GB/T 51399-2019 English PDFUS$959.00 · In stock
Delivery: <= 7 days. True-PDF full-copy in English will be manually translated and delivered via email. GB/T 51399-2019: (Cloud computing infrastructure engineering technical standards) Status: Valid
Basic dataStandard ID: GB/T 51399-2019 (GB/T51399-2019)Description (Translated English): (Cloud computing infrastructure engineering technical standards) Sector / Industry: National Standard (Recommended) Classification of Chinese Standard: P07 Classification of International Standard: 91.010.01 Word Count Estimation: 47,439 Date of Issue: 2019 Date of Implementation: 2020-06-01 Issuing agency(ies): Ministry of Housing and Urban-Rural Development of the People's Republic of China; State Administration for Market Regulation GB/T 51399-2019: (Cloud computing infrastructure engineering technical standards)---This is a DRAFT version for illustration, not a final translation. Full copy of true-PDF in English version (including equations, symbols, images, flow-chart, tables, and figures etc.) will be manually/carefully translated upon your order.1 General 1.0.1 This standard is formulated in order to unify and standardize the construction of cloud computing infrastructure projects in various industries, and make the construction of cloud computing infrastructure projects technologically advanced, safe and reliable, economically reasonable, energy-saving and environmentally friendly. 1.0.2 This standard applies to the construction of new construction, expansion and reconstruction of cloud computing infrastructure. 1.0.3 Project construction should comply with the relevant national requirements on cloud computing infrastructure, implement the national basic construction guidelines and policies, and adhere to the scientificity, rationality and fairness of project construction. 1.0.4 Project construction should select materials and equipment that have issued qualified inspection reports and meet the relevant national technical requirements. 1.0.5 When constructing telecommunication networks in areas with seismic fortification intensity of 7 degrees and above, the requirements for seismic fortification shall be met, and the main telecommunication equipment used shall comply with the provisions of the current industry standard "Code for Seismic Performance Testing of Telecommunications Equipment" YD 5083. 1.0.6 The project construction should be compared with multiple schemes to improve economic efficiency and reduce project cost on the basis of meeting infrastructure quality and safety requirements. 1.0.7 The project construction should follow the principle of openness, and the system should meet the standardization requirements and have the ability to be expanded online. 1.0.8 In addition to complying with the provisions of this standard, the construction of cloud computing infrastructure projects shall also comply with the provisions of the relevant current national standards. 2.1 Terminology 2.1.1 cloud computing cloud computing A model for provisioning and managing scalable, elastic pools of shared physical and virtual resources over a network in an on-demand, self-service manner. 2.1.2 cloud computing infrastructure cloud computing infrastructure The infrastructure supporting cloud computing composed of hardware resources and resource abstract control components, including the software and hardware equipment and cloud management platform required to provide computing, storage, network, and security resources for cloud service customers. 2.1.3 resource pool resource pool A collection of physical or virtual resources that can be taken from a pool or recycled back to a pool. Resources include physical machines, virtual machines, physical network devices, virtual network devices, and IP addresses. 2.1.4 virtual machine virtual machine A virtual data processing system is under the exclusive use of a specific user, but its functions are realized by sharing various resources of the real data processing system. 2.1.5 host machine host machine Refers to the physical server that provides the hardware environment for the virtual machine. 2.1.6 physical machine physical machine Refers to providing users with computing resources in the form of physical servers relative to physical servers of virtual machines. 2.1.7 virtual cluster virtual cluster Consisting of a certain number of physical servers and the virtual machines on them, the virtual machines in the same cluster can support load balancing, online migration, failover and other high-availability functions. 2.1.8 virtual machine template Configure the metadata collection required by the virtual machine, including the number of CPUs, memory size, and disk size. 2.1.9 virtual machine image virtual machine image The file system image corresponding to the virtual machine, including the operating system and software required for running the virtual machine. 2.1.10 object storage object storage Cloud storage that uses objects as storage units and provides object-level access interfaces. 2.1.11 file storage file storage A storage system that connects to the server through Ethernet, and the server accesses data through standard protocols such as NFS, CIFS, HTTP, and FTP. The data is connected to the server through Ethernet, and the data has a packing/unpacking process. 2.1.12 block storage block storage It is connected to the server through SCSI/SAS or FC SAN, and the server directly controls and accesses the data storage system through SCSI/SAS or FC protocol. 2.2 Abbreviations AZ Availability Zone availability zone CIFS Common Internet File System Common Internet File System CPU Central Processing Unit Central Processing Unit CRM Customer Relationship Management Customer Relationship Management FC SAN Fiber Channel Storage Area Network Fiber Channel Storage Area Network FTP File Transfer Protocol file transfer protocol GPU Graphics Processing Unit graphics processing unit HA Host Aggregate host group HTTP Hyper Text Transfer Protocol Hypertext Transfer Protocol IB InfiniBand InfiniBand Technology IDS Intrusion Detection System Intrusion Detection System IOPS Input/Output Operations Per Second The number of read and write operations per second IP Internet Protocol Internet Protocol IPv6 Internet Protocol Version 6 Internet Protocol Version 6 IPS Intrusion Prevention System Intrusion Prevention System IPSec Internet Protocol Security Internet Protocol Security LAN Local Area Network RAID Redundant Array of Independent Disks independent redundant disk array SAN Storage Area Network storage area network SAS Serial Attached SCSI Serial Attached SCSI SCSI Small Computer System Interface small computer system interface SDN Software Defined Network Software Defined Network SLA Service Level Agreement Service Level Agreement TCP Transmission Control Protocol Transmission Control Protocol vCPU Virtual Central Processing Unit virtual processor VLAN Virtual Local Area Network Virtual Local Area Network VxLAN Virtual Extensible LAN Virtual Extended Local Area Network VPN Virtual Private Network virtual private network WAF Web Application Firewall Web Application Firewall 3.1 Overall Architecture 3.1.1 The overall architecture of cloud computing infrastructure should include resource pools and cloud management platforms (Figure 3.1.1). 3.1.2 The resource pool shall include relevant hardware and software such as servers, storage devices, network devices, security devices, virtualization software, and storage software required to provide external computing resources, storage resources, network resources, and security resource services. 3.1.3 The cloud management platform should conduct unified management and scheduling of computing resources, storage resources, network resources and security resources in the resource pool, and should provide services to users. 3.1.4 Cloud computing infrastructure should have protection measures such as physical and environmental security, network and communication security, equipment and computing security, application and data security, and management security, and should comply with the current national standard "Information Security Technology Network Security Level Protection Basic Requirements" GB/T 22239 related regulations. 3.2 Functional requirements 3.2.1 Cloud computing infrastructure should provide users with computing, storage, network and security resources in the form of different service templates according to user needs; it should support on-demand resource usage and billing based on the network, and it should support self-service, rapid deployment and Resource dynamic elastic expansion. 3.2.2 The resource pool function shall comply with the following regulations. 1.Shared or exclusive physical and virtual computing resources should be provided; 2 should provide object storage, file storage or block storage resources; 3 should provide physical or virtual network resources; 4 Security resources such as access control, intrusion prevention, security audit, malicious code prevention, and vulnerability scanning should be provided. 3.2.3 The cloud management platform shall provide resource management, scheduling management, service management, unified portal and interface management functions. 4.1 General requirements 4.1.1 The cloud computing infrastructure should be overall planned and built in a unified manner according to business needs. The recent construction scale should be coordinated with the long-term development plan. The system should meet the requirements of stable performance, safety and reliability, good compatibility, strong scalability, green energy saving, etc. Require. 4.1.2 The software and hardware architecture of cloud computing infrastructure should fully consider the security policies and mechanisms of system operation, and should adopt various technical means to provide comprehensive security technical guarantees. 4.1.3 The cloud computing infrastructure should be divided into different security domains according to business needs, so that logical areas with the same level of protection requirements can share protection methods; means of protection. 4.1.4 Computing resources, storage resources, network resources, security resources, and cloud management platforms of cloud computing infrastructure should abstract resource models in combination with business requirements or live network operating data, and should realize reasonable allocation and optimization of software and hardware resources. 4.1.5 The hardware and software equipment of the cloud computing infrastructure should support IPv6, and standardized design components should be used. 4.1.6 The key equipment of the cloud computing infrastructure should have high reliability, the important components should share the load, the key components should have hot backup, and they should have the function of automatic switching in case of failure. 4.1.7 The virtualization and other software used in the cloud computing infrastructure shall have the ability to be compatible with hardware devices such as servers, networks, and storage from different manufacturers. 4.1.8 The software and hardware used in the cloud computing infrastructure should be easy to install and upgrade, and should provide a friendly user management interface. 4.2 Business Parameters 4.2.1 When the cloud computing infrastructure project is newly built, the business parameters should be determined according to the market business development strategy, user scale and user behavior forecast; when the project is expanded, the business parameters should be determined based on the operating data of the existing network. 4.2.2 Computing resource business parameters shall include virtual machine model, number of virtual machines, ratio of single vCPU to single physical CPU core processing capability, physical machine model, number of physical machines, and floating-point computing capability. 4.2.3 Storage resource business parameters should include storage type, storage capacity, data reliability, IOPS, number of concurrent connections, access delay, and network bandwidth. 4.2.4 Network resource business parameters should include network bandwidth, IP address requirements, and network access methods. 4.2.5 Security resource business parameters should include throughput and number of concurrent connections. 4.3 Network Architecture 4.3.1 When the network traffic of the resource pool is dominated by north-south traffic, a tree network architecture should be adopted, which can be divided into core layer, aggregation layer and access layer (Figure 4.3.1). 4.3.2 When the resource pool network traffic is mainly east-west traffic, a leaf-spine network architecture should be adopted (Figure 4.3.2). For interworking, spine switches should be fully interconnected with leaf switches. 4.4 Hierarchical Division of Resource Pools 4.4.1 Resource pools can divide resources according to the hierarchical structure of region, AZ, and HA according to business needs, and should meet the following requirements. 1 The resource pool can be divided into logically isolated areas according to the geographical concept; 2 It can be divided into physically isolated AZ according to the independent power supply system; 3 Different HAs can be divided according to dimensions such as security domains, hardware specifications, and node uses. 4.4.2 The network interconnection between resource partitions at each level shall comply with the following regulations. 1 Regions should be interconnected through high-bandwidth links; 2 Different AZs in the same area should be interconnected through high-bandwidth, low-latency links, and online migration should be supported. 4.4.3 HA can be composed of multiple server clusters. Server clusters can be divided into virtualized computing clusters, physical server clusters, storage server clusters, and network function virtualization clusters. 4.5 Device configuration 4.5.1 The selection and configuration of equipment in the cloud computing infrastructure should be determined based on factors such as business scenarios, performance indicators, contracted service levels, equipment cost, equipment room conditions, and operation and maintenance requirements, and the types of equipment should not be too many. 4.5.2 The computing and storage resources of the cloud management platform should be configured independently, and should not be shared with computing and storage resources rented by users. 4.5.3 The utilization rate of CPU, memory, network bandwidth, throughput capacity and session processing capacity during busy hours of resource pool equipment should not be greater than 70%. 4.5.4 The selection and configuration of server equipment shall meet the following requirements. 1.It is advisable to choose a general-purpose server; 2 It is advisable to configure a cost-effective multi-core, multi-cache CPU processor, and the CPU should support hardware-assisted virtualization technology; 3 When the business has high requirements for parallel computing capabilities such as image processing, a GPU processor can be added; 4 The configuration ratio of server memory and CPU core should be determined according to actual business needs; 5 The server redundancy factor should be determined according to factors such as server failure, elastic expansion, migration, and resource fragmentation. 4.5.5 The number of hosts in the resource pool shall be calculated according to the following formula. In the formula. Ncore - the number of physical CPU cores; Nvmi——the demand number of the i-th virtual machine, i=1, 2,, n, n is the number of virtual machine demand types; Nvcpui——the number of vCPU cores of the i-type virtual machine; P - the ratio of the processing power of a single vCPU to a single physical CPU core; PR1, PR2 - system redundancy coefficient; Nmem - host memory value; Nvmemi——the memory value of the i-type virtual machine; Acore——the number of CPU cores configured by a single host; Amem—the memory value configured by a single host; N——The number of host machines. 4.5.6 The number of server GPU configurations in the resource pool should be calculated according to the following formula. In the formula. TGPU——GPU total floating-point computing capability requirement; Si——the amount of data to be processed by the i-th function, i=1, 2,, m, m is the number of function items; Ki—for the i-th function, the GPU floating-point computing power required per unit of data; PR3——system redundancy coefficient; NGPU - the number of GPUs required to be configured; T0—Single GPU floating-point computing capability. 4.5.7 The number of servers of the cloud management platform shall be calculated according to the following formula. In the formula. M - the number of physical servers on the management platform; Bcore——the number of CPU cores configured by a single physical server; Fi——the number of physical server CPU cores required by each function, i=1, 2,, n, n is the number of functional modules of the management platform; Ni—the number of network elements that need to be processed by the i-th function, such as the number of virtual machines, the number of physical machines, etc.; Pi—the i-th function processing single network element processing capability; P0——the processing power of a single physical CPU core; PR—system redundancy factor. 4.5.8 The selection and configuration of storage devices shall comply with the following regulations. 1 Different types of storage should be configured according to business storage requirements; 2 The disk array should be equipped with at least dual controllers and multi-path management software, and the cache capacity, disk capacity and IOPS should meet business requirements; 3 For a storage system composed of general server deployment storage software, the hard disk configuration of the server should meet the storage capacity and IOPS requirements. 4.5.9 The storage capacity used for storage resource services shall be calculated according to the following formula. In the formula. S - storage capacity; SR - business data volume; SS - system management data volume; R——system redundancy coefficient. 4.5.10 The storage capacity of the cloud management platform shall be calculated according to the following formula. In the formula. MS - management platform storage capacity; Si——the storage capacity of each management function, Si should be calculated according to the formula (4.5.9) of this standard, i=1, 2,, n, n is the number of function modules of the management platform. 4.5.11 The number of concurrent connections of the storage system shall be calculated according to the following formula. In the formula. Pc - the number of concurrent connections; Au - the total number of users; Pu - user online concurrency; Ta - the average number of threads per user. 4.5.12 The configuration and functions of network equipment shall comply with the following regulations. 1 Network equipment should have good burst traffic buffering capability and support priority control, and should have wire-speed forwarding capability. 2 The switch should be configured according to factors such as traffic convergence ratio, port number and rate, network throughput, etc. The functions and configuration principles of switches at each layer should meet the following requirements. 1) Core switches, aggregation switches, and spine switches should all support Layer 2 functions and VxLAN-based networking and Layer 3 switching functions; 2) Core switches, aggregation switches, and spine switches should have strong backplane processing capabilities, and should be dual-master, slot-type chassis equipment; 3) Access switches should support Layer 2 functions; leaf switches should meet the functional requirements of access switches and should support VxLAN-based networking functions, and should support Layer 3 when interconnecting with external networks of resource pools, firewalls, load balancing and other functional devices swap function. 3 Functional equipment such as firewalls and load balancing should be configured according to factors such as network throughput, number of TCP connections, port number, and speed, and can be deployed in hardware or software according to business needs; it should support network virtualization and hardware multi-tenancy Or software instance deployment method. 4.5.13 The configuration and functions of safety equipment shall meet the following requirements. 1.The selection and configuration of security equipment should ...... Tips & Frequently Asked Questions:Question 1: How long will the true-PDF of GB/T 51399-2019_English be delivered?Answer: Upon your order, we will start to translate GB/T 51399-2019_English as soon as possible, and keep you informed of the progress. The lead time is typically 4 ~ 7 working days. The lengthier the document the longer the lead time.Question 2: Can I share the purchased PDF of GB/T 51399-2019_English with my colleagues?Answer: Yes. The purchased PDF of GB/T 51399-2019_English will be deemed to be sold to your employer/organization who actually pays for it, including your colleagues and your employer's intranet.Question 3: Does the price include tax/VAT?Answer: Yes. Our tax invoice, downloaded/delivered in 9 seconds, includes all tax/VAT and complies with 100+ countries' tax regulations (tax exempted in 100+ countries) -- See Avoidance of Double Taxation Agreements (DTAs): List of DTAs signed between Singapore and 100+ countriesQuestion 4: Do you accept my currency other than USD?Answer: Yes. If you need your currency to be printed on the invoice, please write an email to Sales@ChineseStandard.net. In 2 working-hours, we will create a special link for you to pay in any currencies. Otherwise, follow the normal steps: Add to Cart -- Checkout -- Select your currency to pay. |
