What Does a Data Center Engineer Do?

Cloud computing (English: Cloud Computing) is an Internet-based computing method. In this way, shared software and hardware resources and information can be provided to computers and other devices as needed. Typical cloud computing providers often provide general-purpose network business applications, which can be accessed through software such as a browser or other web services, while software and data are stored on the server. Cloud computing services usually provide universal online business applications accessed through a browser, and software and data can be stored in data centers. Narrow cloud computing refers to the delivery and use model of IT infrastructure, which means to obtain the required resources in an on-demand, easy-to-expand way through the network; broad cloud computing refers to the service delivery and use model, which means that the Way to get the services you need. This service can be related to IT and software, the Internet, or other services. It means that computing power can also be circulated through the Internet as a commodity.

Cloud Computing Data Center

This entry lacks an overview map . Supplementing related content makes the entry more complete and can be upgraded quickly. Come on!

Cloud Computing

Cloud computing is following the 1980s

The latest data released by International Data Corporation IDC shows that the total number of data centers in China in 2010 has reached 504,155. As of March 2012, 13 provinces, municipalities and autonomous regions across the country have planned about 30 large-scale data center construction projects with a scale of 100,000 or more servers, with a total investment of 270 billion yuan.

The industry is urging the call for cloud computing to intensify, and industry insiders have suggested avoiding blind data center investments while promoting cloud computing to enter the commercial stage. So what is the difference between the so-called "next-generation cloud computing data center" compared to traditional data centers? What value-added services? The reporter interviewed 21Vianet on this issue (

For cloud computing, we should focus on improving the data processing capabilities of cloud computing data centers from infrastructure areas such as high-end servers, high-density and low-cost servers, mass storage devices, and high-performance computing devices. Cloud computing requires good infrastructure flexibility, scalability, automation, data movement, multi-tenancy, space efficiency, and support for virtualization. So, what should the architecture of each part of the data center infrastructure be in a cloud computing environment?
1. The overall architecture of the cloud computing data center <br /> < br The cloud computing architecture is divided into two parts: service and management. In terms of services, it mainly provides users with a variety of cloud-based services, including three levels: infrastructure as a service IaaS, platform as a service PaaS, and software as a service SaaS. In terms of management, the cloud management Master, its function is to ensure that the entire cloud computing center can operate safely and stably and can be effectively managed.
2. Cloud computer room structure <br /> <br /> According to Great Wall Electronics' years of experience, in order to meet the needs of cloud computing service flexibility, the cloud computer room adopts a standardized and modular computer room design architecture. Modular computer rooms include container modular computer rooms and building modular computer rooms.
The application of the container modular machine room in the outdoor machine room-less scenario reduces the pressure on the site selection of the builder, helping the builder reduce the construction period of the original half year to two months, and the energy consumption is only 50% of the traditional machine room. It can adapt to the extreme harsh environment of desert hot and arid regions and polar cold regions. The modular machine room of the building adopts leading cooling technologies such as hot and cold air duct isolation, precise air supply, and outdoor cold sources, which can be applied to the building and expansion of large and medium-sized data centers.
3. Cloud computing network system architecture <br /> <br The overall structure planning of the network system should adhere to the regional, hierarchical, and modular design concepts to make the network hierarchy more clear and the functions clearer. The data center network is divided according to the nature of the service or the role of the network equipment, and can be planned from the following aspects.
1) According to the nature of the data transmission service and the user-oriented nature, the network system can be divided into areas such as internal core network, remote business private network, and public service network.
2) According to the role of the equipment in the network structure, the network system can be divided into a core layer, an aggregation layer, and an access layer.
3) Considering the independence of data application services of network services, the mutual access relationship between services, and the security isolation requirements of services, the network system can be logically divided into storage areas, application business areas, front-end areas, and system management areas. , Hosting area, external network access area, internal network access area, etc.
In addition, there is a fabric network architecture. After cloud computing is deployed in the data center, the traditional network structure may make network latency a major bottleneck, which makes low-latency server-to-server communication and higher two-way bandwidth requirements more urgent. This requires the network architecture to develop in a flat direction. The ultimate goal is to minimize the number of network architectures between any two points.
One of the keys to the fabric network structure is to eliminate the concept of network hierarchy. The fabric network architecture can use array technology to flatten the network. It can compress the traditional three-layer structure into two layers, and finally transform it into one layer. Connections to eliminate complexity and network latency. However, there is still no unified standard for this new technology of Fabric, and its promotion and application still need more practice.
4. Cloud computing host system architecture <br /> <br The core of cloud computing is the concentration of computing power and large-scale breakthroughs. The type of computing provided by the cloud computing center determines the hardware infrastructure of the cloud computing center. From the perspective of cloud customer needs, cloud computing centers usually need to provide the following types of computing power on a large scale. The server system can adopt a three (multi) -tier architecture. The first is high-performance, stable and reliable high-end computing, which mainly deals with compact Coupling computing tasks. Such calculations include not only external databases, business intelligence data mining and other key services, but also core systems such as own accounts and billing, which are usually provided by large enterprise-level servers. The second is general-purpose computing for many common applications. , Used to provide low-cost computing solutions, this kind of computing has low hardware requirements, and generally uses high-density, low-cost ultra-density integrated servers to effectively reduce data center operating costs and end-user use costs; Scientific computing, bioengineering and other businesses provide high-performance computing with trillions and trillions of computing capabilities. The hardware foundation is high-performance clusters.
5. Architecture of cloud computing storage system <br /> < br Cloud computing adopts the model of unified and centralized storage of data. In the cloud computing platform, how to place data is a very important issue. In the process of actual use, it is necessary to Data is distributed across multiple disks on multiple nodes. There are currently two ways of storage technology trends that can achieve this, one is to use a cluster file system similar to Google File System, and the other is a block device-based storage area network SAN system.
GFS is a distributed file system designed and implemented by Google. It is based on a cluster system consisting of a large number of ordinary PCs with Linux operating systems installed. The entire cluster system consists of a master and several chunkservers. There are many options for SAN connection. One option is to use a fiber optic network, which can operate fast fiber optic disks, and is suitable for places with high performance and reliability requirements. Another option is to use Ethernet and adopt the iSCSI protocol, which can run in a common LAN environment, thereby reducing costs. Using SAN structure, server to shared storage.
A large amount of data transmission of the device is carried out through the SAN network. The LAN only assumes the communication tasks between the servers. This division of labor allows the storage devices, servers, and LAN resources to be used more efficiently, making the storage system faster and more scalable. And reliability is better.
6. Cloud computing application platform architecture <br /> < br The cloud computing application platform adopts a service-oriented architecture SOA approach. The application platform provides the necessary infrastructure resources for deploying and running application systems. Application infrastructure, so application developers There is no need to care about the underlying hardware and application infrastructure of the application, and the resources required by the application system can be dynamically expanded according to the application requirements. The complete application platform provides the following functional architecture.
1) Application running environment Distributed operating environment Multiple types of data storage dynamic resource scaling 2) Application life cycle support Provides development SDK, IDE, etc. to accelerate application development, testing and deployment.
Public services: Provide public services in the form of APIs, such as queue services, storage services, and cache services.
Monitoring, management, and measurement: Provides management and monitoring functions for resource pools and application systems for accurate measurement. Application uses consumed computing resources.
3) Integration and composite application construction capabilities In addition to providing an application operating environment, it also needs to provide connectivity services, integration services, message services, and process services for building composite applications of SOA architectural style.
The above is some analysis of cloud computing data center architecture. Cloud computing is called "cloud" because it has the characteristics of a cloud in reality in some aspects: clouds are generally large; the scale of clouds can be dynamically scaled, and its boundaries are blurred. The business model of cloud computing provides users with an IT service whose content is also dynamic and flexible over time. Therefore, the architecture of cloud computing data centers will continue to adjust and optimize with the progress of society.