Object storage can be simply understood as unstructured data storage for pictures, audio, video and other data pools. With respect to the host server, has read and write speed, which will help share characteristics. Cedar Rock data summarized here to give you this six.
1, The data separation movement that dramatically improves the performance of web pages
Under normal circumstances, we are all recommended way to use the host server and storage division of labor objects to store website data. The host server is responsible for dynamic data storage site, object storage is used to store static files website. In order to achieve separation of static and dynamic website, when a user visits a Web site, are synchronous read data from the host server and object storage server, can greatly enhance the performance of web pages.
2, as easy a separate file management interface, manage site files and local computer
Whether Tencent cloud, Ali cloud, fir rock, object storage has a separate management console, Tencent Ali clouds and clouds as well as specialized computer client. You do not have to open the site, it can be used like Baidu cloud disk as use object storage to manage your website files, in addition to upload, download, preview and other common functions, you can also picture processing directly on the object storage / media transcoding / Data Analysis Wait.
3, is essentially a "built-in high-capacity hard disk, distributed servers," with a cross-domain file sharing Support
Object storage is the essence of "built-in high-capacity hard disk, distributed server" object store has its own CPU, memory, network and disk systems, with some intelligence, data storage pool of the same object can create different buckets (bucket) , are used to store data for different websites, do not interfere with each other. And can refer to the same file to a different site, it can effectively reduce data redundancy.
4, multi-node storage, cross-regional support for real-time synchronization to achieve Disaster Recovery
Suppose your pictures and other data stored on your own server, you can only protect your data by means of regular data backup. Large amount of data, then each backup will require a lot of time and consume a lot of disk space, and the management is not easy.
If your data is placed in the object storage data pool, and associated with the site. So you can back up files to a static object storage. You only need to back up a small number of dynamic file site. save time and energy.
For example, I use the object store, I still file my nearest node [Shenzhen] to store the site's pictures. But given the extreme circumstances, such as Shenzhen node due to a sudden power failure, it may result in total data loss. But if before I chose [Hangzhou] as the second storage node, and set the data in real time increment [Shenzhen] node synchronization to [Hangzhou] node. So even if the data Shenzhen node is lost, but I'm still in Hangzhou, data can be used. Wait until power is restored, I then synchronize data back to Shenzhen from Hangzhou node node on it. This is what we say off-site disaster recovery.
5, low-cost, resource resilient and elastic, pay-as
Unlike the flow object storage server is fixed, which is included in the cost of the server. Whether or not you use are so many. Object storage can be charged depending on your actual usage.
6, save server space
Why these little on the final say? Because now the price of the server's fairly affordable, activity will be more. Most owners have enough server space problem does not exist. But there is not the object storage space is not enough to solve the problem with the server. Block storage but to bind [], [] file storing their own advantages to achieve efficient sharing and document literacy. But saving server space or facts not win, they still mention it.