KR20180058880A - Load balancing and performance optimizing methods of message queues in the distributed openstack cloud data centers - Google Patents

Load balancing and performance optimizing methods of message queues in the distributed openstack cloud data centers Download PDF

Info

Publication number
KR20180058880A
KR20180058880A KR1020160157100A KR20160157100A KR20180058880A KR 20180058880 A KR20180058880 A KR 20180058880A KR 1020160157100 A KR1020160157100 A KR 1020160157100A KR 20160157100 A KR20160157100 A KR 20160157100A KR 20180058880 A KR20180058880 A KR 20180058880A
Authority
KR
South Korea
Prior art keywords
message
data center
cloud data
message queue
distributed cloud
Prior art date
Application number
KR1020160157100A
Other languages
Korean (ko)
Inventor
장승욱
구경이
Original Assignee
(주)케이아이엔엑스
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by (주)케이아이엔엑스 filed Critical (주)케이아이엔엑스
Priority to KR1020160157100A priority Critical patent/KR20180058880A/en
Publication of KR20180058880A publication Critical patent/KR20180058880A/en

Links

Images

Classifications

    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/50—Queue scheduling
    • H—ELECTRICITY
    • H04—ELECTRIC COMMUNICATION TECHNIQUE
    • H04L—TRANSMISSION OF DIGITAL INFORMATION, e.g. TELEGRAPHIC COMMUNICATION
    • H04L47/00—Traffic control in data switching networks
    • H04L47/50—Queue scheduling
    • H04L47/62—Queue scheduling characterised by scheduling criteria

Landscapes

  • Engineering & Computer Science (AREA)
  • Computer Networks & Wireless Communication (AREA)
  • Signal Processing (AREA)
  • Computer And Data Communications (AREA)
  • Information Retrieval, Db Structures And Fs Structures Therefor (AREA)

Abstract

The present invention relates to a load balancing and performance optimizing method of a message queue in an OpenStack based distributed cloud data center, and more specifically, to a method for load balancing of a dynamic message queue between distributed cloud data centers and optimizing performance of an OpenStack service in an OpenStack based distributed cloud data center. To this end, the method comprises: a step (1) in which a message supplier designates a receiver A to send a message as shown in Fig. 3 so as to disperse a dynamic message queue load between distributed cloud data centers and to optimize performance of an OpenStack service by adding message distributers between the cloud data centers when a plurality of same OpenStack service nodes for each OpenStack based distributed cloud data center exist; a step (2) in which the message distributor collects a message queue for each data center and state information of the receiver A; a step (3) in which the message distributor compares and evaluates a processing volume of the message queue for each data center and the state information and a response speed of the receiver A; a step (4) in which the message distributor selects an optimized datacenter based on a result of comparison and evaluation conducted in the step (3); a step (5) in which entry a is generated by using database in the selected data center; and a step (6) of sending a message to the message queue of the selected data center.

Description

오픈스택 기반의 분산 클라우드 데이터센터 환경에서 메시지 큐 부하 분산 및 성능 최적화 방법 {Load balancing and performance optimizing methods of message queues in the distributed openstack cloud data centers}In this paper, we propose load balancing and performance optimizing methods for message queue load balancing and performance in an open stack based distributed data center environment.

본 발명은 분산 클라우드 데이터센터 환경에서 가상 자원 생성 과정에 있어 주요 오픈스택 서비스 노드간 효율적인 메시지 분배 방법에 관한 것이다.The present invention relates to an efficient message distribution method between major open stack service nodes in a virtual resource creation process in a distributed cloud data center environment.

일반적으로 대용량 분산 시스템의 메시지 분배 방식은 도 1에서와 같이 메시지 공급자가 데이터베이스에 데이터 목록을 게시하고 수신자와 게시된 목록 정보를 지정 후, 메시지 큐에 메시지를 보내면 지정된 수신자가 메시지 큐에서 메시지를 가져가서 게시된 목록 정보를 기반으로 데이터베이스의 수신된 메시지 관련 데이터 목록을 조회하는 방법을 사용하고 있다.In general, the message distribution method of a large-capacity distributed system is such that when a message provider publishes a list of data to a database, specifies a receiver and published list information, and then sends a message to the message queue, as shown in FIG. 1, And uses the method of inquiring the list of the received message related data of the database based on the posted list information.

그 일례로서 분산 클라우드 데이터센터 환경에서 오픈스택 개별 서비스 노드간 상호 작업과 상태 정보 교환을 위하여 메시지 큐를 사용하고 있다.  For example, in a distributed cloud data center environment, message queues are used for interworking and exchanging state information among open-stack individual service nodes.

그러나 분산된 데이터센터복수의 동일한 오픈스택 서비스 노드가 존재한다면 도 2에서와 같이 메시지 공급자는 복수의 데이터센터의 데이터베이스에 데이터 목록을 중복하여 게시하고 동일한 오픈스택 서비스 노드를 수신자로 지정하고 게시된 목록 정보를 추가한 후, 복수의 데이터센터의 메시지 큐에 중복하여 메시지를 보내야 하는 문제점이 있다. However, if there is a plurality of identical open-stack service nodes in the distributed data center, as shown in FIG. 2, the message provider can duplicate and publish the data list in the databases of the plurality of data centers, designate the same open- There is a problem that messages are repeatedly sent to message queues of a plurality of data centers after information is added.

[문헌1] US 20150295844 A1,Asynchronous framework for management of iaas (Mark Perreira, Bryan P. Murray, Rajeev Bharadhwaj, Stephane Herman Maes) 2012.12.3.[Patent Document 1] US 20150295844 A1, Asynchronous framework for management of iaas (Mark Perreira, Bryan P. Murray, Rajeev Bharadhwaj, Stephane Herman Maes)

[문헌1] Wai Kit Sze. Hardening OpenStack Cloud Platforms against Compute Node Compromises, ACM.ISBN 978-1-4503-4233-9/16/05.[Document 1] Wai Kit Sze. Hardening OpenStack Cloud Platforms against Compute Node Compromises, ACM.ISBN 978-1-4503-4233-9 / 16/05.

본 발명은 오픈스택 기반의 분산 클라우드 데이터센터 환경에서 분산 클라우드 데이터센터간 동적인 메시지 큐 부하 분산 및 오픈스택 서비스의 성능 최적화 방법을 구현하는데 그 목적이 있다.  It is an object of the present invention to implement dynamic message queue load balancing and performance optimization of open stack service among distributed cloud data centers in an open stack based distributed cloud data center environment.

본 발명은 오픈스택 기반의 분산된 클라우드 데이터센터별 복수의 동일한 오픈스택 서비스 노드가 존재할 경우 클라우드 데이터센터간 메시지 분배기 추가를 통하여 분산 클라우드 데이터센터간 동적인 메시지 큐 부하 분산 및 오픈스택 서비스의 성능 최적화를 위하여 도 3에서와 같이 메시지 공급자가 1. 수신자 A를 지정하여 메시지 보내기 단계, 메시지 분배기에서 2. 데이터센터별 메시지 큐 및 수신자 A의 상태 정보를 수집하는 단계, 메시지 분배기에서 3. 데이터센터별 메시지 큐의 처리 용량과 수신자 A의 상태정보 및 응답속도 비교평가하는 단계, 메시지 분배기에서 4. 3단계 비교평가의 결과로 최적의 데이터센터를 선택하는 단계, 5. 선택된 데이터센터의 데이터베이스로 entry a를 생성하는 단계, 6. 선택된 데이터센터의 메시지 큐로 메시지를 보내는 단계로 이루어진 것에 특징이 있다.  The present invention optimizes performance of dynamic message queue load balancing and open stack service between distributed cloud data centers by adding a message distributor between cloud data centers when there are a plurality of same open stack service nodes per open-stack distributed cloud data center As shown in FIG. 3, the message provider designates receiver A to send a message, in the message distributor, 2. collects state information of the data center-specific message queue and receiver A, in the message distributor, Comparing the processing capacity of the message queue with the state information of the receiver A and the response speed; evaluating in the message distributor; 4. selecting the optimal data center as a result of the comparison of the three steps; 6. Send the message to the message queue of the selected data center. It is characterized in comprising steps:

본 발명은 오픈스택 기반의 분산 클라우드 데이터센터 환경에서 데이터센터별 메시지 큐 및 오픈스택 서비스 노드별 처리 용량 및 성능 정보 기반으로 분산 클라우드 데이터센터간 동적인 메시지 큐 부하 분산 및 오픈스택 서비스의 성능 최적화를 구현할 수 있다.The present invention optimizes performance of dynamic message queue load balancing and open stack service between distributed cloud data centers based on processing capacity and performance information per message queue and open stack service node per data center in an open stack based distributed cloud data center environment Can be implemented.

도 1은 대용량 분산 시스템의 메시지 분배 단계도
도 2는 분산 클라우드데이터 센터 환경에서 메시지 분배 흐름도
도 3은 메시지 분배기 도입을 통한 분산 클라우드데이터 센터 환경에서 메시지 분배 최적화 흐름도
도 4는 오픈스택 기반의 분산 클라우드데이터 센터 환경에서 메시지 분배기 도입을 통한 클라우드 데이터 센터간 nova-compute 관련 메시지 분배 최적화 흐름도1 is a diagram of a message distribution step of a mass-
Figure 2 illustrates a message distribution flow diagram in a distributed cloud data center environment.
Figure 3 illustrates a message distribution optimization flow in a distributed cloud data center environment with the introduction of a message distributor
FIG. 4 is a flow chart illustrating nova-compute-related message distribution optimization among cloud data centers through the introduction of a message distributor in an open stack-based distributed cloud data center environment

본 발명을 첨부된 도면을 참조하여 상세히 설명하면 다음과 같다.    The present invention will now be described in detail with reference to the accompanying drawings.

도 4는 오픈스택 API cell을 통하여 메시지 분배기에서 1. server-creation(build) 메시지 전송하는 단계, 2. 지역별 메시지 큐 및 서버 자원의 상태 정보를 수집하는 단계, 3. 지역별 메시지 큐 및 서버 자원의 상태 정보 비교 후 최적의 지역으로 데이터센터 1 선택하는 단계, 선택된 데이터센터 1의 데이터베이스에 4. server instance entry a생성하는 단계, 데이터센터 1의 5. 지역 1의 nova-compute에게 server-creation(build) 메시지 보내는 단계, 데이터센터 1의 nova-compute에서 데이터센터 1의 지역 1 메시지 큐로부터 6. server-creation(build) 메시지 가져오는 단계, 데이터센터 1의 nova-compute에서 데이터센터 1의 데이터베이스에 7. server instance entry a 조회하는 단계를 통하여 오픈스택 기반의 분산 클라우드 데이터센터 환경에서 분산 클라우드 데이터센터간 동적인 메시지 큐 부하 분산 및 오픈스택 서비스의 성능 최적화를 구현하게 된다.  FIG. 4 shows a step of transmitting a 1. server-creation (build) message in a message distributor through an open stack API cell, 2. collecting state information of a message queue and a server resource by region, 3. collecting message queue and server resources Selecting the data center 1 as the optimal region after comparing the state information 4. Creating the server instance entry a in the database of the selected data center 1, 5. Creating a server-creation (build ) Message from the nova-compute of data center 1, 6. from the local 1 message queue of data center 1, 6. from the nova-compute of data center 1 to the database of data center 1 server instance entry a through dynamic query queue load balancing between distributed cloud data centers in an open-stack based distributed cloud data center environment, and It will implement performance optimization of open stack service.

Claims (1)

데이터센터별 메시지 큐 및 수신자 A의 상태 정보를 수집하는 단계, 메시지 분배기에서 데이터센터별 메시지 큐의 처리 용량과 수신자 A의 상태정보 및 응답속도 비교평가하는 단계, 메시지 분배기에서 3단계 비교평가의 결과로 최적의 데이터센터를 선택하는 단계를 통한 분산 클라우드 데이터센터간 동적인 메시지 큐 부하 분산 및 오픈스택 서비스의 성능 최적화 방법 Collecting state information of the data center-based message queue and receiver A, comparing the processing capacity of the message queue per data center in the message distributor with the state information and response speed of the receiver A, evaluating the results of the three-step comparison evaluation in the message distributor How to Optimize Performance of Dynamic Message Queue Load Balancing and Open Stack Services Between Distributed Cloud Data Centers by Choosing the Optimal Data Center
KR1020160157100A 2016-11-24 2016-11-24 Load balancing and performance optimizing methods of message queues in the distributed openstack cloud data centers KR20180058880A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
KR1020160157100A KR20180058880A (en) 2016-11-24 2016-11-24 Load balancing and performance optimizing methods of message queues in the distributed openstack cloud data centers

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
KR1020160157100A KR20180058880A (en) 2016-11-24 2016-11-24 Load balancing and performance optimizing methods of message queues in the distributed openstack cloud data centers

Publications (1)

Publication Number Publication Date
KR20180058880A true KR20180058880A (en) 2018-06-04

Family

ID=62628135

Family Applications (1)

Application Number Title Priority Date Filing Date
KR1020160157100A KR20180058880A (en) 2016-11-24 2016-11-24 Load balancing and performance optimizing methods of message queues in the distributed openstack cloud data centers

Country Status (1)

Country Link
KR (1) KR20180058880A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108848039A (en) * 2018-04-24 2018-11-20 平安科技（深圳）有限公司 The method and storage medium that server, message are distributed
CN109168054A (en) * 2018-09-10 2019-01-08 杭州联驱科技有限公司 Display screen play system and control method
CN111338821A (en) * 2020-02-25 2020-06-26 北京思特奇信息技术股份有限公司 Method, system and electronic equipment for realizing data load balance
KR20230010601A (en) 2021-07-12 2023-01-19 홍자영 Method for Providing Message-Based XR Mobile Services Using Augmented Reality and System Supporting It

Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150295844A1 (en) 2012-12-03 2015-10-15 Hewlett-Packard Development Company, L.P. Asynchronous framework for management of iaas

Patent Citations (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US20150295844A1 (en) 2012-12-03 2015-10-15 Hewlett-Packard Development Company, L.P. Asynchronous framework for management of iaas

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
[문헌1] Wai Kit Sze. Hardening OpenStack Cloud Platforms against Compute Node Compromises, ACM.ISBN 978-1-4503-4233-9/16/05.

Cited By (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108848039A (en) * 2018-04-24 2018-11-20 平安科技（深圳）有限公司 The method and storage medium that server, message are distributed
CN108848039B (en) * 2018-04-24 2021-11-02 平安科技（深圳）有限公司 Server, message distribution method and storage medium
CN109168054A (en) * 2018-09-10 2019-01-08 杭州联驱科技有限公司 Display screen play system and control method
CN111338821A (en) * 2020-02-25 2020-06-26 北京思特奇信息技术股份有限公司 Method, system and electronic equipment for realizing data load balance
KR20230010601A (en) 2021-07-12 2023-01-19 홍자영 Method for Providing Message-Based XR Mobile Services Using Augmented Reality and System Supporting It

Similar Documents

Publication Publication Date Title
CN109218355B (en) Load balancing engine, client, distributed computing system and load balancing method
US10733021B2 (en) System of cloud computing and method for detaching load in cloud computing system
Lu et al. Join-idle-queue: A novel load balancing algorithm for dynamically scalable web services
US9712402B2 (en) Method and apparatus for automated deployment of geographically distributed applications within a cloud
Daraghmi et al. A small world based overlay network for improving dynamic load-balancing
Wang et al. Towards network-aware service composition in the cloud
KR20180058880A (en) Load balancing and performance optimizing methods of message queues in the distributed openstack cloud data centers
CN105450618A (en) Operation method and operation system of big data process through API (Application Programming Interface) server
CN111954173B (en) Method, device, server and computer readable storage medium for sending short message
US20100058451A1 (en) Load balancing for services
JP2007529066A (en) Method and system for affinity management
CN102957624B (en) Content routing method and device
CN116708450A (en) Load balancing method, load balancing device, electronic equipment and computer readable storage medium
CN114036031B (en) Scheduling system and method for resource service application in enterprise digital middleboxes
Kumar et al. A K-means clustering based message forwarding model for Internet of Things (IoT)
CN111131333B (en) Business data pushing method and server cluster
WO2019167859A1 (en) Estimating device and estimating method
CN115883559A (en) Stateless network load balancing method, device and storage medium
CN109981726A (en) A kind of distribution method of memory node, server and system
Nasim et al. Mobile publish/subscribe system for intelligent transport systems over a cloud environment
CN108718259B (en) Message processing method and multi-core processor
Varshney et al. A novel approach of Load Balancing in Content Delivery Networks by optimizing the surrogate server
CN104092735A (en) Cloud computing data access method and system based on binary tree
CN111049929B (en) Virtual network resource service method, device, electronic equipment and storage medium
US20210185119A1 (en) A Decentralized Load-Balancing Method for Resource/Traffic Distribution