Detailed Action
The present application, filed on or after March 16, 2013, is being examined under the first inventor to file provisions of the AIA .
The Office Action is in response to claims filed on 4/15/2026 where claims 1-20 are pending and ready for examination.
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.
Claim 4-5 are objected to as being dependent upon a rejected base claim, but would be allowable if rewritten in independent form including all of the limitations of the base claim and any intervening claims.
Applicant's arguments filed 4/15/2026 have been fully considered but they are not persuasive.
The examiner has reviewed the Applicant’s argument submitted on 4/15/2026 in their entirety (Pages 7 – 11)
Applicant’s argument is not persuasive because it addresses Pang in isolation rather than the functionality provided by the combined solution of Li in view of Pang.
Li teaches a distributed database environment having an enabled query cache (Fg. 8A, “query cache is enabled”). Li further teaches query-based analytical tasks, execution of task via queries, and measurement of completed queries per minute. A query is a request for cache data, and a query cache services such query requests. Accordingly, Li teaches servicing cache requests.
Pang teaches redistributing cache data between containers by reading segments from old containers and packing the segments into new containers. Thus, Pang teaches redistribution of cached data from a first container to a second container.
The rejection does not rely upon Pang to teach servicing cache requests. Rather, Pang is relied upon for redistribution of cached data, while Li relied upon for the query cache functionality that services cache requests. The claim does not require the redistributed cache data itself be a request. Instead, the claim requires that redistributed cache data be provided to a second container that services cache requests associated with that redistributed data.
Accordingly, in the combined solution, Pang’s redistributed cached data is provided to the second container, and Li’s query cache requests associated with that redistributed cached data. Therefore the combined solution contemplates redistributing cached data from a first container to a second container to cause the second container to service cache requests associated with the redistributed data, as recited.
Applicant’s argument is not persuasive because it improperly conflates the redistributed cached data with the cache requests. Th claim does not require the redistributed cached data itself to constitute a cache request or to perform cache-request servicing. Rather, the claim recites redistributing cached data to a second container and causing the second container to service cache requests associated with that redistributed data.
Applicant improperly treats the redistributed data and the cache request as the same required disclosure. They are different claim components: the redistributed data is cached content moved to the second container, while the cache request is a request associated with that moved cache content.
As set forth in the rejection, Li teaches servicing cache request through its query-cache functionality. Li teaches an enabled query cache, query-based analytical tasks, execution via queries, and completed queries per minute. A query cache services query requests, and such query requests correspond to cache requests directed to cached data.
Pang teaches redistribution of cached data between containers by reading segments from old containers and packing the segments into new containers. Thus, Pang provides the redistributed cache data, while Li provides the cache request servicing functionality.
Applicant’s arguments with respect to Pang allegedly not disclosing anything related to servicing cache requests within the metes and bounds of the claim are not cogent. One of ordinary skill in the art would understand that a cache request necessarily has metadata associated with that request and and that such information is inherently present when servicing cache requests. Moreover, the metes and bound of the claim provide no constraint regarding the particular type of cache data, cache metadata, or other information that must be generated, maintained, or associated with a cache request. Accordingly, Applicant’s attempt to exclude such information form the scope of the claimed cache-request servicing is unsupported by the claim language.
Applicant’s argument that Pang does not disclose a database is not persuasive because the rejection does not rely upon Pang to teach the distributed database environment. As set forth in the rejection, Li teaches deployment of database containers within a database containerized database architecture and further teaches enabling query cache functionality for the distributed database system. Pang is relied upon for redistribution of cached data between containers based upon storage utilization considerations.
Applicant’s arguments improperly attacks Pang in isolation rather than addressing the combined teachings of Li in view of pang. The rejection does not require Pang, standing alone, to disclose a distributed database system because Li already provides that functionality. One of ordinary skill in the art would have readily applied Pang’s cache-redistribution techniques within Li’s distributed database cache environment to obtain the predictable benefit of managing cached data among containers based on utilization.
Furthermore the claims do not require that the redistribution functionality itself originate from a reference expressly directed to a distributed database. The relevant inquiry is whether the combined solution provides the claimed functionality. Here, Li provides the distributed database and query-cache environment, while Pang provides redistribution of cached data between containers. Accordingly, Applicant’s argument that Pang does not disclose a database does not identify a deficiency in the combined teachings relied upon by the rejection
MPEP 2141.01(a) provides that analogous art includes art that is reasonably pertinent to the problem faced by the inventor, even if the reference is not within the same field of endeavor as the claimed invention. Thus Pang need not itself be directed to a distributed database environment where Pang is relied upon for the reasonably pertinent cache-management problem of redistributing cached data among containers based on utilization/locality.
Furthermore, Applicant’s arguments are not commensurate with the scope of the claims. The claims does not recite the additional restrictions and distinctions asserted by Applicant. Arguments directed to features not appearing in the claims are not persuasive. See In re Self, 671 F.2d 1344, 1348 (CCPA 1982) (“Many of appellant’s arguments fail from the outset because they are not based on limitation appearing in the claims”)
Regarding claim 14, Applicant’s argument is not persuasive because it mischaracterizes the teachings of Kallikuri, As set forth in the rejection, Kallakuri teaches a container service that supports containers deployed within clusters and further teaches data/storage resources including database and database tables. Kallakuri teaches a distributed container environment through AWS container services and clusters of containers. Within that distributed container environment, Kallakuri teaches local cache functionality through the local container image cache associated with the container service/container agents. Accordingly, Kallakuri teaches containers implanting cache functionality in a distributed environment, and Applicant’s argument improperly isolates the local cache disclosure from the distributed container architecture expressly taught by Kallakuri. The claim does not require any particular type of cache or any specialized database ache architecture beyond the recited functionality. Therefore, Applicant’s arguments that Kallakuri fails to teach “a plurality of containers implementing a cache for the distributed cache for the distributed database system” is not persuasive and a mischaracterization.
The examiner notes for purposes of Appeal that Rao (US 20150212744 )expressly teaches a technological environment that a secondary cache may reside on other nodes of a distributed database cluster, confirming that the art recognized cache functionality implemented within distributed database cluster environments:
[0022] It is further noted, that the system and methods of FIG. 1 are provided by way of example. In another example, two or more secondary caches can be populated by a primary cache in a random access memory. In still another example, one secondary cache can be populated during an eviction stage of a primary cache and another secondary cache can be populated based on other metrics and/or triggers (e.g. based on metric and/or triggers that facilitate a `big` data computing process). It is also noted that the secondary cache can be remote and reside in other nodes of a distributed database cluster (e.g. infra). In some embodiments, system 100 can be implemented in a system with SSD cards in a server to layer virtualization methods. In some embodiments system 100 can be implemented in a system with a remote SSD appliance (e.g. can be remotely accessed via a computer network) that is outside of a server (with the CPU and primary cache) and a storage system (with the hard disk drive). Software in the server can implement the population of the secondary cache store in the remote SSD appliance. Accordingly, system 100 can be implemented in a central (e.g. monolithic) storage environment and/or distributed storage systems (local or remote) (e.g. see FIG. 7). In one example of a remote distributed storage system, the local CPU can view the remote secondary cache's SSD appliance as a backend storage.
PNG
media_image1.png
568
603
media_image1.png
Greyscale
Applicant’s arguments are not persuasive because they improperly attack the cited references individually and in a piecemeal fashion rather than addressing the teachings of the references as combined. The rejection relies upon the collection teachings of Kallakuri, Bask, and Jayaram and the functionality that their combined teachings provide to one of ordinary skill in the art. Non obviousness cannot be established by attacking references individually where the rejection is based on a combination of references. See In re Merck & Co., 80F.2d 109, 1097 (Fed. Cir. 1986).
Claim Rejections - 35 USC § 103
The following is a quotation of 35 U.S.C. 103 which forms the basis for all obviousness rejections set forth in this Office action:
A patent for a claimed invention may not be obtained, notwithstanding that the claimed invention is not identically disclosed as set forth in section 102, if the differences between the claimed invention and the prior art are such that the claimed invention as a whole would have been obvious before the effective filing date of the claimed invention to a person having ordinary skill in the art to which the claimed invention pertains. Patentability shall not be negated by the manner in which the invention was made.
Claims 1-2, 6 and 10 are rejected under 35 USC 103 as being unpatentable over Li et al, “Long Live the Image: On Enabling Resilient Production Database Containers for Microservice Applications” , August , 2024 in view of Pang (US 9,594,753)
Regarding claim 1, Li discloses a non-transitory computer-readable medium having program instructions stored thereon that are capable of causing a distributed computing system that includes a plurality of physical nodes implementing a hosting service to perform operations comprising:
deploying, to one or more of the physical nodes, a set of containers that implement a cache for a distributed database system hosted by the hosting service, wherein the set of containers include program instructions executable to store the cache in a memory internal to the one or more physical nodes (Li; Li teaches the utilization of containers spanned across a plurality of physical nodes to realize the deployment of a distributed database and cache system;
Section 3.1
“... after launching containers based on such a database image, we will be able to completely rely on the scheduler and orchestration tools (e.g., Kubernetes) to take care of the database containers at runtime ...”
Section 5.2.2
“When it comes to the deployment, as illustrated in Fig. 5, we employ Amazon’s Elastic Kubernetes Service (EKS) to enable a Kubernetes cluster for each of the microservices. The cluster is composed of one or more Amazon EC2 instances as nodes, each nodes has one or more pods depending on the deployment configurations ...”
See e.g. Fig. 5
Section 6.3
“... And to facilitate the demonstration, we mainly focus on enabling databasing index and query cache ... enabling query cache can bring an exponential improvement to the performance of both a single microservice and the multi-service system ...”
See Fig. 8A “Query cache is enabled”
PNG
media_image2.png
54
808
media_image2.png
Greyscale
PNG
media_image3.png
187
420
media_image3.png
Greyscale
PNG
media_image4.png
238
427
media_image4.png
Greyscale
The examiner notes conventional instructions are necessarily present within the containerized environment to perform said functions and/or features
);
Li does not expressly disclose:
determining a storage utilization for the set of containers; and
based on the determined storage utilization, redistributing data cached by a first of the containers to a second of the containers, to cause the second container to service cache requests associated with the redistributed data.
However in analogous art Pang discloses:
determining a storage utilization for the set of containers (Pang; Pang teaches the determination of a capacity of containers (i.e. a storage utilization);
see e.g. Column 76, Line 66 – Column 7, Line 4 : comparing the total capacity of containers associated with the group to the total size of group data stored in the containers. If the unloading locality exceeds a repair threshold (806), fragmentation repair is performed with respect to the group (808).); and
based on the determined storage utilization, redistributing data cached by a first of the containers to a second of the containers (Pang; Pang teaches if the comparative analysis surpasses a threshold a cache redistribution process between containers is executed;
Column 5, Lines 49 – 57: In various embodiments, the number of containers actually loaded is measured by counting the number of containers loaded or reloaded in a simulated cache when processing the Lx data stream in order and each data stream has its own cache. The simulated caches implement LRU (least recently used) policy. In various embodiments, the measured locality is compared to a detection threshold, e.g., a static threshold, to determine whether fragmentation repair should be performed with respect to the group.
Column 6, Lines 27 – 34: FIG. 6 is a flow chart illustrating an embodiment of a fragmentation repair process. In various embodiments, the process of FIG. 6 is used to implement step 508 of FIG. 5. In the example shown, the repair process (602) reads the segments in the group from the old containers (604) and packs them into new containers (606). The loading locality of the group will now become lower because reading its segments will load the new containers which are fewer in number than the old containers. The simulated cache is adjusted to replace the old containers loaded by the current group with the new containers (608) so that loading locality of future groups referencing segments in old/new containers can be correctly measured.
See e.g. Fig. 6).
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Pang’s cache redistribution scheme. The motivation being the combined solution provides for implementing a known technique resulting in increased efficiencies of placing cache data.
Li in view of Pang disclose:
based on the determined storage utilization, redistributing data cached by a first of the containers to a second of the containers, to cause the second container to service cache requests associated with the redistributed data (The combined solution as Li teaches a distributed database environment having an enabled query cache (Fg. 8A, “query cache is enabled”). Li further teaches query-based analytical tasks, execution of task via queries, and measurement of completed queries per minute. A query is a request for cache data, and a query cache services such query requests. Accordingly, Li teaches servicing cache requests. Pang teaches redistributing cache data between containers by reading segments from old containers and packing the segments into new containers. Thus, Pang teaches redistribution of cached data from a first container to a second container. Accordingly the combined solution provides one of ordinary skill in the art to realize second container service cache requests associated with the redistributed data)
Pang teaches redistributing cache data between containers by reading segments from old containers and packing the segments into new containers. Thus, Pang teaches redistribution of cached data from a first container to a second container.
.
Regarding claim 2, Li in view of Pang disclose the non-transitory computer-readable medium of claim 1, wherein the operations further comprise:
determining a first storage utilization for the first container and a second storage utilization for the second container, wherein the data is redistributed based on a difference between the first storage utilization and the second storage utilization (Pang’s “comparing the total capacity of containers ... to the total size of group data in the containers” constitutes determining storage utilization for containers by evaluating stored data size relative to container capacity. Because the comparison is performed with respect to “containers” associated with the group, the comparison necessarily yields storage utilization information for at least two distinct containers (i.e. a first container and a second container) among the containers associated with the group, thereby determining a first storage utilization for the first container and a second storage utilization for the second container. Redistribution is performed based on a difference in storage utilization between a source container (from which segments are read) and a destination container (into which segments are packed), i.e., based on the difference between the first and second storage utilizations.
Regarding claim 6, Li in view of Pang disclose the non-transitory computer-readable medium of claim 1, wherein the redistributed cached by the first of the containers to the second of the containers comprises a set of data extents, wherein the set of data extents comprises a set of data fragments ( The combined solution per Pang discloses redistributing cached data between containers using segments/fragments as the discrete units of data movement. The fragments are equivalent to the data extents).
Regarding claim 10, claim 10 comprises the same and/or similar subject matter as claim 1 and is considered an obvious variation; therefore it is rejected under the same rationale.
Claim 3 is rejected under 35 USC 103 as being unpatentable over Li in view of Pang and in further view of Boppanna (US 20250370813)
Regarding claim 3, Li in view of Pang disclose the non-transitory computer-readable medium of claim 2, Li does not expressly disclose wherein the operations further comprise:
determining an average utilization for the set of containers; and
determining if the first storage utilization exceeds the average utilization.
However in analogous art Boppanna discloses:
determining an average utilization for the set of containers (Boppanna;
see e.g. [0069] In embodiments of FIG. 4, the utilization calculator module 110 receives historical data (e.g., pod data and resource usage) from an external system (e.g., external container system) and calculates a
resource utilization metric based on the historical data. For example, the utilization calculator module 110 calculates the resource utilization metric RU for each pod based on an exponential weighted moving average (EWMA), static weights W1, W2, W3, probability of being in the 90.sup.th percentile P.sup.90th, and a simple moving average SMA.)
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Boppanna’s scheme. The motivation being the combined solution provides for implanting a known technique resulting in increased efficiencies of utilizing resources.
Li in view of Pang and in further view of Boppanna disclose:
determining if the first storage utilization exceeds the average utilization (The combined solution per Boppanna provides for a comparison scheme to be utilized based on the average utilization).
Claims 7 – 8 and 11 are rejected under 35 USC 103 as being unpatentable over Li in view of Pang and in further view of Kallakuri (US 12,443,427) and in further view of Boppanna (US 20250370813)
Regarding claim 7, Li in view of Pang disclose the non-transitory computer-readable medium of claim 1, Li does not expressly disclose wherein the operations further comprise:
selecting an auditor from the set of containers, wherein the auditor determines an average utilization for the set of containers.
However in analogous art Kallakuri discloses:
selecting an auditor from the set of containers, wherein the auditor determines an average utilization for the set of containers (Kallakuri; Kallakuri teaches a control plane selects an agent assigned to and residing at a particular container and where the agent may provide monitoring data with respects to its container (e.g. determining an average utilization for the set of containers
see e.g. Column 9, Lines 1-1 4: As described herein, a container agent 120A-120N can be a software module that runs on each compute instance 122 within a cluster and can send information to the control plane about the instance's current running tasks, resource utilization amounts (e.g., CPU, memory, networking), etc., and can start and stop tasks whenever it receives a request to do so from the control plane. In some examples,).
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Kallakuri’s control plane and container agent schemes. The motivation being the combined solution provides for increased efficiencies in provisioning and managing container resources.
As evidence of the rationale above Boppanna discloses:
average utilization for the set of containers (Boppanna;
see e.g. [0069] In embodiments of FIG. 4, the utilization calculator module 110 receives historical data (e.g., pod data and resource usage) from an external system (e.g., external container system) and calculates a
resource utilization metric based on the historical data. For example, the utilization calculator module 110 calculates the resource utilization metric RU for each pod based on an exponential weighted moving average (EWMA), static weights W1, W2, W3, probability of being in the 90.sup.th percentile P.sup.90th, and a simple moving average SMA.)
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Boppanna’s scheme. The motivation being the combined solution provides for implanting a known technique resulting in increased efficiencies of utilizing resources.
Regarding claim 8, Li in view of Pang discloses the non-transitory computer-readable medium of claim 1, Li does not expressly disclose wherein the operations further comprise:
determining whether an average utilization for the set of containers satisfies a threshold; and
in response to determining that the average utilization satisfies the threshold, adding additional containers to the set of containers.
However in analogous art Kallakuri discloses:
determining whether an average utilization for the set of containers satisfies a threshold (Kallakuri; Kallakuri teaches resource utilization (e.g. average utilization for the set of containers satisfies a threshold) may be calculated for subsequent determination of scaling container infrastructure which would require a threshold to be necessarily present;
see e.g. Column 5, Line 61 – Column 6, Line 12: As introduced earlier, a container service 102 (referred to in various implementations as a container service, cloud container service, container engine, or container cloud service), such as the Amazon Elastic Container Service (ECS) (TM), can be a highly scalable, high performance container management service that supports containers (e.g., Docker containers) and allows users to easily run applications on a managed cluster of compute instances, eliminating the need for users to install, operate, and scale their own cluster management infrastructure. With simple API calls, users can launch and stop container-enabled applications, query the complete state of their cluster”
See e.g. Column 9, Lines 1-14 : As described herein, a container agent 120A-120N can be a software module that runs on each compute instance 122 within a cluster and can send information to the control plane about the instance's current running tasks, resource utilization amounts (e.g., CPU, memory, networking), etc., and can start and stop tasks whenever it receives a request to do so from the control plane.; and
in response to determining that the average utilization satisfies the threshold, adding additional containers to the set of containers (Kallakuri; Kallakuri teaches the Amazon Elastic Container Service facilitates scaling containers based on resources (e.g. average utilization)
see e.g. Column 5, Line 61 – Column 6, Line 12: As introduced earlier, a container service 102 (referred to in various implementations as a container service, cloud container service, container engine, or container cloud service), such as the Amazon Elastic Container Service (ECS) (TM), can be a highly scalable, high performance container management service that supports containers (e.g., Docker containers) and allows users to easily run applications on a managed cluster of compute instances, eliminating the need for users to install, operate, and scale their own cluster management infrastructure. With simple API calls, users can launch and stop container-enabled applications, query the complete state of their cluster”)
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Kallakuri’s scheme. The motivation being the combined solution provides for implementing a known technique resulting in increased efficiencies of provisioning and managing container resources.
As evidence of the rationale above Boppanna discloses:
average utilization for the set of containers (Boppanna;
see e.g. [0069] In embodiments of FIG. 4, the utilization calculator module 110 receives historical data (e.g., pod data and resource usage) from an external system (e.g., external container system) and calculates a resource utilization metric based on the historical data. For example, the utilization calculator module 110 calculates the resource utilization metric RU for each pod based on an exponential weighted moving average (EWMA), static weights W1, W2, W3, probability of being in the 90.sup.th percentile P.sup.90th, and a simple moving average SMA.)
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Boppanna’s scheme. The motivation being the combined solution provides for implanting a known technique resulting in increased efficiencies of utilizing resources.
Regarding claim 11, Li in view of Pang disclose the non-transitory computer-readable medium of claim 10, Li does not expressly disclose wherein the operations further comprise:
sending, by the first container, a first storage utilization to a metadata server, wherein the received storage utilization is an average utilization determined based on the first storage utilization and a second storage utilization associated with the second container.
However in analogous art Kallakuri discloses:
sending, by the first container, a first storage utilization to a metadata server, wherein the received storage utilization is an average utilization determined based on the first storage utilization and a second storage utilization associated with the second container (Kallakuri; Kallakuri teaches a control plane server (i.e., metadata server) exchanges resource utilization with agent(s) associated with containers relating to utilization (i.e., average utilization)
see e.g. Column 4, Line 37 – Line 59: Generally, the traffic and operations of a provider network can broadly be subdivided into two categories: control plane operations carried over a logical control plane and data plane operations carried over a logical data plane. While the data plane represents the movement of user data through the distributed computing system, the control plane represents the movement of control signals through the distributed computing system. The control plane generally includes one or more control plane components distributed across and implemented by one or more control servers. Control plane traffic generally includes administrative operations, such as system configuration and management (e.g., resource placement, hardware capacity management, diagnostic monitoring, system state information.
see e.g. Column 9, Lines 1-1 4: As described herein, a container agent 120A-120N can be a software module that runs on each compute instance 122 within a cluster and can send information to the control plane about the instance's current running tasks, resource utilization amounts (e.g., CPU, memory, networking), etc., and can start and stop tasks whenever it receives a request to do so from the control plane
)
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Kallakuri’s scheme. The motivation being the combined solution provides for implementing a known technique resulting in increased efficiencies of provisioning and managing container resources.
As evidence of the rationale above Boppanna discloses:
average utilization for the set of containers (Boppanna;
see e.g. [0069] In embodiments of FIG. 4, the utilization calculator module 110 receives historical data (e.g., pod data and resource usage) from an external system (e.g., external container system) and calculates a
resource utilization metric based on the historical data. For example, the utilization calculator module 110 calculates the resource utilization metric RU for each pod based on an exponential weighted moving average (EWMA), static weights W1, W2, W3, probability of being in the 90.sup.th percentile P.sup.90th, and a simple moving average SMA.)
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Boppanna’s scheme. The motivation being the combined solution provides for implanting a known technique resulting in increased efficiencies of utilizing resources.
Claims 8 and 9 are rejected under 35 USC 103 as being unpatentable over Li in view of Pang and in further view of Shao (US 12,489, 716)
Regarding claim 9, Li in view of Pang disclose the non-transitory computer-readable medium of claim 1, wherein the operations further comprise:
determining whether an average utilization for the set of containers satisfies a threshold; and
in response to determining that the average utilization satisfies the threshold, removing one or more containers from the set of containers.
However in analogous art Shao discloses:
determining whether an average utilization for the set of containers satisfies a threshold (Shao;
see e.g. Column 5. Lines 3 -15: The HPA may be an example of a reactive autoscaler, or a tool that scales pods in reaction to real-time changes in service load. For example, the HPA may automatically increase or decrease a number of pod replicas in a container-based cluster based on comparing current resource utilization metrics to predefined thresholds and/or metrics. As such, the HPA (e.g., a reactive autoscaler) may help to procure a responsive and efficient system by closely monitoring the resource utilization of service(s) in real-time (or near real-time) and performing immediate scaling action(s), when necessary. Further, the HPA may enable available pod capacity to more closely track current service utilization requirements.
See e.g. Column 13, Lines 12-14: In certain embodiments, applying logic 646 includes logic for applying a smoothing filter to the resource utilization metrics to obtain smoothed resource utilization metrics.
See e.g. Column 22, Line 15 -29: Column In certain embodiments, adjusting logic 648 includes logic for adjusting each of the smoothed resource utilization metrics by a nominal value. In certain embodiments, adjusting logic 648 includes logic for automatically adjusting one or more configuration parameters for the container-based cluster to modify a state of the container-based cluster based on the future resource utilization determined for the service. In certain embodiments, adjusting logic 648 includes logic for adjusting at least one of: a number of pods to be deployed in the container-based cluster; or a number of nodes to be deployed in the container-based cluster. In certain embodiments, adjusting logic 648 includes logic for automatically adjusting the one or more configuration parameters based on the future resource utilization being above or below a threshold); and
in response to determining that the average utilization satisfies the threshold, removing one or more containers from the set of containers (Shao;
see e.g. Column 5. Lines 3 -15) The HPA may be an example of a reactive autoscaler, or a tool that scales pods in reaction to real-time changes in service load. For example, the HPA may automatically increase or decrease a number of pod replicas in a container-based cluster based on comparing current resource utilization metrics to predefined thresholds and/or metrics. As such, the HPA (e.g., a reactive autoscaler) may help to procure a responsive and efficient system by closely monitoring the resource utilization of service(s) in real-time (or near real-time) and performing immediate scaling action(s), when necessary. Further, the HPA may enable available pod capacity to more closely track current service utilization requirements.
See e.g. Column 13, Lines 12-14:In certain embodiments, applying logic 646 includes logic for applying a smoothing filter to the resource utilization metrics to obtain smoothed resource utilization metrics.
See e.g. Column 22, Line 15 -29: In certain embodiments, adjusting logic 648 includes logic for adjusting each of the smoothed resource utilization metrics by a nominal value. In certain embodiments, adjusting logic 648 includes logic for automatically adjusting one or more configuration parameters for the container-based cluster to modify a state of the container-based cluster based on the future resource utilization determined for the service. In certain embodiments, adjusting logic 648 includes logic for adjusting at least one of: a number of pods to be deployed in the container-based cluster; or a number of nodes to be deployed in the container-based cluster. In certain embodiments, adjusting logic 648 includes logic for automatically adjusting the one or more configuration parameters based on the future resource utilization being above or below a threshold).
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Shao’s scheme. The motivation being the combined invention provides for implementing a known technique resulting in increased efficiencies of provisioning and managing compute resources.
Regarding claim 8, claim 8 comprises the same and/or similar subject matter as claim 9 and is considered an obvious variation; therefore it is rejected under the same rationale.
Claim 12 is rejected under 35 USC 103 as being unpatentable over Li in view of Pang and in further view of Antinori (US 2025/0094310)
Regarding claim 12, Li in view of Pang disclose the non-transitory computer-readable medium of claim 10, wherein the operations further comprise:
reporting, by the first container, a request rate indicating a frequency the first container receives read and write requests, wherein the redistributing second data is further based on the request rate.
However in analogous art Antinori discloses:
reporting, by the first container, a request rate indicating a frequency the first container receives read and write requests (Antinori;
[0023] At block 204, the example processing device measures the one or more performance metrics. For example, the processing device 110 may determine that the performance metric 150 (e.g. the measured rate of request processing) for the target container environment 190 is five requests per second over a period of one minute, while a lower bound of the window of values 140 specifies a request processing rate of at least six requests per second over a period of one minute)
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Antinori’s scheme. The motivation being the combined solution provides for incorporating a known technique resulting in increased efficiencies of provisioning and managing containers.
Li in view of Pang and in further view of Antinori disclose:
reporting, by the first container, a request rate indicating a frequency the first container receives read and write requests, wherein the redistributing second data is further based on the request rate (The combined solution per Antinori provides for redistribution of data via Pang is based on the request rate).
Claim 13 is rejected under 35 USC 103 as being anticipated by Li in view of Pang and in further view of Xiao (US 20230195522) and in further view of Jayaram US 20240135161)
Regarding claim 13, Li in view of Pang disclose the non-transitory computer-readable medium of claim 10, Li does not expressly disclose wherein the operations 13. further comprise:
participating in an election to determine an auditor within a set of containers including the first and second containers, wherein the auditor provides an average storage utilization, wherein the average storage utilization is the received storage utilization.
However in analogous art Xiao discloses:
participating in an election to determine an auditor within a set of containers including the first and second containers (Xiao;
see e.g. [0153] In other implementations of the FIG. 13 embodiment, the VMs/container sets 1302 comprise respective containers implemented using virtualization infrastructure 1304 that provides operating system level virtualization functionality, such as support for Docker containers running on bare metal hosts, or Docker containers running on VMs. The containers are illustratively implemented using respective kernel control groups of the operating system. Such implementations can also provide multi-leader election functionality in a distributed computing system of the type described above. For example, a container host device supportingmultiple containers of one or more container sets can implement logic instances, data structures and/or other components for implementing multi-leader election functionality in the system 100.)
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Xiao’s elections scheme. The motivation being the combined solution provides for implementing a known technique which can provide the ability to elect conventional agents in distributed processing (e.g. containers). Thus increasing the overall efficiency of processing data.
Li in view of Pang and in further view of Xiao does not expressly disclose:
wherein the auditor provides an average storage utilization, wherein the average storage utilization is the received storage utilization.
However in analogous art Jayaram discloses:
wherein the auditor provides an average storage utilization, wherein the average storage utilization is the received storage utilization. (Jayaram;
see e.g. [0044] In order to build and update the historical infrastructure and utilization repository 122, the monitoring, collection and logging layer 121 of the data collection engine 120 extracts and collects parameters corresponding to resource infrastructure and resource utilization of various components of existing or previously deployed computing environments (e.g., existing or previously deployed private cloud environments). The parameters may be collected from the compute host devices 103, storage systems 105 and network systems 107 and/or from applications used for monitoring component metrics. The parameters comprise, for example, virtual instance types (e.g., VM, container, pod, etc.), virtual instance identifiers (e.g., VM ID, container ID, pod ID, etc.), compute quantity, compute size (e.g., number of CPU cores (millicores)), memory size (e.g., size of RAM), storage size (e.g., ephemeral storage size), time period (e.g., one or more timestamps identifying when (e.g., date, time) certain parameters were collected, CPU utilization, memory utilization, storage utilization
see e.g. [0053] As noted herein, historical infrastructure and utilization data is used for training the multi-target classification and regression models. FIG. 4 depicts example training data in an illustrative embodiment. As can be seen in the table 400, the training data identifies user/customer information, and the following data associated with each user/customer (“Cust.”): virtual instance types (“Instance Type”) (e.g., VM, container (Cont.), Mixed (combination of different virtual instances)), compute quantity, compute size (e.g., number of CPU cores (millicores)), storage size (e.g., ephemeral storage size (MiB)), average CPU utilization (Avg. CPU utilization (%)), average memory utilization (Avg. memory utilization (%)), average storage utilization (Avg. storage utilization (%)),
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Jayaram’s scheme. The motivation being the combined solution provides for implementing a known echnque resulting in increased efficiencies of provisioning and managing compute resources.
Claims 14, 16 and 18-19 are rejected under 35 USC 103 as being unpatentable over Kallakuri in view of Basak (US 20160349992) and in further view of Jayarman
Regarding claim 14, Kallakuri discloses a computer-implemented method, comprising:
receiving, via a metadata server, storage utilizations associated with a plurality of containers deployed to a plurality of physical nodes implementing a hosting service that hosts a distributed database system, wherein the plurality of containers implement a cache for the distributed database system (Kallakuri; Kallakuri teaches a control plane comprising at least one control server (i.e. metadata server) which exchanges (i.e. receives) storage utilization from agents and where the technological environment comprises a distributed database system and cache implanted within the context of containers associated with a Container Service (i.e. hosting service)
Column 5, Line 61 – Column 6, Line 12 “As introduced earlier, a container service 102 (referred to in various implementations as a container service, cloud container service, container engine, or container cloud service), such as the Amazon Elastic Container Service (ECS) (TM), can be a highly scalable, high performance container management service that supports containers (e.g., Docker containers) and allows users to easily run applications on a managed cluster of compute instances, eliminating the need for users to install, operate, and scale their own cluster management infrastructure. With simple API calls, users can launch and stop container-enabled applications, query the complete state of their clusters, and/or utilize provider network features such as virtual firewalls, load balancing, virtual block storage volumes, and/or Identity Access Management (IAM) roles. Users can use the container service 102 to schedule container placement across a cluster based on their unique resource needs and availability requirements or can integrate their own schedulers or third-party schedulers to meet business or application specific requirements.”
See e.g. Column 20, Lines 48 – 65: From an instance of the virtual computing system(s) 992 and/or another customer device 990 (e.g., via console 994), the customer can access the functionality of a storage service 910, for example via the one or more APIs 902, to access data from and store data to storage resources 918A-918N of a virtual data store 916 (e.g., a folder or “bucket,” a virtualized volume, a database, etc.) provided by the provider network 900. In some embodiments, a virtualized data store gateway (not shown) can be provided at the customer network 950 that can locally cache at least some data, for example frequently accessed or critical data, and that can communicate with the storage service 910 via one or more communications channels to upload new or modified data from a local cache so that the primary store of data (the virtualized data store 916) is maintained. In some embodiments, a user, via the virtual computing system 992 and/or another customer device 990, can mount and access virtual data store 916 volumes via the storage service 910 acting as a storage virtualization service, and these volumes can appear to the user as local (virtualized) storage 998.
See e.g. Column 3, Lines 3-23: FIG. 1 is a diagram illustrating an environment for optimizing the retrieval of container images by container instances running on computing infrastructure external to a cloud provider network using temporally staggered container image pull requests according to some examples. A provider network 100 (or, “cloud” provider network) provides users with the ability to use one or more of a variety of types of computing-related resources such as compute resources (e.g., executing virtual machine (VM) instances and/or containers, executing batch jobs, executing code without provisioning servers), data/storage resources (e.g., object storage, block-level storage, data archival storage, databases and database tables, etc.), network-related resources (e.g., configuring virtual networks including groups of compute resources, content delivery networks (CDNs), Domain Name Service (DNS)), application resources (e.g., databases, application build/deployment services), a
See e.g. Column 9, Line 1 -1 4: As described herein, a container agent 120A-120N can be a software module that runs on each compute instance 122 within a cluster and can send information to the control plane about the instance's current running tasks, resource utilization amounts (e.g., CPU, memory, networking), etc., and can start and stop tasks whenever it receives a request to do so from the control plane. In some examples, a container agent can optionally cache container images in a local container image cache 136. Thus, it can perform local actions under the control of the control plane 134 (e.g., responsive to commands sent by the control plane or configurations initiated from the control plane) and can report back metadata to the control plane 134 or a separate monitoring or logging service of the provider network 100.
See e.g. Column 4, Lines 37 – 59: Generally, the traffic and operations of a provider network can broadly be subdivided into two categories: control plane operations carried over a logical control plane and data plane operations carried over a logical data plane. While the data plane represents the movement of user data through the distributed computing system, the control plane represents the movement of control signals through the distributed computing system. The control plane generally includes one or more control plane components distributed across and implemented by one or more control servers. Control plane traffic generally includes administrative operations, such as system configuration and management (e.g., resource placement, hardware capacity management, diagnostic monitoring, system state information.
See e.g. Column 9, Line 1 -1 4: As described herein, a container agent 120A-120N can be a software module that runs on each compute instance 122 within a cluster and can send information to the control plane about the instance's current running tasks, resource utilization amounts (e.g., CPU, memory, networking), etc., and can start and stop tasks whenever it receives a request to do so from the control plane ... , it can perform local actions under the control of the control plane 134 (e.g., responsive to commands sent by the control plane or configurations initiated from the control plane) and can report back metadata to the control plane 134 or a separate monitoring or logging service of the provider network 100.
determining an average storage utilization for the plurality of containers, wherein the average storage utilization is based on the storage utilizations for the plurality of containers (Kallakuri; Kallakuri teaches the determination of storage utilization which one of ordinary art is readily able to perform conventional post processing of the data to yield average storage utilization;
See e.g. Column 9, Line 1 -1 4:As described herein, a container agent 120A-120N can be a software module that runs on each compute instance 122 within a cluster and can send information to the control plane about the instance's current running tasks, resource utilization amounts (e.g., CPU, memory, networking), etc., and can start and stop tasks whenever it receives a request to do so from the control plane
) and
providing a report to the metadata server, wherein the report includes the average storage utilization and wherein the report is accessible to the plurality of containers (Kallakuri; The agents report the utilization metrics to the control server (i.e metadata server) and subsequently the control server can exchange and/or share the information to other agents associated with containers; The agents can take commands from metadata server for configuration purposes)
See e.g. Column 9, Line 1 -1 4:As described herein, a container agent 120A-120N can be a software module that runs on each compute instance 122 within a cluster and can send information to the control plane about the instance's current running tasks, resource utilization amounts (e.g., CPU, memory, networking), etc., and can start and stop tasks whenever it receives a request to do so from the control plane ... , it can perform local actions under the control of the control plane 134 (e.g., responsive to commands sent by the control plane or configurations initiated from the control plane) and can report back metadata to the control plane 134 or a separate monitoring or logging service of the provider network 100.).
Kallakuri does not expressly disclose;
redistribute data among the plurality of containers
However in analogous art Basak discloses:
redistribute data among the plurality of containers j(Basak teaches the redistribution of workloads (e.g.) data based upon storage utlization;
see e.g. [0062] In some aspects, the regression module 120 can compute the utilization for workloads on all aggregates in the storage system (340). The utilizations of the aggregates can then be displayed to a storage server admin or user along with workload characteristics on the dashboard 140 (342). The admin can choose to provision new workloads 142 to minimally utilized aggregates or provision to aggregates where the dominating workload has similar characteristics to the workload to be provisioned, for example. In other aspects, the provisioning advisor 105 can use the utilization data and workload characteristics to redistribute workloads between nodes in a cluster in order to maximize performance (344).
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Basak’s redistribution scheme. The motivation being the combined solution provides for increased efficiencies of managing data.
As evidence of the rationale above Jayaram discloses:
Average storage utilization: ((Jayaram;
see e.g. [0053] As noted herein, historical infrastructure and utilization data is used for training the multi-target classification and regression models. FIG. 4 depicts example training data in an illustrative embodiment. As can be seen in the table 400, the training data identifies user/customer information, and the following data associated with each user/customer (“Cust.”): virtual instance types (“Instance Type”) (e.g., VM, container (Cont.), Mixed (combination of different virtual instances)), compute quantity, compute size (e.g., number of CPU cores (millicores)), storage size (e.g., ephemeral storage size (MiB)), average CPU utilization (Avg. CPU utilization (%)), average memory utilization (Avg. memory utilization (%)), average storage utilization (Avg. storage utilization (%)),
see e.g. [0044] In order to build and update the historical infrastructure and utilization repository 122, the monitoring, collection and logging layer 121 of the data collection engine 120 extracts and collects parameters corresponding to resource infrastructure and resource utilization of various components of existing or previously deployed computing environments (e.g., existing or previously deployed private cloud environments). The parameters may be collected from the compute host devices 103, storage systems 105 and network systems 107 and/or from applications used for monitoring component metrics. The parameters comprise, for example, virtual instance types (e.g., VM, container, pod, etc.), virtual instance identifiers (e.g., VM ID, container ID, pod ID, etc.), compute quantity, compute size (e.g., number of CPU cores (millicores)), memory size (e.g., size of RAM), storage size (e.g., ephemeral storage size), time period (e.g., one or more timestamps identifying when (e.g., date, time) certain parameters were collected, CPU utilization, memory utilization, storage utilization)
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Jayaram’s scheme. The motivation being the combined solution provides for implementing a known technique resulting in increased efficiencies of provisioning and managing compute resources.
Regarding claim 16, Kalluri in view of Basak and in further view of Jayaram disclose the computer-implemented method of claim 14, wherein the computer-implemented 16. method is performed by one of the plurality of containers (Per Independent claim 14 the agents may collect and calculate metrics)
Regarding claim 18, Kalluri in view of Basak and in further view of Jayaram disclose The computer-implemented method of claim 14, Kalluri does not expressly disclose wherein the average storage utilization is based on the data cached by the plurality of containers.
However the examiner take official notice that this feature is well known and conventional in the art and it would have been obvious to one of ordinary skill in the art to implement this feature in order to optimize container and cache resources.
Regarding claim 19, Kalluri in view of Basak and in further view of Jayaram disclose the computer-implemented method of claim 14, further comprising:
determining whether the average storage utilization for the plurality of containers satisfies a threshold (The combined solution per Jayarma see e.g. [0046], [0053], [0044] ); and
in response to determining that the average storage utilization satisfies the threshold, changing the number of containers within the plurality of containers (The combined solution per Jayadram and Kallatur as detailed in indepent claim 14 provides for a scalable platform based on performance).
Claim 15 is rejected under 35 USC 103 as being unpatentable over , Kallakuri in view of Basak and in further view of Jayaram and in further view of Takahashi (US 20230185924)
Regarding claim 15, Kallakuri in view of Basak and in further view of Jayaram disclose The computer-implemented method of claim 14, further comprising, Kallakuri does not expressly disclose determining an average access frequency associated with the plurality of containers, wherein the report includes the average access frequency.
However in analgous art Takahashi discloses:
access frequency associated with the plurality of containers (Takahashi; [[045]
The access frequency factor calculation part 412 calculates an access frequency factor based on access frequency information (recorded in an access frequency management DB 140 to be described later), the access frequency factor being an evaluation value of a vulnerability deriving from a form of communication performed by each container 11
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Takahashi scheme. The motivation being the combined solution provides for increased efficiencies of managing containers.
Kallakuri in view of Basak and in further view of Jayaram and in further vie of Takahashi disclose:
determining an average access frequency associated with the plurality of containers, wherein the report includes the average access frequency. (The combined solution provides for using conventional means to calculate an average acess frequency associated with containers for reporting purposes)
Claim 17 is rejected under 35 USC 103 as being unpatenable over Kallaturi in view of Basak and in further view of Jayaram and in further vie of Xiao
Regarding claim 17, Kallaturi in view of Basak and in further view of Jayaram The computer-implemented method of claim 16, Kallakuri does not expressly disclose wherein the one container is elected. from among the plurality of containers.
However in analogous art Xiao discloses:
container is elected. from among the plurality of containers.(Xiao;
see e.g. [0153] In other implementations of the FIG. 13 embodiment, the VMs/container sets 1302 comprise respective containers implemented using virtualization infrastructure 1304 that provides operating system level virtualization functionality, such as support for Docker containers running on bare metal hosts, or Docker containers running on VMs. The containers are illustratively implemented using respective kernel control groups of the operating system. Such implementations can also provide multi-leader election functionality in a distributed computing system of the type described above. For example, a container host device supportingmultiple containers of one or more container sets can implement logic instances, data structures and/or other components for implementing multi-leader election functionality in the system 100.)
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Xiao’s elections scheme. The motivation being the combined solution provides for implementing a known technique which can provide the ability to elect conventional agents in distributed processing (e.g. containers). Thus increasing the overall efficiency of processing data.
Claim 20 is rejected under 35 USC 103 as being unpatentable over Kallaturi in view of Basak and in further view of Jayaram and in further view of Mufti (US 2023/0236939)
Regarding claim 20, Kallaturi in view of Basak and in further view of Jayaram disclose the computer-implemented method of claim 14,Kallaturi does not expressly disclose wherein a first container within the 20. plurality of containers is deployed to a first physical node located in a first area zone and a second container within the plurality of containers is deployed to a second physical node located in a second area zone, wherein the first area zone and the second area zone are in different geographical locations.
Mufiti discloses:
plurality of containers is deployed to a first physical node located in a first area zone and a second container within the plurality of containers is deployed to a second physical node located in a second area zone, wherein the first area zone and the second area zone are in different geographical locations.(Mufti;
[0278] For example, a recovery policy may, for a given one or more container system resources, indicate conditions or parameters to be satisfied by backup data. In this example, a recovery policy may indicate one or more of: a specified geographic region for recovery data, that recovery data be within a fault zone that is different than resources being backed up, a range of financial costs, performance characteristics of a computing environment comprising recovery data, reliability characteristics of a computing environment comprising recovery data, communication latency from a specified location to a computing environment comprising recovery data, that recovery data be at least a minimum distance away from a geographical region associated with container system resources (e.g., cluster, containerized application, user data) being backed up, among other parameters used as a basis for the controller 502 to determine a computing environment used for backing up one or more container system resources associated with the recovery policy.)
Therefore it would have been prima facie obvious to one of ordinary skill in the art before the effective filing date of the claimed invention to incorporate Mufti’ geographical scheme. The motivation being the combined solution provides for implementing a known technique resulting in increased efficiencies of data reliability and integrity.
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 TODD L. BARKER whose telephone number is (571) 270 0257. The Examiner can normally be reached on Monday through Friday, 7:30am to 5:00pm.
If attempts to reach the Examiner by telephone are unsuccessful, the Examiner's supervisor Vivek Srivastava can be reached on (571) 272 7304.
/TODD L BARKER/Primary Examiner, Art Unit 2449
Prosecution Insights