Software-Defined Storage, SLOs, and the Protection Storage Architecture — The Story Continues, Part I

Window to future

“Software-defined [fill in the blank]” has already made a huge industry impact. Admittedly, there’s a lot of buzz around the term, but it’s no small feat to triumph over “big data” so quickly. And as with many over-hyped terms, there is some real substantive change behind it. So, overlooking the obvious cynical jokes (e.g., thank goodness we don’t need that pesky hardware to store data anymore!) what will software-defined storage mean to data protection and the teams that provide it?

Software-Defined Storage—It’s about SLOs

Software-defined storage (SDS) is about delivering service levels to your applications with your storage assets (compute and media).

Too often, people assume  SDS means that it’s finally time to build a storage system out of software to run on commodity hardware. They’re about 15 years too late; it’s already happened. What makes a Data Domain different from a VMAX? The software. So, if so much of the value in storage has already moved to software, why the noise about software-defined storage? One word… simplicity.

Managing storage environments is excruciating. Each type of array has a unique set of functionality—a storage personality—that must be managed differently and on dedicated islands of hardware.

Customers like the distinct functionality (e.g., Data Domain’s space optimization and data durability, VMAX’s predictable performance and availability, Isilon’s scale) but hate the operational complexity. In some cases, operational simplicity wins and customers select a one-size-fits-all “good enough” storage solution. In other cases, they grind through the complexity. In either case, they have to settle.

Software-defined storage promises to simplify storage management by delivering service level objectives across the various storage systems. Instead of having to be an expert on the intricacies of the VMAX, VNX, and Isilon—imagine a software layer that selects and configures the appropriate storage personality for your workload.

That’s why software-defined storage is so exciting—you can have your cake (all the unique storage functionality) and eat it too (none of the agonizing management complexity). While most customers immediately focus on service levels objectives like response time, throughput, and availability, that’s not where you’ll find the maximum value in software-defined storage. It’s in protection. And it can help you achieve a whole new level of IT productivity.

Software-Defined Storage—It’s about Protection SLOs

Protection has created the greatest amount of complexity in storage environments. While each storage array has a different personality, each also has a well-established set of performance and availability capabilities. In other words, most people know the difference between a VMAX and an Isilon. However, each array offers multiple native protection methods (e.g., SRDF, TimeFinder clones, RecoverPoint) in addition to traditional (e.g. backup client) and next-generation (hypervisor or application-level) backup techniques. The complexity multiplier is staggering. If storage management is excruciating, protection management is soul crushing; it’s impossible to make the right choice.

How can software-defined storage address the protection management challenges?

  • First, customers need to extend their SLO expectations to include Recovery Point Objective (RPO), Recovery Time Objective (RTO), retention, and recovery resiliency (e.g., geography, number of copies, etc.).
  • Second, they need to select a protection storage personality that integrates with the data movement and control mechanisms from their key data sources (e.g., primary storage).
  • Third, they need to connect the protection movement to the application.
  • Finally, they need to demand data management software that can span all the different protection mechanisms. If this sounds familiar… it should. The protection storage architecture recognizes that in the “software-defined” world, storage will take a much more prominent role in protection than it has.

The ultimate goal for software-defined storage is to enable a customer to provision protected storage to meet their SLOs.

The Future Won’t Look Like the Past

While the software-defined storage battles currently are more sound and fury than substance (e.g., a “one-size-fits-all” storage OS is “software defined” in the same way that Michael Bay’s films are “diverse”). Ignore the petty debates and focus on the substance—the storage market has become a breakneck race to see who can deliver SLO-based storage provisioning and protection.

Software-defined storage will have profound implications on the roles of the backup and storage administrators and how companies build (and purchase) protection solutions, and it lays the groundwork for the next massive shift in our industry—from data protection to data management. If you thought my last series was long… wait until you see this one.

Stephen Manley

Stephen Manley

CTO, Data Protection and Availability Division
Over the past 15 years at both EMC and NetApp, I have traveled the world, helping solve backup and recovery challenges - one customer at a time (clearly, I need to optimize my travel arrangements!). My professional mission is to transform data protection so that it accelerates customers’ businesses. I have a passion for helping engineers pursue technical career path(without becoming managers), telling stories about life on the road and NDMP (yes, that’s NDMP).

One thought on “Software-Defined Storage, SLOs, and the Protection Storage Architecture — The Story Continues, Part I

  1. Pingback: Software Defined Storage Delivers New Level of IT Productivity - Federal Technology Insider

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>