Hyperconverged infrastructure ("HCI" for short) means that there are not only resources and technologies such as computing, network, storage and server virtualization in the same group of unit devices, but also elements such as backup software, snapshot technology, deduplication and online data compression. Multiple groups of unit devices can be aggregated through the network to realize modular seamless horizontal expansion and form a unified resource pool.
What is a server SAN?
The term server SAN comes from Wikibon, which is just a storage resource pool composed of multiple independent servers directly connected to storage. High-speed interconnection (such as InfiniBand or low-latency Ethernet) is realized between DAS, and unified management is carried out through software. Server SAN can be realized by standard X86 server and high-speed general network, which saves the cost of special equipment and network and achieves higher cost performance for enterprises. The server SAN architecture combines the technical advantages of ultra-large-scale, convergence and flash memory, so that both computing and storage can enjoy the network, and it also has a good performance in scalability and manageability. Very large companies, such as Facebook, Google and Amazon, have developed server SAN solutions.
What is the relationship between hyper-converged architecture and server SAN?
Both of them are an implementation form of software-defined storage, but the difference is that hyper-converged architecture emphasizes the integration of virtualization functions, and it is deployed in a virtualized environment, while Server SAN deploys software-defined storage (SDS) to replace traditional SAN on multiple physical nodes.