Distributed file system means that the physical storage resources managed by the file system are not necessarily directly connected to local nodes, but connected to nodes through computer networks (which can be simply understood as a computer). The design of distributed file system is based on client/server mode. A typical network may include multiple servers accessed by multiple users. In addition, peer-to-peer functionality allows some systems to play the dual roles of client and server. For example, users can "publish" directories that other clients can access. Once accessed, this directory is like a client using a local drive.
What is superconvergence architecture?
Hyper-converged infrastructure (HCI), also known as hyper-converged architecture, means that the same set of unit equipment (x86 server) not only has resources and technologies such as computing, networking, storage, server virtualization, but also includes cache acceleration, deduplication, online data compression, backup software, snapshot technology and other elements, and multiple nodes can be aggregated through the network to achieve modularization and seamless.
What is the difference between hyper-converged products and distributed file systems?
Hyper-converged architecture generally includes storage module, virtualization module, control management module, data protection and disaster tolerance module. This kind of storage generally adopts distributed storage. Distributed storage is not necessarily a distributed file system, but also a distributed block storage. For example, SmartX's hyper-converged architecture is based on distributed block storage, and some vendors are based on distributed file storage.
However, if we deeply study the implementation technology of distributed block storage, we will find that some distributed block storage is based on distributed file system. The general principles are as follows: first, manage physical disks with a distributed file system to form a pool (much like Google GFS); Create files from this pool, such as 500GB files. Then through the block device gateway, such as iSCSI gateway, this 500GB file will be turned into a 500GB virtual hard disk (block device) to provide services to the outside world.
Hyperconverged system architecture