IP Library › Granted Patent US 12,405,817
Granted Patent B2
US 12,405,817 · App. 17/926,936 · Granted Sep 2, 2025

Container resource designing device, container resource designing method, and program

Inventors: Masazumi Ota (Musashino, JP); Makoto Hamada (Musashino, JP)
Assignee: Nippon Telegraph and Telephone Corporation
G06F9/45558G06F2009/45575G06F2009/45591
View Patent ↗
Loading inventors, assignments & file history…
Monitor This Case
Get email alerts when status or documents change.
Order Certified Copies
Most orders are placed with the USPTO same day — all within 24 business hours.
Order via The Patent Place →
Pre-filled with this patent's details
Quick Facts
Patent No.
US 12,405,817
App. No.
17/926,936
Granted
Sep 2, 2025
Kind
B2
Abstract

A master node, which is a container resource design device, includes: a container setting reception unit configured to acquire definition information including a replica set count indicating a number of redundancies of a pod, an affinity rule indicating a condition for starting up containers, and a failure tolerance count; a resource calculation unit configured to calculate a required worker node count while satisfying the replica set count and satisfying the condition defined in the affinity rule, even if a failure of the failure tolerance count or less has occurred in a worker node, and transmit resource dispensation instruction information to a resource management mechanism; and a container setting unit configured to determine placement of the pods on dispensed worker nodes and transmit container setting information.

Claims (26)

1. A container resource design device for performing resource design for a virtualized container system,

wherein the container system comprises a plurality of worker nodes configured to provide a service due to containers being placed thereon, and the container resource design device for performing generation and operation of the worker nodes, and

the container resource design device comprises a processor configured to perform operations comprising:

acquiring definition information including a replica set count indicating a number of redundancies of a pod in units of pods, a pod being a set of one or more of the containers, an affinity rule indicating a condition for starting up the containers, and a failure tolerance count indicating a number of the worker nodes for which occurrence of a failure is allowed in order to maintain a state in which self-healing is possible;

extracting the replica set count, the affinity rule, and the failure tolerance count from the definition information;

calculating a required worker node count indicating a minimum required number of the worker nodes while satisfying the replica set count and satisfying the condition defined by the affinity rule, even if a failure of the failure tolerance count or less has occurred in the worker nodes;

dispensing the worker nodes of the required worker node count as resources by transmitting resource dispensation instruction information, which is information instructing dispensation of resources, to a resource management mechanism, the resource dispensation instruction information including the calculated required worker node count;

determining placement of the pods of the number indicated by the replica set count on the dispensed worker nodes; and

transmitting container setting information instructing setting of containers constituting the pods to the worker nodes for which the placement of the pods was determined, wherein in response to determining a label indicating a type of a resource for starting up the containers is attached to the affinity rule, the processor distinguishes the type of the resource indicated by the label, calculates a required label node count indicating the minimum required number of the worker nodes in the resource of the distinguished type, and transmits the required label node count to the resource management mechanism in a state of being included in the resource dispensation instruction information.

2. The container resource design device according to claim 1 ,

wherein acquiring the definition information comprises acquiring a plurality of pieces of the definition information related to respectively different pods, and

calculating the required worker node count comprises calculating the required worker node count that is the minimum required overall while satisfying the replica set count and satisfying the condition defined by the affinity rule even if a failure of the failure tolerance count or less has occurred in the worker nodes, each pod being indicated by the respective plurality of pieces of definition information.

3. A container resource design method for a container resource design device for performing resource design for a virtualized container system,

wherein the container system comprises a plurality of worker nodes configured to provide a service due to containers being placed thereon, and the container resource design device for performing generation and operation of the worker nodes, and

the container resource design method comprises:

acquiring definition information including a replica set count indicating a number of redundancies of a pod in units of pods, a pod being a set of one or more of the containers, an affinity rule indicating a condition for starting up the container, and a failure tolerance count indicating a number of the worker nodes for which occurrence of a failure is allowed in order to maintain a state in which self-healing is possible;

extracting the replica set count, the affinity rule, and the failure tolerance count from the definition information, and calculating a required worker node count indicating a minimum required number of the worker nodes while satisfying the replica set count and satisfying the condition defined in the affinity rule, even if a failure of the failure tolerance count or less has occurred in the worker nodes;

dispensing the worker nodes of the required worker node count as resources by transmitting resource dispensation instruction information, which is information instructing dispensation of resources, to a resource management mechanism, the resource dispensation instruction information including the calculated required worker node count; and

determining placement of the pods of the number indicated by the replica set count on the dispensed worker nodes, and transmitting container setting information instructing setting of containers constituting the pods to the worker nodes for which the placement of the pods was determined, wherein in response to determining a label indicating a type of a resource for starting up the containers is attached to the affinity rule, a processor distinguishes the type of the resource indicated by the label, calculates a required label node count indicating the minimum required number of the worker nodes in the resource of the distinguished type, and transmits the required label node count to the resource management mechanism in a state of being included in the resource dispensation instruction information.

4. A non-transitory computer-readable medium storing program instructions that cause a computer to perform operations comprising:

acquiring definition information including a replica set count indicating a number of redundancies of a pod in units of pods, a pod being a set of one or more of containers, an affinity rule indicating a condition for starting up the containers, and a failure tolerance count indicating a number of worker nodes for which occurrence of a failure is allowed in order to maintain a state in which self-healing is possible;

extracting the replica set count, the affinity rule, and the failure tolerance count from the definition information;

calculating a required worker node count indicating a minimum required number of the worker nodes while satisfying the replica set count and satisfying the condition defined by the affinity rule, even if a failure of the failure tolerance count or less has occurred in the worker nodes;

dispensing the worker nodes of the required worker node count as resources by transmitting resource dispensation instruction information, which is information instructing dispensation of resources, to a resource management mechanism, the resource dispensation instruction information including the calculated required worker node count;

determining placement of the pods of the number indicated by the replica set count on the dispensed worker nodes; and

transmitting container setting information instructing setting of containers constituting the pods to the worker nodes for which the placement of the pods was determined, wherein in response to determining a label indicating a type of a resource for starting up the containers is attached to the affinity rule, a processor distinguishes the type of the resource indicated by the label, calculates a required label node count indicating the minimum required number of the worker nodes in the resource of the distinguished type, and transmits the required label node count to the resource management mechanism in a state of being included in the resource dispensation instruction information.

Assignments (2)
CHANGE OF NAME Recorded Aug 14, 2025
From: NIPPON TELEGRAPH AND TELEPHONE CORPORATION
To: NTT, INC.
Reel/Frame 072412/0498 →
ASSIGNMENT OF ASSIGNOR'S INTEREST Recorded Nov 21, 2022
From: OTA, MASAZUMI; HAMADA, MAKOTO
To: NIPPON TELEGRAPH AND TELEPHONE CORPORATION
Reel/Frame 061845/0230 →
Continuity (1)
Related Publication 20230195497A1 · Jun 22, 2023
References Cited (34)
US 10191778B1 · Yang · 2019 [cited by examiner]
US 10601740B1 · Harding · 2020 [cited by examiner]
US 11005721B1 · Patki · 2021 [cited by examiner]
US 11134098B1 · Lieberman · 2021 [cited by examiner]
US 11171950B1 · Zhuravlev · 2021 [cited by examiner]
US 11422844B1 · Filiz · 2022 [cited by examiner]
US 11853807B1 · Coult · 2023 [cited by examiner]
US 20070180314A1 · Kawashima et al. · 2007 [cited by applicant]
US 20160205518A1 · Patel · 2016 [cited by examiner]
US 20190235775A1 · Zwiegincew · 2019 [cited by examiner]
US 20190379590A1 · Rimar · 2019 [cited by examiner]
US 20200133731A1 · Barbara · 2020 [cited by examiner]
US 20200136987A1 · Nakfour · 2020 [cited by examiner]
US 20200250006A1 · Parekh · 2020 [cited by examiner]
US 20200403872A1 · Shivashankara · 2020 [cited by examiner]
US 20210117868A1 · Sriharsha · 2021 [cited by examiner]
US 20210303290A1 · Schmit · 2021 [cited by examiner]
US 20210326763A1 · Bernat · 2021 [cited by examiner]
US 20210349767A1 · Asayag · 2021 [cited by examiner]
US 20220147391A1 · Balcha · 2022 [cited by examiner]
US 20220156102A1 · Bandarupalli · 2022 [cited by examiner]
US 20220156129A1 · Li · 2022 [cited by examiner]
US 20220253335A1 · Bequet · 2022 [cited by examiner]
US 20220329499A1 · Smith · 2022 [cited by examiner]
US 20230094937A1 · Arndt · 2023 [cited by examiner]
US 20230138344A1 · Bequet · 2023 [cited by examiner]
CN 109117265A · 2019 [cited by examiner]
CN 110300130A · 2019 [cited by examiner]
CN 111694633A · 2020 [cited by examiner]
JP 2007207219 · 2007 [cited by applicant]
Jeremy Casas, Mist: PVM with Transparent Migration and Checkpointing. (Year: 1995). [cited by examiner]
ID Wilson, Container stowage planning: a methodology for generating computerised solutions. (Year: 2000). [cited by examiner]
[No Author Listed] [online], “Start Nginx with Kubernetes Manifest file,” Leaf of Thought, Apr. 16, 2019, retrieved from URL <https://noumenon-th.net/programming/2019/04/16/manifest01>, 9 pages (with English Translation… [cited by applicant]
Nishijima, “Running Kubernetes manifests on Magnum,” Think IT, Feb. 12, 2016, retrieved from URL <https://thinkit.co.jp/article/9378>, 33 pages (with English Translation). [cited by applicant]