A comprehensive interpretation of 8 core technologies in cloud computing

Cloud computing "turned out" so many people see it as a new technology, but in fact its prototype has been for many years, only in recent years began to make relatively rapid development. To be exact, cloud computing is the product of large-scale distributed computing technology and the evolution of its supporting business model, and its development depends on virtualization, distributed data storage, data management, programming mode, information security and other technologies, and the common development of products. In recent years, the evolution of business models such as trusteeship, post-billing and on-demand delivery has also accelerated the transition to the cloud computing market. Cloud computing not only changes the way information is delivered, it also subverts the delivery model of traditional ICT systems. Cloud computing is not so much a technological innovation as a shift in thinking and business models.

Here is a talk about the core technology of cloud computing.

Cloud computing is a data and processing capacity-centric dense computing model, which integrates a number of ICT technologies, is the traditional technology "smooth evolution" of the product. Among them, virtualization technology, distributed data storage technology, programming model, large-scale data management technology, distributed resource management, information security, cloud computing platform management technology, green energy-saving technology is the most critical.

1, Virtualization Technology Virtualization is one of the most important core technologies of cloud computing, which provides the infrastructure support for cloud computing services, and is the main driving force for ICT services to rapidly move to cloud computing. It can be said that without virtualization technology there will be no cloud computing services to the landing and success. As the application of cloud computing continues to warm up, the industry's emphasis on virtualization technology also mentions a new height. At the same time, our survey found that many people are aware of cloud computing and virtualization are mistaken, that cloud computing is virtualization. In fact, virtualization is an important part of cloud computing, but not all.

Technically, virtualization is a computational form that simulates computer hardware in software and provides services to users with virtual resources. The aim is to rationally allocate computer resources to provide services more efficiently. It breaks the physical division between the hardware of the application system, thus realizes the dynamic structure and realizes the centralized management and use of physical resources. The greatest benefit of virtualization is to enhance the elasticity and flexibility of the system, reduce costs, improve services, improve resource utilization efficiency.

From the form of manifestation, virtualization is divided into two kinds of application mode. One is a powerful server virtual into a number of independent small servers, service different users. The second is to make multiple servers virtual into a powerful server, complete the specific functions. The core of the two models is unified management, dynamic allocation of resources, improve resource utilization. In cloud computing, both of these models have more applications.

2, distributed data storage technology another big advantage of cloud computing is the ability to deal with massive data quickly and efficiently. This is crucial today in the data explosion. To ensure the high reliability of data, cloud computing typically uses distributed storage technology to store data in different physical devices. This model not only free from the limitations of hardware devices, but also better scalability, can quickly respond to changes in user needs.

Distributed storage is not exactly the same as traditional networked storage, the traditional networked storage system uses centralized storage server to store all the data, and the storage server becomes the bottleneck of the system performance, and can not meet the needs of large-scale storage applications. Distributed networked storage system, which uses scalable system structure, uses multiple storage servers to share storage load, and locates storage information by location server, it not only improves the reliability, usability and access efficiency of the system, but also expands easily.

In the current cloud computing world, Google's GFS and Hadoop open source system HDFs are two of the more popular cloud computing distributed storage systems.

GFS (Google File System) technology: Google's non-open source GFs (googlefile System) cloud computing platform meets the needs of a large number of users, providing services in parallel to a large number of users. The data storage technology of cloud computing has the characteristics of high throughput rate and high transmission rate.

HDFS (Hadoop distributed File System) technology: Most ICT vendors, including Yahoo and Intel's "cloud" program, adopt HDFS data storage technologies. Future developments will focus on large scale data storage, data encryption and security assurances, and continue to improve I/O rates.

3, the programming mode is essentially, cloud computing is a multi-user, multi-task, support concurrent processing system. Efficient, simple, fast is its core concept, it is designed to network the powerful server computing resources easily distributed to end users, while ensuring low-cost and good user experience. In this process, the choice of programming mode is of paramount importance. The distributed parallel programming model will be widely used in cloud computing projects.

The purpose of the distributed parallel programming model is to make use of software and hardware resources more efficiently and to enable users to use applications or services more quickly and simply. In distributed parallel programming mode, background complex task processing and resource scheduling are transparent to users, so the user experience can be greatly improved. MapReduce is one of the current mainstream parallel programming patterns in cloud computing. The MapReduce mode automatically divides tasks into subtasks and achieves the task's height and distribution in a large scale computing node through both map and reduce steps.

MapReduce is a Java, Python, C + + programming model developed by Google, mainly for parallel operations with large datasets (larger than 1TB). The idea of the MapReduce mode is to decompose the problem into a map (map) and reduce (simplify) by the map program to cut the data into unrelated chunks, allocate (dispatch) to a large number of computers, and achieve the results of distributed operations. The result is then remitted through the reduce program.

4, large-scale data management processing massive data is a big advantage of cloud computing. So how to deal with many aspects of things, so efficient data processing technology is also one of the core technologies of cloud computing. Data management faces huge challenges for cloud computing. Cloud computing is not only to ensure data storage and access, but also to be able to carry out a specific search and analysis of massive data. Because cloud computing needs to process and analyze a large amount of distributed data, data management technology must be able to manage a lot of data efficiently.

Google's BigTable Data management technology and the open source data management module developed by the Hadoop team HBase is a typical large-scale data management technology in the industry.

BT (BigTable) Data management technology: BigTable database is a distributed, persistent storage multidimensional ranking map.bigtable based on GFS, Scheduler, Lock service and MapReduce, Unlike a traditional relational database, it handles all the data as objects, forming a huge table to distribute and store large-scale structured data. BigTable is designed to reliably handle PB-level data and can be deployed to thousands of machines.

Open source data Management module Hbase:hbase is a subproject of the Apache Hadoop project, positioned in a distributed, column-oriented open source database. Unlike a generic relational database, HBase is a database suitable for unstructured data storage. Another difference is that HBase is based on columns rather than on a row based pattern. As a highly reliable distributed storage system, HBase has a better performance and scalability. Using hbase technology, large structured storage clusters can be built on inexpensive PC servers.

5. Distributed resource Management cloud computing uses distributed storage technology to store data, so it is natural to introduce distributed resource management technology. In a multi-node concurrent execution environment, the States of each node need to be synchronized, and when a single node fails, the system needs an effective mechanism to ensure that other nodes are not affected. The Distributed resource management system is just such a technology, it is the key to ensure the state of the system.

In addition, cloud computing system processing resources are often very large, fewer than hundreds of servers, many tens of thousands of units, and may span multiple regions. And the application of the cloud platform is also thousands of applications, how to effectively manage this batch of resources to ensure that they provide services normally, the need for strong technical support. Therefore, the importance of distributed resource management technology is conceivable.

The Global Cloud Computing solutions/service providers are actively engaged in the development of relevant technologies. Google's internal use of Borg technology is highly praised by the industry. In addition, Microsoft, IBM, Oracle/sun and other cloud computing giants have a corresponding solution proposed.

6, Information Security survey data show that security has become an obstacle to the development of cloud computing is one of the main reasons. Data show that 32% of organizations that already use cloud computing and 45% of ICT management in organizations that have not yet used cloud computing have made cloud security the biggest obstacle to further deployment of the cloud. Therefore, to ensure long-term stability and rapid development of cloud computing, security is the first problem to be solved.

In fact, cloud computing security is not a new problem, the traditional internet has the same problem. Only after the advent of cloud computing has security issues become more pronounced. In the cloud computing system, security involves many aspects, including network security, server security, software security, system security, and so on. Therefore, some analysts believe that the development of cloud security industry, will be the traditional security technology to mention a new stage.

Now, both software security vendors and hardware security vendors are actively developing cloud computing security products and solutions. Including traditional anti-virus software vendors, soft and hard firewall manufacturers, IPS vendors, including all levels of security providers have joined the cloud security field. I believe that in the near future, cloud security issues will be well resolved.

7, cloud computing platform Management cloud computing resources Large scale, the number of servers and distributed in different locations, while running hundreds of applications, how to effectively manage these servers, to ensure that the entire system to provide uninterrupted service is a huge challenge. The platform management technology of cloud computing system needs the ability to efficiently deploy a large number of server resources to make it work better. It is the key of cloud computing platform management technology to easily deploy and open new business, quickly discover and recover system fault, realize large-scale system reliable operation through automation and intelligence.

For providers, cloud computing can have three deployment patterns, namely, the public cloud, the private cloud, and the hybrid cloud. Three modes are very different for platform management requirements. As for the users, because of the control of ICT resource sharing, the requirements of system efficiency and the budget of ICT cost, the scale and manageability of the cloud computing system required by enterprises are very different. Therefore, the Cloud computing platform management scheme should take into account the customization requirements, can meet the application requirements of different scenarios.

Many vendors, including Google, IBM, Microsoft, Oracle/sun and others, have cloud computing platform management solutions. These programs can help enterprises to achieve infrastructure integration, enterprise hardware resources and software resources Unified management, unified distribution, unified deployment, unified monitoring and unified backup, break the monopoly of the application of resources, so that the value of the enterprise cloud computing platform can be fully realized.

8, green energy-saving technology energy conservation and environmental protection is the global era of the major theme. Cloud computing is also said to be low-cost and efficient. Cloud computing has huge economies of scale, and it saves a lot of energy while increasing the efficiency of resource utilization. Green energy-saving technology has become the essential technology of cloud computing, and more and more energy-saving technologies will be introduced into cloud computing in the future.

Carbon Disclosure Project (the Carbon Disclosure Project, CDP) recently published a study on cloud computing that helps reduce carbon emissions. The report notes that U.S. companies migrating to the cloud can reduce their carbon emissions by 85.7 million tonnes a year, equivalent to the total carbon emitted by 200 million barrels of oil.

In short, cloud services providers need to continually improve their technology to make cloud computing greener.