What are the domestic companies that do distributed storage research and development?

Distributed storage is a data storage technology that uses the disk space on each machine in the enterprise through the network, and these scattered storage resources form a virtual storage device, and the data is stored in all corners of the enterprise.

I have done some integration projects, and I know a little about the domestic cluster NAS (distributed file system), and I will talk about it casually, which is limited to the general cluster NAS. There are many companies that advertise self-developed cluster NAS, most of them are OEMs, not many do R&D, and even fewer are completely self-developed. Let's list some of the ones that have been contacted, are completely self-developed, and have a relatively high degree of productization.

Bigger, Huawei (Oceanstor 9000), Dawning (Parastor). There are more of them in the market. Smaller, Long Cun, this is an old brand.

Jucun, not many people know this. There are a lot of companies based on ceph in recent years, and I have used one of them, and the block is okay, but the document will take time. Distributed storage is actually a relatively large field.

There are distributed databases, distributed file systems, distributed blocks (serversan), distributed object storage, and so on. There are a lot of companies that do it, but there are very few that are really their own, and many of them are open source. In terms of distributed data, Internet companies generally use a lot, such as Ali and Tencent, which have their own distributed databases.

In terms of domestic distributed file systems, a group of people from the Chinese Academy of Sciences are the earliest. People who came out of the Chinese Academy of Sciences are basically divided into three, Shuguang, Longcun, and Davao. These three are basically self-developed, and the application time is the longest in China.

The initial team was the core R&D team from DB2 Lab in North America, and the database was developed in 2011. At that time, it was considered that: 1) the open-source distributed database technology was immature, and instead of modifying it, it was better to build a more enterprise-level database based on its own database development strength. 2) The technical continuity of packaging and open source is weak, which has a great impact on the subsequent commercial operation and productization.

3) As a database R&D team, with such technical strength, I hope to truly build my own product. 4) Seeing the unique needs of the Chinese market, we firmly follow the idea of independent research and development, commercial software, and open source operation. Now, after more than 5 years, our team has basically matured, with a leading database R&D team in China, and at the same time, we have introduced a pre-sales and after-sales and technical support team including DB2 level2 technical support expert group in China, as well as many big data technology experts.

Jushan has also taken root in many industries, and it can be said that it has realized the process of successfully commercializing and marketizing from completely independent technology products to the market. Therefore, our team has been recognized by the capital side, and the company has also obtained DCM's nearly 100 million yuan in Series B financing in the second half of 16 years. <>

I. CEPH

CEPH originated from Sage's Ph.D. work, which was published in 2004 and subsequently contributed to the open source community. After years of development, it has been supported by many cloud computing and storage vendors, and has become the most widely used open source distributed storage platform.

2. GFS

GFS is Google's distributed file storage system, which is designed to store massive search data, and was proposed in 2003 as a closed-source distributed file system. It is suitable for a large number of sequential reads and sequential appends, such as reads and writes of large files. Focus on the continuous and stable bandwidth of large files, rather than the latency of a single read and write.

3. HDFS

HDFS (Hadoop Distributed File System) is a distributed file system suitable for running on commodity hardware, which is the core sub-project of Hadoop, which is developed based on the needs of accessing and processing large files in the streaming data mode. The system is modeled after the Google File System (GFS) and is a simplified and open-source version of GFS.