CN109542586B - Node resource state updating method and system - Google Patents
Node resource state updating method and system Download PDFInfo
- Publication number
- CN109542586B CN109542586B CN201811376424.2A CN201811376424A CN109542586B CN 109542586 B CN109542586 B CN 109542586B CN 201811376424 A CN201811376424 A CN 201811376424A CN 109542586 B CN109542586 B CN 109542586B
- Authority
- CN
- China
- Prior art keywords
- resource state
- node
- increment
- scheduling
- updating
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Active
Links
- 238000000034 method Methods 0.000 title claims abstract description 24
- 238000012163 sequencing technique Methods 0.000 claims abstract description 6
- 238000004364 calculation method Methods 0.000 claims description 6
- 238000013459 approach Methods 0.000 claims description 5
- 238000007726 management method Methods 0.000 description 4
- 238000004422 calculation algorithm Methods 0.000 description 2
- 230000009286 beneficial effect Effects 0.000 description 1
- 238000007405 data analysis Methods 0.000 description 1
- 230000007812 deficiency Effects 0.000 description 1
- 238000013461 design Methods 0.000 description 1
- 238000011161 development Methods 0.000 description 1
- 238000010586 diagram Methods 0.000 description 1
- 238000002955 isolation Methods 0.000 description 1
- 238000012423 maintenance Methods 0.000 description 1
- 238000012986 modification Methods 0.000 description 1
- 230000004048 modification Effects 0.000 description 1
- 238000013468 resource allocation Methods 0.000 description 1
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F9/00—Arrangements for program control, e.g. control units
- G06F9/06—Arrangements for program control, e.g. control units using stored programs, i.e. using an internal store of processing equipment to receive or retain programs
- G06F9/44—Arrangements for executing specific programs
- G06F9/455—Emulation; Interpretation; Software simulation, e.g. virtualisation or emulation of application or operating system execution engines
- G06F9/45533—Hypervisors; Virtual machine monitors
- G06F9/45558—Hypervisor-specific management and integration aspects
Landscapes
- Engineering & Computer Science (AREA)
- Software Systems (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention provides a method and a system for updating node resource states, which comprises the following steps: updating the node resource state; in the period of updating the node resource state, the node resource state is updated in an incremental mode; including calibration increments and compute node resource states. The step of updating the node resource state comprises the following steps: the scheduling main body regularly acquires the resource state of each node from the Kube-apiserver; sequencing the nodes according to the condition of the residual resources of the nodes; after the scheduling main body acquires the task of creating the Pod, selecting the node with the most residual resources from the sequenced nodes, and scheduling the Pod to be created to the node; and after the scheduling is finished, the scheduling main body reports the scheduling result to the Kube-apiserver. The computing node resource states include: every time scheduling is carried out, a value, namely increment K, is subtracted from the residual resource of the scheduled node recorded in the scheduling main body, so that the resource state of the node is close to the actual resource state.
Description
Technical Field
The invention relates to the technical field of computer resource management algorithms, in particular to a node resource state updating method and system.
Background
Since the birth of 2013, Docker immediately sees the great revolutionary significance brought by Docker by software providers, and accordingly Docker is drawn together in a dispute, and an ecosystem meeting various requirements is built for Docker. Therefore, Docker is rapidly developed and becomes one of the most fierce topics in the cloud computing field. Docker has the advantages of isolation, resource control and portability, and can provide great convenience for software development, deployment and maintenance, so that great attention is paid to the IT world since birth. The currently popular container cluster management tools are kubernets by Google and the distributed operating system CoreOS. Of the two, because kubernets have perfect copy management and access proxy functions, powerful development teams, and active open source communities, the basic concepts in kubernets are: pod, playback Controller, Service, Label, and Selector.
1) And Pod. A Pod is the smallest unit of a Kubernetes management Container, and multiple containers may be contained within one Pod.
2) Reproduction Controller (hereinafter referred to as RC). An RC is a group of Pods with identical properties, primarily to ensure that the number of copies of a Pod of that type conforms to a user-defined number at any time.
3) And Service. Service is an abstraction of services provided by kubernets to a set of Pod, through which a set of exposed ports of a Pod can be encapsulated as services provided to the outside.
4) Label and Selector: the use of Label and Selector is generally known in the preceding explanations of RC and Service: both the Label and the Selector are in the form of Key-Value, the Label being used to add a Label to an element, and the Selector being used to query elements containing certain labels.
The scheduling of the Pod is uniformly completed by Kube-scheduler and Kube-apiserver components, if a large number of requests for creating the Pod are generated in a node state updating period, because the node resource state is not updated in the period of time, the node list is not reordered, the Kube-scheduler schedules the pods to the same node, and the situations of excessive pods on the node and uneven resource allocation are caused. This problem can be inherently achieved by reducing the node resource status update period, but this in turn increases the number of calls to the Restful API, affecting scheduling time.
Disclosure of Invention
In order to overcome the deficiencies in the prior art, the present invention provides a method and a system for updating node resource status, so as to solve the technical problems.
The technical scheme of the invention is as follows:
a node resource state updating method comprises the following steps:
updating the node resource state;
in the period of updating the node resource state, the node resource state is updated in an incremental mode; including calibration increments and compute node resource states.
Further, the step of updating the node resource status includes:
the scheduling main body regularly acquires the resource state of each node from the Kube-apiserver;
sequencing the nodes according to the condition of the residual resources of the nodes;
after the scheduling main body acquires the task of creating the Pod, selecting the node with the most residual resources from the sequenced nodes, and scheduling the Pod to be created to the node;
and after the scheduling is finished, the scheduling main body reports the scheduling result to the Kube-apiserver.
Further, in the step of performing incremental update on the node resource state in the node resource state update period, calculating the node resource state includes:
every time scheduling is carried out, a value, namely increment K, is subtracted from the residual resource of the scheduled node recorded in the scheduling main body, so that the resource state of the node is close to the actual resource state.
Further, in the step of updating the node resource state in an incremental manner within the period of updating the node resource state, the calibration increment, that is, the calibration increment K, includes:
after the node resource state is updated every time, calibrating the value of the increment K according to the error between the node resource state and the resource state stored in the scheduling main body, which is acquired from the Kube-apiserver; so that the value of the increment K can be made to gradually approach the true value.
Further, in the step of performing incremental update on the node resource state in the node resource state update period, the calibration increment, that is, the calibration increment K, specifically includes:
incorporating the current increment K0And obtaining the increment K in the updating period through actual calculationrThe incremental calibration is set to a new increment K', and the process of calibrating the increment K is expressed by equation (1-1)
K'＝iK0+(1-i)Kr (1-1)
Wherein the value of the variable i affects the value of the calibrated increment: if i is larger, the new increment is closer to the original increment K0If i is smaller, the new increment is closer to the actual increment K in the updating periodr。
The technical scheme of the invention also provides a node resource state updating system, which comprises a node resource state updating module and an increment updating module;
the node resource state updating module is used for updating the node resource state;
and the incremental updating module is used for updating the node resource state in an incremental manner in the period of updating the node resource state by the node resource state updating module.
Further, the node resource state updating module comprises a scheduling main body and a Kube-apiserver unit;
the scheduling main body is used for regularly acquiring the resource state of each node from the Kube-apiserver and sequencing the nodes according to the residual resource condition of the nodes;
the scheduling main body is also used for selecting the node with the most residual resources from the sequenced nodes and scheduling the Pod to be created to the node;
and the scheduling main body is also used for reporting the scheduling result to the Kube-apiserver unit after the scheduling is finished.
Further, the increment updating module comprises a calibration increment unit and a computing node resource state unit;
and the calculating node resource state unit is used for subtracting a value, namely increment K, from the residual resource of the scheduled node recorded in the scheduling main body when scheduling is performed every time, so that the resource state of the node is close to the actual resource state.
And the calibration increment unit is used for calibrating the increment K, namely calibrating the value of the increment K according to the error between the node resource state acquired from the Kube-apiserver unit and the resource state stored in the scheduling main body after the node resource state is updated every time.
The Kube-apiserver is responsible for providing Restful API call, the Kube-scheduler is a scheduling main body, the scheduling main body works in the Kube-scheduler to complete the Kube-scheduler to periodically acquire the resource state of each node from the Kube-apiserver, and the nodes are sequenced according to the residual resource condition of the nodes; after the Kube-scheduler acquires the task of creating the Pod, selecting the node with the most residual resources from the sorted nodes, and scheduling the Pod to be created to the node; after scheduling is completed, the Kube-scheduler reports a scheduling result to the Kube-apiserver to complete the whole scheduling process, and two steps of calibrating increment K and calculating the node resource state are added on the basis of the original scheduling process, namely, the calculation of the node resource state means that each time scheduling is performed, a value, namely the increment K, is subtracted from the residual resource of the scheduled node recorded in the Kube-scheduler, so that the resource state of the node is close to the actual resource state; the calibration increment is to calibrate the value of the increment K according to an error between the actual resource state of the node (the resource state acquired from the Kube-api server) and the estimated resource state of the node (the resource state stored by the Kube-scheduler) after updating the resource state of the node each time, so that the value of the increment K gradually approaches to the true value.
According to the technical scheme, the invention has the following advantages: the scheme provides a node state incremental calculation algorithm, which is used for simulating the node resource state in real time in an incremental calculation mode in a node resource state information updating period and adjusting the sequence of a node list in time, so that the problem is avoided.
In addition, the invention has reliable design principle, simple structure and very wide application prospect.
Therefore, compared with the prior art, the invention has prominent substantive features and remarkable progress, and the beneficial effects of the implementation are also obvious.
Drawings
Fig. 1 is a schematic diagram illustrating a node resource status updating method.
Detailed Description
The cloud platform has the greatest advantage that the individual physical host resources are virtualized into the resource pool, and a user only needs to apply for the resources from the resource pool when using the cloud platform, and does not need to be concerned about which physical host the resources are specifically located on. However, as a platform provider, a policy for scheduling resources requested by a user needs to be considered, so that the resources of the physical host can be fully utilized as much as possible. An important criterion for evaluating the scheduling policy is the balance of resource scheduling, i.e. whether the resources requested by the user are scheduled to different physical hosts in a balanced manner. According to the node resource state updating method, the node resource states are calculated in real time, a large number of Pod are prevented from being dispatched to the same node in the node resource state updating period, the distribution balance of the pods is improved, and therefore resources of a physical host can be fully utilized.
In order to make the objects, technical solutions and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the drawings in the embodiments of the present application, and it is obvious that the described embodiments are some embodiments of the present application, but not all embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present application.
Example one
As shown in fig. 1, an embodiment of the present invention provides a method for updating a node resource status, including the following steps:
s1: updating the node resource state;
in this step, the process of updating the node resource state includes: the scheduling main body regularly acquires the resource state of each node from the Kube-apiserver and sequences the nodes according to the node residual resource condition; after the scheduling main body acquires the task of creating the Pod, selecting the node with the most residual resources from the sequenced nodes, and scheduling the Pod to be created to the node;
and after the scheduling is finished, the scheduling main body reports the scheduling result to the Kube-apiserver.
S2: in the period of updating the node resource state, the node resource state is updated in an incremental mode; including calibration increments and compute node resource states.
The method comprises the steps of adding two steps of calibrating increment K and calculating node resource state on the basis of the original scheduling process, namely, calculating the node resource state, namely, subtracting a value, namely the increment K, from the residual resource of the scheduled node recorded in the Kube-scheduler every time scheduling is carried out, so that the resource state of the node is close to the actual resource state; the calibration increment is to calibrate the value of the increment K according to an error between the actual resource state of the node (the resource state acquired from the Kube-api server) and the estimated resource state of the node (the resource state stored by the Kube-scheduler) after updating the resource state of the node each time, so that the value of the increment K gradually approaches to the true value.
Wherein, the resource state of the computing node is relatively simple and only needs to be in the original nodeAdding increment K to the point resource state as the scheduled node resource state, wherein the calibration increment K is relatively complex and needs to be combined with the current increment K0And obtaining the increment K in the updating period through actual calculationrThe increment is calibrated to a new increment K', and therefore the implementation of the calibration increment K is mainly described below.
The procedure for calibrating the increment K is expressed by equation (1-1)
K'＝iK0+(1-i)Kr (1-1)
Wherein the value of the variable i affects the value of the calibrated increment: if i is larger, the new increment is closer to the original increment K0If i is smaller, the new increment is closer to the actual increment K in the updating periodr。
Example two
The technical scheme of the invention also provides a node resource state updating system, which comprises a node resource state updating module and an increment updating module;
the node resource state updating module is used for updating the node resource state;
and the incremental updating module is used for updating the node resource state in an incremental manner in the period of updating the node resource state by the node resource state updating module.
The node resource state updating module comprises a scheduling main body and a Kube-apiserver unit;
the scheduling main body is used for regularly acquiring the resource state of each node from the Kube-apiserver and sequencing the nodes according to the residual resource condition of the nodes;
the scheduling main body is also used for selecting the node with the most residual resources from the sequenced nodes and scheduling the Pod to be created to the node;
and the scheduling main body is also used for reporting the scheduling result to the Kube-apiserver unit after the scheduling is finished.
The increment updating module comprises a calibration increment unit and a computing node resource state unit;
and the calculating node resource state unit is used for subtracting a value, namely increment K, from the residual resource of the scheduled node recorded in the scheduling main body when scheduling is performed every time, so that the resource state of the node is close to the actual resource state.
And the calibration increment unit is used for calibrating the increment K, namely calibrating the value of the increment K according to the error between the node resource state acquired from the Kube-apiserver unit and the resource state stored in the scheduling main body after the node resource state is updated every time.
The Kube-apiserver is responsible for providing Restful API call, the Kube-scheduler is a scheduling main body, the scheduling main body works in the Kube-scheduler to complete the Kube-scheduler to periodically acquire the resource state of each node from the Kube-apiserver, and the nodes are sequenced according to the residual resource condition of the nodes; after the Kube-scheduler acquires the task of creating the Pod, selecting the node with the most residual resources from the sorted nodes, and scheduling the Pod to be created to the node; after scheduling is completed, the Kube-scheduler reports a scheduling result to the Kube-apiserver to complete the whole scheduling process, and two steps of calibrating increment K and calculating the node resource state are added on the basis of the original scheduling process, namely, the calculation of the node resource state means that each time scheduling is performed, a value, namely the increment K, is subtracted from the residual resource of the scheduled node recorded in the Kube-scheduler, so that the resource state of the node is close to the actual resource state; the calibration increment is to calibrate the value of the increment K according to an error between the actual resource state of the node (the resource state acquired from the Kube-api server) and the estimated resource state of the node (the resource state stored by the Kube-scheduler) after updating the resource state of the node each time, so that the value of the increment K gradually approaches to the true value.
The node resource state updating method provided by the invention is based on data analysis, mainly when the number of created Pod is small, the container starting stage is the bottleneck of the whole process, when the number of created Pod is large, the container scheduling gradually becomes the bottleneck of the whole process, because the container starting time is limited by the Docker, the shortest time is about 2s, therefore, when the number of the start-up Pod is small, the time of about 2s becomes the bottleneck of the whole flow, whereas when the number of the start-up Pod is large, because the task of starting Pod can be distributed to different nodes and performed simultaneously, the time for starting container is not greatly increased, however, the current implementation of container scheduling in Kubernetes is single-thread sequential scheduling, and the scheduling time is in direct proportion to the number of the starting Pod, so when the number of the starting pods is large, the container scheduling becomes a bottleneck of the whole process. Starting from the container scheduling stage, the flow of creating Pod is optimized, and the deployment and expansion speed of the application is improved.
The terms "first," "second," "third," "fourth," and the like in the description and in the claims, as well as in the drawings, if any, are used for distinguishing between similar elements and not necessarily for describing a particular sequential or chronological order. It is to be understood that the data so used is interchangeable under appropriate circumstances such that the embodiments of the invention described herein are capable of operation in sequences other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," as well as any variations thereof, are intended to cover non-exclusive inclusions.
The previous description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.
Claims (2)
1. A node resource state updating method is characterized by comprising the following steps:
updating the node resource state;
in the period of updating the node resource state, the node resource state is updated in an incremental mode; the method comprises the steps of calibrating increment and calculating node resource state;
the step of updating the node resource state comprises the following steps:
the scheduling main body regularly acquires the resource state of each node from the Kube-apiserver;
sequencing the nodes according to the condition of the residual resources of the nodes;
after the scheduling main body acquires the task of creating the Pod, selecting the node with the most residual resources from the sequenced nodes, and scheduling the Pod to be created to the node;
after the scheduling is finished, the scheduling main body reports the scheduling result to the Kube-apiserver;
in the incremental updating of the node resource state in the node resource state updating period, the step of calculating the node resource state comprises the following steps:
each time scheduling is carried out, subtracting a value, namely increment K, from the residual resource of the scheduled node recorded in the scheduling main body to enable the resource state of the node to be close to the actual resource state;
in the updating period of the node resource state, in the incremental updating of the node resource state, the step of calibrating the increment, namely calibrating the increment K, comprises the following steps:
after the node resource state is updated every time, calibrating the value of the increment K according to the error between the node resource state and the resource state stored in the scheduling main body, which is acquired from the Kube-apiserver;
in the incremental updating of the node resource state in the node resource state updating period, the specific step of calibrating the increment, namely calibrating the increment K, comprises the following steps:
incorporating current deltas
And obtaining the increment in the updating period through actual calculation
Setting the delta calibration to the new delta
Over of the calibrated increment KThe equation is expressed by the formula (1-1)
Wherein, variable
The value of (c) affects the value of the calibrated increment: if it is not
If the value is larger, the new increment is closer to the original increment
If, if
The value is smaller, and the new increment is closer to the actual increment in the updating period
。
2. A node resource state updating system is characterized by comprising a node resource state updating module and an increment updating module;
the node resource state updating module is used for updating the node resource state;
the incremental updating module is used for updating the node resource state in an incremental manner in the period of updating the node resource state by the node resource state updating module;
the node resource state updating module comprises a scheduling main body and a Kube-apiserver unit;
the scheduling main body is used for regularly acquiring the resource state of each node from the Kube-apiserver and sequencing the nodes according to the residual resource condition of the nodes;
the scheduling main body is also used for selecting the node with the most residual resources from the sequenced nodes and scheduling the Pod to be created to the node;
the scheduling main body is also used for reporting the scheduling result to the Kube-apiserver unit after the scheduling is finished;
the increment updating module comprises a calibration increment unit and a computing node resource state unit;
a node resource state calculating unit, configured to subtract a value, namely an increment K, from the remaining resource of the scheduled node recorded in the scheduling main body every time scheduling is performed, so that the resource state of the node approaches an actual resource state;
and the calibration increment unit is used for calibrating the increment K, namely calibrating the value of the increment K according to the error between the node resource state acquired from the Kube-apiserver unit and the resource state stored in the scheduling main body after the node resource state is updated every time.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811376424.2A CN109542586B (en) | 2018-11-19 | 2018-11-19 | Node resource state updating method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811376424.2A CN109542586B (en) | 2018-11-19 | 2018-11-19 | Node resource state updating method and system |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN109542586A CN109542586A (en) | 2019-03-29 |
| CN109542586B true CN109542586B (en) | 2021-10-29 |
Family
ID=65848302
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811376424.2A Active CN109542586B (en) | 2018-11-19 | 2018-11-19 | Node resource state updating method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN109542586B (en) |
Families Citing this family (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN112860387A (en) * | 2019-11-27 | 2021-05-28 | 上海哔哩哔哩科技有限公司 | Distributed task scheduling method and device, computer equipment and storage medium |
| CN111459612A (en) * | 2020-03-27 | 2020-07-28 | 新浪网技术(中国)有限公司 | Updating method and device for Pod in Kubernetes system |
Family Cites Families (9)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US8793690B2 (en) * | 2009-10-09 | 2014-07-29 | International Business Machines Corporation | Generating timing sequence for activating resources linked through time dependency relationships |
| CN102043673B (en) * | 2009-10-21 | 2015-06-03 | Sap欧洲公司 | Calibration of resource allocation during parallel processing |
| CN103885836A (en) * | 2014-01-14 | 2014-06-25 | 无锡城市云计算中心有限公司 | Intelligent cloud computing data center resource scheduling method |
| CN103823714B (en) * | 2014-01-15 | 2017-05-24 | 浙江大学 | Virtualization-based method and device for adjusting QoS (quality of service) of node memory of NUMA (non uniform memory access architecture) |
| CN103986766B (en) * | 2014-05-19 | 2017-07-07 | 中国工商银行股份有限公司 | Adaptive load balancing job task dispatching method and device |
| CN104168332A (en) * | 2014-09-01 | 2014-11-26 | 广东电网公司信息中心 | Load balance and node state monitoring method in high performance computing |
| CN107179945A (en) * | 2017-03-31 | 2017-09-19 | 北京奇艺世纪科技有限公司 | A kind of resource allocation methods and device |
| CN107045455B (en) * | 2017-06-19 | 2019-06-11 | 华中科技大学 | A kind of Docker Swarm cluster resource method for optimizing scheduling based on load estimation |
| CN107908479B (en) * | 2017-12-11 | 2021-03-02 | 北京奇艺世纪科技有限公司 | Node resource allocation method and device |
-
2018
- 2018-11-19 CN CN201811376424.2A patent/CN109542586B/en active Active
Also Published As
| Publication number | Publication date |
|---|---|
| CN109542586A (en) | 2019-03-29 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US12069128B2 (en) | Automatic scaling of resource instance groups within compute clusters | |
| EP3270289B1 (en) | Container-based multi-tenant computing infrastructure | |
| US11204793B2 (en) | Determining an optimal computing environment for running an image | |
| US10496503B2 (en) | Healing cloud services during upgrades | |
| Morreale et al. | Software defined networking: Design and deployment | |
| US9722853B2 (en) | Endpoint data centers of different tenancy sets | |
| US9553782B2 (en) | Dynamically modifying quality of service levels for resources running in a networked computing environment | |
| US9430257B2 (en) | Scheduling virtual machines using user-defined rules | |
| US11740936B2 (en) | Method and system for managing cloud resources | |
| US20170249283A1 (en) | Event resolution as a dynamic service | |
| US20230259415A1 (en) | Dynamic management of network policies between microservices within a service mesh | |
| US10606489B2 (en) | Sidefiles for management of data written via a bus interface to a storage controller during consistent copying of data | |
| Arnold et al. | Workload orchestration and optimization for software defined environments | |
| CN109542586B (en) | Node resource state updating method and system | |
| Chang et al. | Write-aware replica placement for cloud computing | |
| US20160125488A1 (en) | Methods and systems to allocate physical network cost to tenants of a data center | |
| Alyas et al. | Resource Based Automatic Calibration System (RBACS) Using Kubernetes Framework. | |
| US20190005255A1 (en) | Protecting restricted information when importing and exporting resources | |
| US11570042B2 (en) | Software-defined network controller communication flow | |
| US20240168734A1 (en) | Identifying involvement of application services in a distributed application | |
| Bumgardner et al. | Constellation: A secure self-optimizing framework for genomic processing | |
| Hawilo | Elastic Highly Available Cloud Computing | |
| Liu | Optimizing Big Data Analytics Frameworks in Geographically Distributed Datacenters | |
| WO2024025538A1 (en) | Apparatus and method for providing resource management policy in telecommunications system | |
| Bairley et al. | An application aware approach to scalable online placement of data center workloads |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| GR01 | Patent grant | ||
| GR01 | Patent grant |