InfraDMS Storage Alternatives.

Attached.  Networked. Offline.
Hard Disk Drive On White

Traditional vs. Cloud-Based Storage Alternatives

Architects of traditional, on-premises IT infrastructures and applications have numerous potential data storage choices, including the following:

  • Memory—In-memory storage, such as file caches, object caches, in-memory databases, and RAM disks, provide very rapid access to data.
  • Message Queues—Temporary durable storage for data sent asynchronously between computer systems or application components.
  • Storage area network (SAN)—Block devices (virtual disk LUNs) on dedicated SANs often provide the highest level of disk performance and durability for both business-critical file data and database storage.
  • Direct-attached storage (DAS)—Local hard disk drives or arrays residing in each server provide higher performance than a SAN, but lower durability for temporary and persistent files, database storage, and operating system (OS) boot storage than a SAN.
  • Network attached storage (NAS)—NAS storage provides a file-level interface to storage that can be shared across multiple systems. NAS tends to be slower than either SAN or DAS.
  • Databases—Structured data is typically stored in some kind of database, such as a traditional SQL relational database, a NoSQL non-relational database, or a data warehouse. The underlying database storage typically resides on SAN or DAS devices, or in some cases in memory.
  • Backup and Archive—Data retained for backup and archival purposes is typically stored on non-disk media such as tapes or optical media, which are usually stored off-site in remote secure locations for disaster recovery.

Each of these traditional storage options differs in performance, durability, and cost, as well as in their interfaces. Architects consider all these factors when identifying the right storage solution for the task at hand. Notably, most IT infrastructures and application architectures employ multiple storage technologies in concert, each of which has been selected to satisfy the needs of a particular subclass of data storage, or for the storage of data at a particular point in its
lifecycle. These combinations form a hierarchy of data storage tiers.