什么是块存储？

The main pro for block storage is performance, and is ideal for applications that require low latency, and which will make a tremendous number of IOPS calls.

• Storage Utilization Optimization — Because block storage is capable of saving data blocks to multiple devices, and environments, storage systems can pool resources and maximize storage. By utilizing every inch of data storage capacity, regardless of the environment, teams can compact and optimize their storage usage, and avoid leaving unused capacity sitting in data centers.
• Data Consistency — Because of how block storage is architectured, and how it performs over object storage, block storage is considered to have a “strong” data consistency model. This means that once data is changed at the block-level, modifications are immediately made available to the system and other users. Whereas object storage, though with its own benefits, is considered “eventually” consistent, which denotes that data available is not immediate. This is because data needs to propagate in object storage, objects must be replicated to be modified and then replace outdated objects, and until it has, data remains outdated.
• Performance — Block storage is often chosen for its high speed due to the uniform size of blocks. Additionally, in comparison to object storage, which utilizes a metadata system for accessing data rather than an address style system. Applications that require random access to data benefit greatly from how block storage access data, which allows block-level edits.

Advancements in technology have reduced many of the disadvantages that block storage used to have, like the inability to store across environments, which is now a block storage capability and advantage. But there are still drawbacks to deploying block storage.

• Dedicated Network Fabric — Block storage is heavily relied upon for storage area networks (SAN). SANs utilize multiple block-level data storage devices, consolidating them into a storage network that is then connected to the main network, seamlessly providing storage capacity to users, while enabling storage scaling. But, these storage networks are held together with its own network “fabric” that is complex, and potentially costly.

Complex Management — As alluded to above, SANs using block storage are more complex than directly attaching storage (DAS), or network attached storage (NAS). Therefore, it is essential to ensure that the right skill set to manage these systems is within the IT.

Block storage is used by enterprises to support business critical applications and services.

One example, transactional systems, those which require the highest performance, such as ordering, inventory, and payments systems, need real-time data consistency to ensure that each system is drawing from the freshest data sources. Block storage provides the reliability and low-latency necessary to enable these high-performance interconnected systems.

VMs are another challenge that block storage is suited to solve. In the modern cloud, which uses orchestration and automation to create and provision VMs, the number of VMs and their association to storage has grown, introducing challenges that block storage helps to solve. From a capacity standpoint, block storage enables VMs to access large amounts of storage. Regarding performance, block storage can handle the increasing number of random I/O calls produced by VMs. And thirdly, resilient block storage is highly available and can enable the concentration of VMs into less storage units

Block storage can be compared to two other common storage formats, file storage and object storage. These formats aim to store, organize, and allow access to data in specific ways that benefit certain data applications. For instance, file storage, commonly seen on desktop computers as a file and folder hierarchy, presents information intuitively to users. This intuitive format, though, can hamper operations when data becomes voluminous. Block storage and object storage both help to overcome the scaling of data in their own ways. Block storage does this by “chunking” data into arbitrarily sized data blocks that can be easily managed by software, but provides little data about file contents, leaving that to the application to determine. Object storage decouples the data from the application, using metadata as a file organization method which then allows object stores to span multiple systems, but still be easily located and accessed.