Cloud storage technology

The technological revolution must have originated from a technological breakthrough, so what is the origin of cloud computing? From a technical perspective, the breakthrough of cloud computing is the breakthrough of mass storage technology and parallel computing technology based on this. It is these breakthroughs that have created emerging Internet companies such as Google, Facebook, and Twitter. On the other hand, enterprise information systems are becoming increasingly complex and huge. The almost unlimited expansion, massive computing, and high-availability features demonstrated by the above-mentioned technological breakthroughs are what enterprises urgently need.

1. What Cloud Storage Technology Brings

It is understandable that cloud storage technology first originated from Internet companies, and the amount of access these companies face is not encountered by any enterprise before. You can know from one piece of data: Alipay now has 300 million new records every day. Obviously, the amount of data and operations on this are not supported by traditional relational databases.

The cloud storage technology mentioned here does not specifically refer to a certain technology, but a general term for a large class of technologies, generally from any database that has the following characteristics can be regarded as cloud storage technology. The first is a total of almost unlimited expansion capacity, which can support hundreds of terabytes of data up to PB level; in addition, it uses a parallel computing mode to obtain massive computing power. In short, when the computing power is insufficient, whether it is storage or computing, for the demander, it is simply added to the machine. A further feature of cloud storage is high availability, which means that the system can be used normally at any time, even if a machine fails. At present, the common systems that meet such characteristics include Google's GFS and BigTable, Apache Foundation's Hadoop (HDFs and HBaSe), originally from Amazon and now also belongs to the Apache Foundation's Cassandra, in addition to MongoDB, CouchDB, Hypetable, Redis and many more.

As scalability refers to the system architecture allows the system to provide more services without reducing the performance characteristics. It is the easiest way to achieve scalability by increasing the capacity of the hardware and the vertical expansion of the memory through the existing machine, but there is a limit. The horizontal expansion requires adding more machines, each machine provides all or part of the data. In this way, all hosts do not have to bear all business requests. But the software itself needs an internal mechanism to ensure data synchronization between nodes in the cluster. The scalability brought by cloud storage technology is almost unlimited, and the investment (hardware investment) and output (providing more services) are almost linear for investors.

In the final analysis, horizontal expansion is to use more hosts to undertake operations. Assuming that a host has n failures in one year of operation, then the number of cattle failures of 20 hosts in one year of operation is 20 × n, which shows that when a host in a cluster The number has reached a certain level, the probability of failure in a year will be very large, and even a machine failure every day is not alarmist. Many cloud storage technologies use this as a basic design prerequisite, and this cloud storage technology inherently has good high availability and fault tolerance.

Is it possible to replace all current enterprise applications? No. Before implementing the application, it is necessary to satisfy EricBrewer's CAP theory (proven by theory in 2002). According to this theory, a large-scale distributed database system has the following three requirements:

1 Consistency (Consistency): For all database clients using the same query can get the same result, even when there are concurrent updates.

2 Availability (AvailabiLity): All database clients can always read and write data.

3 Partition Tolerance (PartitionTolerance): The database can be distributed to multiple machines, even if a network failure occurs, it is divided into multiple partitions and can still provide services.

The CAP theory states that only two of these three characteristics can be available at the same time. The traditional relational database emphasizes consistency (C) and availability (A), and the support for partition tolerance (P) is very limited. This essentially reveals the problems of the above relational database. Looking at the cloud storage technology, the emphasis is on partition tolerance (P) to make up for the shortcomings of relational data inventory. The next difference is whether to choose availability (A) or consistency (C). For the CP system, what is abandoned is availability (A), and the data can be kept consistent, but if a node fails, some data will still be inaccessible; and for the AP system, what is abandoned is consistency (C), then you System may return less accurate data.

The above technical characteristics determine that cloud storage technology has some special areas. For example, the access traffic may be large, that is, the amount of data accessed at any time is large, which requires large-scale distributed deployment. Investigate the ratio of read and write operations, especially suitable for statistical analysis work. Sometimes the requirements for data consistency are not high. It can be tolerated that after a certain data is written, some users may read the data before writing within a reasonable period of time. The search service is a typical example.

Cloud storage technology

Stainless Steel Garden Edging

Flexible garden edge Flexible to create the shapes you want, from intricate curves to endless straight lines, even right angle bends without a cut and join. At thesame time, it is cost-effective, wide range of applications, installation is very convenient.Stainless steel garden edge is stronger than aluminium and traditional plastic edging materials. It holds its shape for many years. It also has high strength, good quality, weather resistance and durability. size can be customized, it is not only beautiful but also good concealment after installation.The retaining plate is a temporary retaining structure to prevent the collapse of the earth in the trench and foundation pit.Garden fences can be made of cast iron, sturdy, and decorated with green vines, which can be very beautiful.The use of fence is very wide, fence network is often used for boundary isolation or regional division, regional protection.Widely used in: industry, agriculture, animal husbandry, forestry, highway, railway, national defense, airport, port, municipal fence, greening, grassland, orchard and other sites of isolation and protection of the fence net.

74

Stainless Steel Garden Edging,Stainless Steel Lawn Edging,Stainless Steel Edging Landscape

Henan Jinbailai Industrial Co.,Ltd , https://www.jblbbqgrill.com