Download the authoritative guide: Enterprise Data Storage 2018: Optimizing Your Storage Infrastructure
Block-level storage is a foundational data storage technology. In the enterprise, it helps makes today's large-scale databases and business applications possible.
Here's what you should know about block-level storage and how it has shaped the enterprise data storage technology landscape.
What is block storage?
Closely associated with storage area networks (SANs), block storage refers to saving data in raw storage volumes called blocks. Storage blocks, in turn, can each function as an individual hard drive.
Operating systems can access these blocks using the Fibre Channel, Fibre Channel over Ethernet (FCoE) or iSCSI protocols. Block storage is often used in workloads where network-based, low-latency storage operations are required. Examples include databases, critical applications, virtual machines, RAID implementations and even file systems. Block storage can also be used to boot an operating system, if so configured.
Block storage is popular in enterprise IT environments because of its flexibility, but it is comparatively expensive and complex to the storage technologies explored below.
How block differs from file storage
As mentioned earlier, block storage can be configured with a file system. Regardless of this apparent overlap, at its core, block storage is very different from file-level storage.
In file storage, data is stored in files and folders, or directories, typically in network-attached storage (NAS) deployments. It's a hierarchical approach that, apart from being familiar to practically any PC user, is easy to implement and simplifies data management and file sharing. Protocols used in this environment are Network File System (NFS), Common Internet File System (CIFS) and Server Message Block (SMB).
One major difference between the two is the use of metadata. File-level storage uses metadata to arrange files appropriately. Metadata is simply data or information that describes other data. In most cases, this metadata contains a file's basic attributes, including its name, file type and when it was created and last updated.
In block storage, metadata is lacking. Server operating systems directly access data blocks by their unique address instead of traversing directories and file hierarchies. It is up to the operating system or application to track and manage these blocks of data.
Although block and file storage may seem at odds, and indeed they are two very different storage technologies, they can coexist in the same enterprise storage system. Many vendors, including Dell EMC and HPE, have been rolling out unified storage systems that offer both block and file storage over the years.
How block differs from object storage
Object storage has been surging in popularity lately, thanks in large part to runaway adoption of cloud-computing services.
As the term implies, object storage packages data into objects. These objects, along with their metadata, are stored in a flat structure or address space, rather than in files, folders or as data blocks that can be reassembled into files.
Objects are typically comprised of files, and any bundled metadata, of course. Each object is assigned a unique identifier, or object ID, enabling each object to be retrieved from a single repository or pool of storage, offering enterprises greater flexibility on how and where, data is placed.
Object-based storage is also rich in metadata. Compared to the relatively little metadata used in file-based storage and the dearth of it in block storage, object storage metadata can contain much more than some basic information about a file. Businesses can add custom information, yielding additional context to support search, analytics, advanced management and other use cases beyond simply storing and retrieving data.
All told, this approach enables massive scalability using distributed architectures, along with flexible data protection and management compared to SAN environments.
Block storage advantages
Block, file and object storage: Each has its benefits and shortcomings, depending on the types of workloads your organization is running.
Here are some of the reasons why block storage may make sense in your IT environment:
- Block storage offers low-latency IO for consistent and predictable performance.
- Flexible management.
- Easy management of access and control privileges.
- Each storage volume in block storage setup can be used as an independent disk drive by an external server.
- External servers can boot up from block storage.
Drawbacks include the aforementioned lack of metadata along with the complexity and expense of deploying SANs, which often require highly skilled storage professionals to maintain and keep running optimally.
Block storage use cases
- Databases: Fast, reliable performance along with easy database storage management make block storage the go-to technology.
- Enterprise applications: Like databases, block storage helps ensure that many large and transaction-based business applications don't keep their users waiting.
- Virtualization: VMware customers can use block storage for their virtual machine file systems.
- Email servers: Running Exchange? Microsoft's email server explicitly requires block-level storage and doesn't support NAS file storage systems.
- RAID: Block storage enables organizations to configure individual disks into a RAID array for data redundancy and improved performance.
Block Storage Vendors
Block-level storage's flexibility and performance characteristics make the technology particularly well suited to enterprise-grade applications and workloads.
All major enterprise storage vendors offer SAN products, and therefore block storage solutions. Top-tier providers include Dell EMC, HPE, Hitachi Data Systems, IBM and NetApp. SAN arrays are also available from Fujitsu, along with number of all-flash storage vendors like Pure Storage, Kaminario, Tegile and Tintri.
As noted earlier, enterprises looking for arrays that can handle both their block and file workloads can find systems that fit the bill. Options include VMAX from Dell EMC and HPE 3PAR, to name a few.
In the cloud, Amazon Web Services Elastic Block Store, or AWS EBS, provides scalable block storage that can be used by Elastic Compute Cloud (EC2) instances. This allows customers to create storage volumes and place databases, enterprise applications and other workloads that typically run on a SAN on Amazon's cloud instead.
"Amazon EBS provides a range of options that allow you to optimize storage performance and cost for your workload," states the AWS EBS product page. "These options are divided into two major categories: SSD-backed storage for transactional workloads, such as databases and boot volumes (performance depends primarily on IOPS), and HDD-backed storage for throughput intensive workloads, such as MapReduce and log processing (performance depends primarily on MB/s)."
Similar elastic block storage offerings from other cloud providers include Linode Block Storage and Rackspace's Openstack-powered Cloud Block Storage service.