If you’ve been hearing people in the IT profession talking a lot about object storage lately, there’s a good reason. It is a scalable, cost-effective unstructured data storage solution for handling large quantities of information outside of a database. It’s not a new technology, but it is one whose time has come.
The explosive growth of unstructured data, ranging from medical images to content for online catalogs, has become a significant burden for many organizations. Regulatory compliance mandates, corporate data governance and business intelligence initiatives increasingly require retention of this data for long periods of time with the caveat that it can be accessed when needed.
The problem is that traditional network-attached storage (NAS) — whether employing file-level or block storage — can’t cost-effectively store what are often hundreds terabytes, even petabytes, of data. Object storage can. While there are a number of object storage services on the market, few can match the benefits offered by Peak 10® Object Storage.
How It Works
Before discussing Peak 10 Object Storage, here’s a brief overview of how object storage works in general.
Traditional storage systems organize data in files or blocks while object storage works with variable-sized data containers, known as objects. Each object contains the data; metadata, which is contextual information such as policies for retention and deletion; and a unique ID. The objects are stored in buckets in a flat address space called a storage pool. Because the objects aren’t restricted by a hierarchical file structure, the number of object IDs can be increased for almost limitless scalability.
Using an HTTP-based REST API, applications can quickly retrieve the right data through the object’s ID or by querying the metadata using simple commands such as “get,” “put” and “delete.” This eliminates the need to dive into deep file structures.
For resiliency, objects created on one node are copied across one or more data centers. If one node fails, the data can still be made available, typically without any effect on the application or end user. The cost is usually pennies per gigabyte.
Peak 10 Differentiators
As mentioned in the overview of object storage, one of its key attributes is resiliency. Most vendors achieve this by replicating data across multiple data centers, charging for storage at each facility as well as for transferring data between them. This is one of the areas where Peak 10 Object Storage stands out.
Peak 10 strengthens the resiliency of its object storage by erasure coding and spreading data across three geographically diverse, enterprise-class data centers — all for a single base price and with no additional charge for data transfers between sites. Peak 10’s aggressive “pay-as-you-go” pricing also includes lower cost bandwidth compared to leading competitors, as well as 24/7/365 technical support. Many companies would charge two to three times what Peak 10 does for comparable cloud enabled storage across three data centers. Peak 10 also doesn’t charge API transaction fees.
In addition, Peak 10’s ISO/IEC 27001:2013 certified data centers are located more than 200 miles apart and in a configuration to help ensure safety from region-specific natural disasters. As such, Peak 10 Object Storage can also be used as part of disaster recovery plan. While other vendors offer replication, there is often an additional charge if it is across multiple data centers or regions. Charges can come in the form of additional features, storage, or bandwidth. Some only offer replication within a single data center.
While object storage is often for long-term or archival storage, Peak 10 nonetheless ensures data is readily accessible when needed with an SLA-backed availability of 99.999%. Data can be accessed from any device, anywhere and anytime over the Internet via an S3-compatiable API. Data is also always encrypted in transit and at rest with customers always maintaining sole control of their keys.
In addition, Peak 10 Object Storage is compatible with Commvault® applications and NetApp AltaVault devices for backup or archive storage.
Peak 10 Storage for All Needs
Object storage is designed for storing non-transactional data, making it ideal for use cases such as backup and archiving, and static content distribution. However, it may not meet an organization’s complete data storage needs. Peak 10 covers that with higher performance storage options that can be used with object storage to create a multi-tiered storage strategy.
When it comes to unstructured data storage, the benefits of object storage speak for themselves. For storing unstructured data even more cost effectively, efficiently and with greater peace of mind, the benefits of Peak 10 Object Storage speak a little louder. Watch for future blogs that provide more detail about some of the distinctive features of Peak 10 Object Storage.