, where the reducing (316) may include migrating a workload to an alternate device (target device) (transmitting migration data to the target ...) within the networked computing environment, where the migrating is in dependence upon the mapping and topological information ... mapping (302) may include determining a mapping of a source device to a target device (... via the route to cause the workload to be migrated to the target via the route) ... a mapping of a source device to a target device may be determined based on a correspondence to a route through devices corresponding to a previous data communication for an access from a workload to a workload target-where a source device may be the first device along the route and the target device may be the last device along the route (transmitting migration data to the target via the route to cause the workload to be migrated to the target via the route) or a source device may be a device on which a workload executes [0083-84] ... there may be different numbers of levels and different numbers of connections in a given network topology representing a networked computing environment (208) ... such that a data communication path is provided from a workload to a workload target and the network topology may be hierarchically organized, where different devices at different levels may have difference characteristics, such as different performance characteristics (transmitting migration data to the target via the route to cause the workload to be migrated to the target via the route) [0080] ... topological information (352) may be used to determine which other source devices, target devices, or routes through intermediary devices, may be available for remapping (migrating) such that workload on previously mapped devices are at least partially offloaded onto newly mapped devices (transmitting migration data to the target via the route to cause the workload to be migrated to the target via the route) [0088;95]).
Patel, Pabon, Banerjee and Crowe are analogous art because they are from the same field of endeavor with respect to utilization of infrastructure.
Before the effective filing date, for AIA , it would have been obvious to a person of ordinary skill in the art to incorporate the strategies by Crowe into the method by Patel, Pabon and Banerjee. The suggestion/motivation would have been to provide for migration of a workload from one source device to another, or a migration of a workload from one target device to another to improve balanced utilization of the infrastructure of the data center (Crowe: [0021]).
As to claim 2, see similar rejection to claim 1 where the method is taught by the method.
As to claim 2, Patel, Pabon, Banerjee and Crowe further disclose wherein determining the target for the migration of the workload comprises determining the target for the migration of the workload based on a consensus of the plurality of control nodes, wherein the consensus is based on the data associated with the decentralized hierarchy (Pabon: fig 1-7, [0002-236]: ... authority owners have exclusive right to modify entities, to migrate entities from one non-volatile solid state storage unit to another non-volatile solid state storage unit (wherein determining the target for the migration of the workload comprises ...), and to add and remove copies of entities and this allows for maintaining the redundancy of the underlying data ... transient failures make it non-trivial to ensure that all non-faulty machines agree upon the new authority location (... determining the target for the migration of the workload ...) and ... can be achieved automatically by a consensus protocol such as Paxos (see with title & [0220;225] above - based on a consensus of the plurality of control nodes, wherein the consensus is based on the data associated with the decentralized hierarchy), hot-warm failover schemes, via manual intervention by a remote system administrator, or by a local hardware administrator (such as by physically removing the failed machine from the cluster, or pressing a button on the failed machine) and in some embodiments, a consensus protocol is used, and failover is automatic [0096];
Patel: fig 1-4, [0004-129]: ... negotiating, by the leader and using the temporary deployment plan, agreements with the data processing systems to implement at least one of the roles (wherein determining the target for the migration of the workload comprises ...) ... negotiating agreements may include, for a role of the roles: selecting a data processing system of the data processing systems to perform the role (... determining the target for the migration of the workload ...); distributing proposals to all of the data processing systems indicating the selected data processing system for the role; obtaining responses from all of the data processing systems based on the distributed proposals; in a first instance of the responses that indicate unanimous assent, recording the data processing system as performing the role in the temporary deployment plan (... based on a consensus of the plurality of control nodes, wherein the consensus is based on the data associated with the decentralized hierarchy); and in a second instance of the responses that does not indicate unanimous assent, selecting a different data processing system of the data processing systems to perform the role [0018-19]).
For motivation, see rejection of claim 1.
As to claim 3, see similar rejection to claims 1-2.
As to claim 3, Patel, Pabon, Banerjee and Crowe further disclose wherein the target comprises at least one of:
a control node in the plurality of control nodes,
a non-control node in the decentralized hierarchy (Patel: fig 1-4, [0004-129]: ... for example, consider a scenario where deployment 110 provides tiered data storage services to clients 100 and to provide the tiered data storage services, data processing systems 112-113 may need to (i) intake data, (ii) select a storage location for data, (iii) preprocess the data prior to storage ( e.g., deduplication), (iv) store the data in the storage location, and (v) migrate the data between storage locations (a non-control node in the decentralized hierarchy) so as to properly tier the data as its relevance/ importance changes overtime [0030]),
an edge device in the decentralized hierarchy,
cloud computing resources associated with the decentralized hierarchy, or
a virtual machine associated with the decentralized hierarchy.
For motivation, see rejection of claim 1.
As to claim 4, see similar rejection to claims 1-3.
As to claim 4, Patel, Pabon, Banerjee and Crowe further disclose wherein the target is one of a plurality of targets, wherein each of the plurality of targets is associated with a weight from a set of weights, and wherein determining the target for the migration of the workload is based on the set of weights (Banerjee: fig 1-14, [0027-335]: ... once the cluster, that is, the first set of nodes 121 may have been identified, nodes comprised in it, such as the second node 112, may cast a vote to elect the leader node of the newly defined set of nodes 111 (see with [0126;129;132 below - wherein determining the target for the migration of the workload is based on the set of weights) [0125] ... a first indication may indicate that a second node 112 is comprised in the first set of nodes 121 and by receiving the first indication, the second node 112 may be enabled to know which potential candidates it may choose from in order to elect a leader node (wherein the target is one a plurality of targets ...) [0126] ... the second node 112 determines a vote for one of the nodes comprised in the first set of nodes 121 to be candidate for leader node of the first set of nodes 121 in the handling of the prediction of an event (... and wherein determining the target for the migration of the workload is based on ...) [0129] ... each leadership score of a node may be calculated from a computing latency and a communication constraint of the node (... wherein determining the target for the migration of the workload is based on the set of weights) ... communication constraint calculated from a degree and an energy cost of the node and, for node a, the degree of node may be denoted as d(a) and degree stands for number of edges, that is, a line joining two nodes of a graph, which are incident to the node a and E(ab) may be understood as the amount of energy that may be required to send an electric signal from node a to node b and a minimum weighted spanning tree may be prepared by finding the spanning arborescence of minimum weight in the graph and the weight in each edge uv may be denoted as E(uv) and If d(a) is high and energy cost C=Summation (I element of d(a))E(ai) is low, then the node "a" may have more chances to spread the information to a large portion of the communications system 100 and the higher the degree and smaller the energy cost of a node is, the faster the node may spread the information (see with [0125-126;129] above- wherein each of the plurality of targets is associated with a weight from a set of weights, and wherein determining the target for the migration of the workload is based on the set of weights) [0132] ... EF calculated by a weighted sum of these decision factors (see with [0125-126;129;132] above- wherein each of the plurality of targets is associated with a weight from a set of weights, and wherein determining the target for the migration of the workload is based on the set of weights) [0133]).
For motivation, see rejection of claim 1.
As to claim 5, see similar rejection to claims 1-4.
As to claim 5, Patel, Pabon, Banerjee and Crowe further disclose wherein the route includes at least two layers of the decentralized hierarchical control plane or at least two nodes of the plurality of control nodes in the decentralized hierarchy (Patel: fig 1-4, [0004-129]: fig 1 & 2A-F ... the distributed control plane may vest decision making authority at different levels of a hierarchy used to manage data processing systems (the route includes at least two layers of the decentralized hierarchical control plane) [0012];
Banerjee: fig 1-14, [0027-335]: ... a third node (control node)113 sends a fifth indication indicating the updated machine-learning model to another node (control node) 114 in the communications system 100 (the route includes ...) and at least one of the following applies: i) the another node 114 is a leader node of a second set of nodes 122 comprised in the first plurality of nodes 110, and all nodes in the first plurality of nodes 110 belong to a same layer 151 of nodes in the communications system 100, see FIG. 2a), and ii) the communications system 100 comprises multiple pluralities of nodes 140 hierarchically organized in a plurality of layers 150, wherein the another node 114 is in a different layer 152 of the plurality of layers 150 than a layer 151 wherein the third node 113 is comprised (the route includes at least two layers of the decentralized hierarchical control plane or at least two nodes of the plurality of control nodes in the decentralized hierarchy) [0173]).
For motivation, see rejection of claim 1.
As to claim 6, see similar rejection to claims 1-5.
As to claim 6, Patel, Pabon, Banerjee and Crowe further disclose wherein the route is one a plurality of routes, wherein each of the plurality of routes is associated with a weight from a set of weights, and wherein determining the route for the migration of the workload is based on the set of weights (Banerjee: fig 1-14, [0027-335]: ... once the cluster, that is, the first set of nodes 121 may have been identified, nodes comprised in it, such as the second node 112, may cast a vote to elect the leader node of the newly defined set of nodes 111 (see with [0126;129;132] below & [0107-110; 0125-126;129;132] above - wherein determining the route for the migration of the workload is based on the set of weights) [0125] ... a first indication may indicate that a second node 112 is comprised in the first set of nodes 121 and by receiving the first indication, the second node 112 may be enabled to know which potential candidates it may choose from in order to elect a leader node (wherein the target is one a plurality of targets ...) [0126] ... the second node 112 determines a vote for one of the nodes comprised in the first set of nodes 121 to be candidate for leader node of the first set of nodes 121 in the handling of the prediction of an event (... and wherein determining the target for the migration of the workload is based on ...) [0129] ... each leadership score of a node may be calculated from a computing latency and a communication constraint of the node (... wherein determining the route for the migration of the workload is based on the set of weights) ... communication constraint calculated from a degree and an energy cost of the node and, for node a, the degree of node may be denoted as d(a) and degree stands for number of edges, that is, a line joining two nodes of a graph, which are incident to the node a and E(ab) may be understood as the amount of energy that may be required to send an electric signal from node a to node b and a minimum weighted spanning tree may be prepared by finding the spanning arborescence of minimum weight in the graph and the weight in each edge uv may be denoted as E(uv) and If d(a) is high and energy cost C=Summation (I element of d(a))E(ai) is low, then the node "a" may have more chances to spread the information to a large portion of the communications system 100 and the higher the degree and smaller the energy cost of a node is, the faster the node may spread the information (see with [0126;129;132] & [0107-110; 0125-126;129;132] above - wherein the route is one a plurality of routes, wherein each of the plurality of routes is associated with a weight from a set of weights) [0132] ... EF calculated by a weighted sum of these decision factors (see with [0126;129;132] & [0107-110; 0125-126;129;132] above - - wherein the route is one a plurality of routes, wherein each of the plurality of routes is associated with a weight from a set of weights) [0133]).
For motivation, see rejection of claim 1.
As to claim 7, see similar rejection to claims 1-2 & 4-6.
As to claim 7, Patel, Pabon, Banerjee and Crowe further disclose wherein determining the route to the target comprises determining the route to the target based on a consensus of the plurality of control nodes, wherein the consensus is based on the data associated with the decentralized hierarchy (Banerjee: fig 1-14, [0027-335]: ... learning criteria may require every node in the layer L(n) (plurality of control nodes) to reach consensus on weights that may best fit the dataset coming from end devices, distributed over the entire layer and learning algorithm may generate the information that nodes may exchange with each other, such as model parameters/weights, and may dictate the merging of all the information gathered at each node (see with [0126;129;132-133] & [0107-110; 0125-126;129;132] above wherein determining the route to the target comprises determining the route to the target based on a consensus of the plurality of control nodes, wherein the consensus is based on the data associated with the decentralized hierarchy) [0155] ... for example, according to the algorithm below, which may run an election and find out the leader node n(L) from all nodes of a local cluster such as the first set of nodes 121 and the Election Factor (EF) may be used by each of the nodes in a cluster, e.g., each node of the first set of nodes 121, to determine respective votes from all nodes, as described earlier (see with [0126;129;132-133] & [0107-110; 0125-126;129;132] above & see algorithm in [0143] - wherein determining the route to the target comprises determining the route to the target based on a consensus of the plurality of control nodes, wherein the consensus is based on the data associated with the decentralized hierarchy) [0143]).
For motivation, see rejection of claim 1.
As to claim 8, see similar rejection to claims 1-7.
As to claim 8, Patel, Pabon, Banerjee and Crowe further disclose obtaining the data associated with the decentralized hierarchy from at least one of a control node of the decentralized hierarchy or a non-control node of the decentralized hierarchy, wherein determining the target for the migration of the workload is additionally based on the obtained data (Banerjee: fig 1-14, [0027-335]: ... fig 9 ... clustering provides modularity in a large hierarchical and widely spread network ... leader selection methodology enables ... decentralized learning of embodiments ... enables transfer global and local beliefs (see with [0236] below- obtaining the data associated with the decentralized hierarchy from at least one of a control node of the decentralized hierarchy or a non-control node of the decentralized hierarchy) and not entire model parameters, among leader and follower fog nodes ... leader-based learning may in turn be understood to allow context awareness and modularity in fog layers and learnt global beliefs may be sent to higher layers for making complex decisions ... allows fog-based analytics to act fast for dynamic events and to the decision making of nodes connected for information related to specific locations (see with [0126;129;132-133] & [0107-110; 0125-126;129;132;143] above & see algorithm in [0143] ... wherein determining the target for the migration of the workload is additionally based on the obtained data) [0025] ... solid black arrows show how the local beliefs from the member nodes are sent to the leader node of each local cluster ... dotted black arrows show how the global belief is broadcasted from the leader node to all the nodes in a local cluster ... striped arrows show how the local belief from a member cluster is sent to the leader cluster in the same first layer 151, which is here the second set of nodes 122 ... empty arrows show how the global belief is broadcasted from the leader cluster to all the local clusters in the same first layer 151 (see with [0126;129;132-133] & [0107-110; 0125-126;129;132;143] above & see algorithm in [0143] ... wherein determining the target for the migration of the workload is additionally based on the obtained data) [0236]).
For motivation, see rejection of claim 1.
As to claim 9, see similar rejection to claims 1-8.
As to claim 9, Patel, Pabon, Banerjee and Crowe further disclose wherein the data associated with the decentralized hierarchy comprises at least one of:
resource utilization of at least one of a control node, a non-control node, a virtual machine, or an edge device,
network latency associated with the decentralized hierarchy,
device capabilities of devices in the decentralized hierarchy (Patel: fig 1-4, [0004-129]: ... selecting by the local control plane and using the intermediate representation, a data processing system of the data processing systems as a leader; obtaining ... self-reported role fit data from the data processing systems for each of the roles, the role fit data indicating estimates of an ability of each of the data processing systems to fulfill each of the roles [0017] ...during role assignment processes... evaluate their capacities for performing the roles ... (i) identify hardware and/or software needed to perform the roles as well as other characteristics to meet the criteria for each role ( e.g., power availability, connectivity, etc.), (ii) compare their available hardware, software, and/or other characteristics (e.g., in aggregate the "evaluation metrics") to the criteria for each role to identify their capability to perform each of the roles, (iii) rank or otherwise grade their evaluation metrics for each of the roles based on their capabilities to perform the roles, and/or (iv) provide role fit data to local control plane 114 ... role fit data may indicate their self-report ability for each of the roles [0071]),
device policies of the devices in the decentralized hierarchy,
device locations of the devices in the decentralized hierarchy,
energy consumption in the decentralized hierarchy, or
load balancing across the decentralized hierarchy.
For motivation, see rejection of claim 1.
As to claim 10, see similar rejection to claims 1-9.
As to claim 10, Patel, Pabon, Banerjee and Crowe further disclose transmitting, to at least one control node in the plurality of control nodes, a vote for a first proposed target for the migration of the workload; and receiving, from the at least one control node in the plurality of control nodes, votes for a second proposed target for the migration of the workload, wherein determining the target for the migration of the workload comprises determining the target based on the vote and the votes (Banerjee: fig 1-14, [0027-335]: ... leader nodes may be selected from clustered fog nodes by a Cluster Leader Selection Mechanism run by the second node (transmitting, to at least one control node in the plurality of control nodes, a vote for a first proposed target for the migration of the workload; and receiving, from the at least one control node in the plurality of control nodes ...) 112, where the election may be done to find out the leader node in a local cluster ... an Election Factor (EF) may be used to determine the respective votes from all the nodes ... leadership score of a node may be calculated from the computing latency and the communication constraint of the node (see with [0126;129;132-133] & [0107-110; 0125-126;129;132;143] above & see algorithm in [0143] & [0236] below - receiving, from the at least one control node in the plurality of control nodes, votes for a 1st 2nd ... n proposed target(s) for the migration of the workload, wherein determining the target for the migration of the workload comprises determining the target based on the vote and the votes) [232] ... fig 9 ... clustering provides modularity in a large hierarchical and widely spread network ... leader selection methodology enables ... decentralized learning of embodiments ... enables transfer global and local beliefs (see with [0126;129;132-133] & [0107-110; 0125-126;129;132;143;232] above & see algorithm in [0143] & [0236] below - receiving, from the at least one control node in the plurality of control nodes, votes for a 1st 2nd ... n proposed target(s) for the migration of the workload, wherein determining the target for the migration of the workload comprises determining the target based on the vote and the votes) and not entire model parameters, among leader and follower fog nodes ... leader-based learning may in turn be understood to allow context awareness and modularity in fog layers and learnt global beliefs may be sent to higher layers for making complex decisions ... allows fog-based analytics to act fast for dynamic events and to the decision making of nodes connected for information related to specific locations (see with [0126;129;132-133] & [0107-110; 0125-126;129;132;143;232] above & see algorithm in [0143] ... wherein determining the target for the migration of the workload is additionally based on the obtained data) [0025] ... solid black arrows show how the local beliefs from the member nodes are sent to the leader node of each local cluster ... dotted black arrows show how the global belief is broadcasted from the leader node to all the nodes in a local cluster ... striped arrows show how the local belief from a member cluster is sent to the leader cluster in the same first layer 151, which is here the second set of nodes 122 ... empty arrows show how the global belief is broadcasted from the leader cluster to all the local clusters in the same first layer 151 (see with [0126;129;132-133] & [0025;107-110; 0125-126;129;132;143;232] above & see algorithm in [0143] - receiving, from the at least one control node in the plurality of control nodes, votes for a 1st 2nd ... n proposed target(s) for the migration of the workload, wherein determining the target for the migration of the workload comprises determining the target based on the vote and the votes) [0236]).
For motivation, see rejection of claim 1.
As to claim 11, see similar rejection to claims 1-10.
As to claim 11, Patel, Pabon, Banerjee and Crowe further disclose wherein obtaining the indication of the workload associated with the decentralized hierarchical control plane comprises obtaining the indication of the workload based on a device executing the workload in the decentralized hierarchy becoming inactive or based on the device being predicted to become inactive (Banerjee: fig 1-14, [0027-335]: ... determines a vote for one of the nodes comprised in the first set of nodes 121 to be candidate for leader node of the first set of nodes 121 in the handling of the prediction of the event ... vote(s) is/are based on at least one of the following with respect to the candidate: i) available energy resources, e.g., how much battery the candidate may have to be to perform a task; ii) average uptime, wherein uptime may be understood as an average time for which the node may remain up, that is, the node may have internal battery power; iii) available computing power, e.g., CPU power, memory and storage; iv) computing latency, that is, an amount of time, e.g., in milliseconds of seconds, that may be required to perform a computation job, v) a communication constraint, the communication constraint being based on an energy cost and a number of connections, e.g., hops, of the candidate with the other nodes in the first set of nodes 121,( wherein obtaining the indication of the workload associated with the decentralized hierarchical control plane comprises ...) and vi) a probability of state change of the candidate, e.g., changing of state from ON to OFF, ACTIVE to INACTIVE etc (... obtaining the indication of the workload based on a device executing the workload in the decentralized hierarchy becoming inactive or based on the device being predicted to become inactive) [0129]).
For motivation, see rejection of claim 1.
As to claims 16-17, see similar rejection to claims 1-2, respectively, where the system is taught by the method.
As to claims 19-20, see similar rejection to claims 1-2, respectively, where the medium is taught by the method.
Claim 12 is rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Publication No. 2024/0236109 to Patel et al. (“Patel”) in view of U.S. Patent Publication No. 2025/0077295 to Pabon et al. (“Pabon”), U.S. Patent Publication No. 2023/0169356 to Banerjee et al. (“Banerjee”), U.S. Patent Publication No. 2017/0201423 to Crowe et al. (“Crowe”) and further in view of U.S. Patent Publication No. 2022/0141099 to Kumar et al. (“Kumar”).
As to claim 12, see similar rejection to claims 1-11 where the method is taught by the method.
For motivation, see rejection of claim 4.
Patel did not explicitly disclose wherein causing the workload to be migrated to the target comprises causing the workload to be migrated from a first layer of the decentralized hierarchy to a second layer of the decentralized hierarchy.
Kumar discloses wherein causing the workload to be migrated to the target comprises causing the workload to be migrated from a first layer of the decentralized hierarchy to a second layer of the decentralized hierarchy (Kumar: fig 1-7, [0006-168]: ... migration refers to reallocation of the workload amongst the one or more nodes having untapped computing resources ... path of the communication channel after migration depends on the one or more nodes within the network that are identified to have untapped computing resources (wherein causing the workload to be migrated to the target comprises ...) ... if a mixture of utilised nodes and unutilised nodes are identified ... a new node for migrating workload that is in respect of an application can lie in any part of the network, such as on premise datacentres; or cloud infrastructure, e.g. private cloud, public cloud and edge cloud (... causing the workload to be migrated from a first layer of the decentralized hierarchy to a second layer of the decentralized hierarchy) [0049]).
Patel, Pabon, Banerjee, Crowe and Kumar are analogous art because they are from the same field of endeavor with respect to multi-layered control and multi-point control.
Before the effective filing date, for AIA , it would have been obvious to a person of ordinary skill in the art to incorporate the strategies by Kumar into the method by Patel, Pabon, Banerjee and Crowe. The suggestion/motivation would have been to simplify deployment, operation and optimization that guarantees applications performance (Kumar: [0005]).
Claim 13 is rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Publication No. 2024/0236109 to Patel et al. (“Patel”) in view of U.S. Patent Publication No. 2025/0077295 to Pabon et al. (“Pabon”), U.S. Patent Publication No. 2023/0169356 to Banerjee et al. (“Banerjee”), U.S. Patent Publication No. 2017/0201423 to Crowe et al. (“Crowe”), U.S. Patent Publication No. 2022/0141099 to Kumar et al. (“Kumar”) and further in view of U.S. Patent Publication No. 2025/0371349 to Xu et al. (“Xu”).
As to claim 13, see similar rejection to claims 1-12 where the method is taught by the method.
As to claim 13, Patel, Pabon, Banerjee, Crowe and Kumar further disclose establishing a baseline state of the decentralized hierarchy (Kumar: fig 1-7, [0006-168]: autonomous optimisation decisions made using predictive analytics to ensure that application performance always meets expected baseline [0165]).
For motivation, see rejection of claim 12.
Patel did not explicitly disclose wherein determining the target for the migration of the workload comprises: providing the baseline state and the data associated with the decentralized hierarchy as input to at least one of a heuristic procedure or a machine learning (ML) model; and obtaining, as an output of at least one of the heuristic procedure or the ML model, an indication of the target for the migration.
Xu discloses wherein determining the target for the migration of the workload comprises: providing the baseline state and the data associated with the decentralized hierarchy as input to at least one of a heuristic procedure or a machine learning (ML) model (Xu: fig 1-15, [0004-166]: fig 1 ... computing system 102 includes a first model training controller 104A e.g., a first instance of the model training controller 104AE, a second model training controller 104B e.g., a second instance of the model training controller 104A-E, a third model training controller 104C e.g., a third instance of the model training controller 104A-E), a fourth model training controller 104D e.g., a fourth instance of the model training controller 104A-E, and a fifth model training controller 104E e.g., a second instance of the model training controller 104A-E, collectively referred to herein as the model training controller 104A-E (the decentralized hierarchy) [0059] ... fig 13 ... block 1302, at which the example model training controller 104A-E selects a baseline machine learning model to train ... determines a hardware configuration of a target hardware platform on which the machine learning model is to be executed, for example, the configuration determiner 220 (FIG. 2) can determine at least one of processor configuration information, memory configuration information, or caching configuration information included in the hardware configuration(s) 122 of FIG. 1 (see with [0059] above - wherein determining the target for the migration of the workload comprises: providing the baseline state and the data associated with the decentralized hierarchy as input to at least one of a heuristic procedure or a machine learning (ML) model) [0156-157] );
and obtaining, as an output of at least one of the heuristic procedure or the ML model, an indication of the target for the migration (Xu: fig 1-15, [0004-166]: fig 13 ... at block 1308, the example model training controller 104A-E determines an embedding state for the layer, for example, the model training handler 240 (FIG. 2) can direct, instruct, and/or otherwise invoke the architectural benchmark handler 520 to output the environment output data 512, which can include the embedding state as defined by the example of Array (1) above ... block 1320, the example model training controller 104A-E deploys the trained machine learning model to the target hardware platform, for example, the deployment controller 260 (FIG. 2) can deploy, distribute, and/or otherwise provide the best reward sparse model 506 to one or more target hardware platforms (see with [0059;156-157] above - and obtaining, as an output of at least one of the heuristic procedure or the ML model, an indication of the target for the migration) [0159; 166]).
Patel, Pabon, Banerjee, Crowe, Kumar and Xu are analogous art because they are from the same field of endeavor with respect to machine learning.
Before the effective filing date, for AIA , it would have been obvious to a person of ordinary skill in the art to incorporate the strategies by Xu into the method by Patel, Pabon, Banerjee, Crowe and Kumar. The suggestion/motivation would have been to provide model training controller that can train the neural network to achieve high performance on the target hardware platform relative to a baseline version of the neural network (Xu: [0041]).
Claim 14 is rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Publication No. 2024/0236109 to Patel et al. (“Patel”) in view of U.S. Patent Publication No. 2025/0077295 to Pabon et al. (“Pabon”), U.S. Patent Publication No. 2023/0169356 to Banerjee et al. (“Banerjee”), U.S. Patent Publication No. 2017/0201423 to Crowe et al. (“Crowe”), U.S. Patent Publication No. 2022/0141099 to Kumar et al. (“Kumar”) and further in view of U.S. Patent Publication No. 2020/0403847 to Goodman et al. (“Goodman”).
As to claim 14, see similar rejection to claims 1-13 where the method is taught by the method.
As to claim 14, Patel, Pabon, Banerjee, Crowe and Kumar further disclose herein obtaining the indication of the workload, determining the target for the migration of the workload, and causing the workload to be migrated to the target are performed by a control node in the plurality of control nodes, wherein the control node possesses a state of the decentralized hierarchy (Kumar: fig 1-7, [0006-168]: fig 6 ... step 602 application performance metrics of data communication between the devices 108, 110 and 310 is measured (obtaining the indication of the workload) ... step 604 the application performance metrics against performance requirements is compared (determining the target for the migration of the workload) ... step 606 nodes having untapped computing resources within the network 100 detected (wherein the control node possesses a state of the decentralized hierarchy) and in response to the application performance metrics being below the performance requirements ... step 608, operation parameters achieving service at the performance requirements are determined ... step 610 one or more of the nodes having untapped computing resources are commanded to function at the operation parameters ... step 612 at least a portion of workload associated with the data communication is migrated amongst the one or more nodes commanded to function at the operation parameters (... causing the workload to be migrated to the target are performed by a control node in the plurality of control nodes, wherein the control node possesses a state of the decentralized hierarchy) [0148]).
For motivation, see rejection of claim 12.
Patel did not explicitly disclose wherein the state is less than a full state of the decentralized hierarchy.
Goodman discloses wherein the state is less than a full state of the decentralized hierarchy (Goodman: fig 1-11, [0004-76]: ... while availability zones are typically configured with data plane capacity as well as their own control planes, a child area may only include data plane capacity or some limited control plane functionality which, for latency or other reasons, require close proximity to the data plane ... a child area can be implemented in a variety of ways ... a child area may be a subset of the provider network that has more limited capacity than a typical area ... this "outpost" of the provider network can include a limited quantity of capacity e.g., compute and storage resources such that provision of a local control plane within the outpost would constitute a significant percentage reduction in the capacity (wherein the state is less than a full state of the decentralized hierarchy) [0022] ... a child area may only update the asset monitoring service 122 when new hardware resources are added to the child area e.g., when a new server rack is installed, or other resources are added (wherein the state is less than a full state of the decentralized hierarchy) [0027]).
Patel, Pabon, Banerjee, Crowe, Kumar and Goodman are analogous art because they are from the same field of endeavor with respect to child areas.
Before the effective filing date, for AIA , it would have been obvious to a person of ordinary skill in the art to incorporate the strategies by Goodman into the method by Patel, Pabon, Banerjee, Crowe and Kumar. The suggestion/motivation would have been to provide remote/ geographically distinct child areas that may appear as a pool of a parent area’s capacity (Goodman: [0018]).
Claim 15 is rejected under 35 U.S.C. 103 as being unpatentable over U.S. Patent Publication No. 2024/0236109 to Patel et al. (“Patel”) in view of U.S. Patent Publication No. 2025/0077295 to Pabon et al. (“Pabon”), U.S. Patent Publication No. 2023/0169356 to Banerjee et al. (“Banerjee”), U.S. Patent Publication No. 2017/0201423 to Crowe et al. (“Crowe”), U.S. Patent Publication No. 2022/0141099 to Kumar et al. (“Kumar”) and further in view of U.S. Patent Publication No. 2023/0244544 to Chen et al. (“Chen”).
As to claim 15, see similar rejection to claims 1-14 where the method is taught by the method.
As to claim 15, Patel, Pabon, Banerjee, Crowe and Kumar further disclose wherein the decentralized hierarchy comprises a plurality of clusters including a first cluster comprises edge devices and a second cluster comprising cloud devices (Kumar: fig 1-7, [0006-168]: ... the system is advantageous in situations where the network is comprised of clusters which are interconnected by different service providers (wherein the decentralized hierarchy comprises a plurality of clusters ...) and nodes that relay data transmitted between devices that are connected to the network may then also be interconnected by different organisations [0050] ... one or more of the nodes may be located in any one of the following locations within the network: a network edge (see with [0050] above - ... including a first cluster comprises edge devices), a telecommunication network or a cloud computer network (see with [0050] above - ... and a second cluster comprising cloud devices) [0030] ... new node for migrating workload that is in respect of an application can lie in any part of the network, such as on premise datacentres; or cloud infrastructure, e.g. private cloud, public cloud (see with [0050] above - ... and a second cluster comprising cloud devices) and edge cloud (see with [0050] above - ... including a first cluster comprises edge devices) [0049]).
For motivation, see rejection of claim 12.
Patel did not explicitly disclose wherein causing the workload to be migrated to the target comprises causing the workload to be migrated to the first cluster to the second cluster, or vice versa.
Chen discloses wherein causing the workload to be migrated to the target comprises causing the workload to be migrated to the first cluster to the second cluster, or vice versa (Chen: fig 1-8, [0007-]: ... federated operator component 101 may generate a connectivity map that defines a mesh network to enable communication between the various compute services deployed across different clusters ... may maintain a routing table that translates between virtual and physical addresses of the cache service deployed on far edge cluster 240 (wherein causing the workload to be migrated to the target comprises causing the workload to be migrated to the first cluster ...) and the persistent storage service deployed on regional/central cluster 250 (... to the second cluster) and since both the cache service and the persistent storage service may be part of the same compute service e.g., a content delivery network the federated operator component 101 may maintain and update routing tables to ensure reliable communication between different compute resources deployed across different clusters (wherein causing the workload to be migrated to the target comprises causing the workload to be migrated to the first cluster to the second cluster, or vice versa) ... the connectivity map generated by the federated operator component 101 may enable access control, circuit breaking, quality of service, etc ... may automatically generate the service mesh e.g., the connectivity map when a compute service e.g., a CDN is provisioned (wherein causing the workload to be migrated to the target comprises causing the workload to be migrated to the first cluster to the second cluster, or vice versa) and, additionally, the federated operator component 101 may maintain session affinity and routing table updates when individual compute resources/services on clusters are altered, re-balanced, and/ or transitioned from one node or cluster to another (wherein causing the workload to be migrated to the target comprises causing the workload to be migrated to the first cluster to the second cluster, or vice versa) [0036]).
Patel, Pabon, Banerjee, Crowe, Kumar and Chen are analogous art because they are from the same field of endeavor with respect to clusters.
Before the effective filing date, for AIA , it would have been obvious to a person of ordinary skill in the art to incorporate the strategies by Chen into the method by Patel, Pabon, Banerjee, Crowe and Kumar. The suggestion/motivation would have been to a new and innovative system, methods and apparatus for deployment and maintenance of compute services across multiple container orchestration clusters (Chen: [0003]).
Conclusion
The following prior art made of record and not relied upon is considered pertinent to applicant’s disclosure.
A) US 20260169823 – Yammada
A decentralized infrastructure enables devices to contribute compute, storage, and networking resources to a shared distributed environment. A registering device provides a registration request to one or more control devices, which determine the device's resource capabilities and direct creation of resource modules that expose those capabilities for use within the infrastructure. The control devices register the device in a pool of resource devices and receive service requests from other devices. A selected device is chosen based on current resource availability, provisioned with workloads, and instructed to execute those workloads and generate results, which are communicated back to the requesting device.
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 JUNE SISON whose telephone number is (571)270-5693. The examiner can normally be reached 9:00 am - 5:00 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, Emmanuel Moise can be reached at 571-272-3865. 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.
/JUNE SISON/Primary Examiner, Art Unit 2455
Prosecution Insights