Avinash Lakshman, CEO and Founder of Hedvig
Data&StorageAsean: Why are traditional hardware storage vendors seeing revenues decline? Is SDS in part responsible for this?
Avinash: There are many, converging trends. First and foremost, enterprises are learning that traditional storage simply cannot keep pace with ever-changing business requirements. Traditional storage is simply not flexible enough. A great example is Docker, which stormed onto the scene just a few years ago. Traditional storage is not designed for the distributed nature of containers, so it’s not a good fit for Docker. As a result, organizations are abandoning traditional storage as they look at more modern application architectures. This is compounded by the rise and maturation of software-defined storage. SDS solutions now offer parity in terms of feature set, yet offer the flexibility and scalability that traditional storage cannot match. Many larger enterprises are actually creating “software-defined mandates” where they will no longer buy server, storage, and network technology that is tied to hardware. It needs to be decoupled to maximize flexibility.
Data&StorageAsean: Does moving to SDS storage mean throwing away existing storage hardware investment?
Avinash: Most companies will not throw away their existing storage hardware investment when moving to software-defined storage. In many cases, they will have spent millions on those assets. Instead, most companies look to SDS when their traditional storage solutions are near a refresh. They use this as an inflection point to look at additional solutions. However, many SDS solutions can actually reuse the traditional storage array as backend capacity. This only is recommended for a narrower set of workloads like archiving, but does provide a way of extending the life of that original storage asset but still providing an SDS “front end” that is more flexible.
Data&StorageAsean: Is SDS already being overtaken by cloud storage?
Avinash: It depends on your definition of cloud storage. In many cases, software-defined storage integrates with cloud storage. Many customers will not put 100% of their data in a public cloud. It’s either too costly, too slow, or they are prohibited based on data sovereignty laws. Because SDS can be a truly software-only solution, it can be run in the public cloud. In this case, the storage is actually run in a hybrid manner -- stretching across both private and public cloud sites. This actually expands the definition of cloud storage and solves a broader set of data challenge facing modern enterprises.
Data&StorageAsean: What are the key drivers that will drive people to implement SDS?
Avinash: Cost savings and flexibility are the two primary benefits driving organizations to adopt software-defined storage. According to Forrester, data is growing 10x faster than storage budgets. The traditional storage infrastructure simply does not provide viable economics given this exponential growth rate. SDS uses less expensive commodity hardware, ensures the latest in hardware innovations can be accommodated immediately, and provides operational savings through simpler provisioning and ongoing maintenance. But for many organizations, it’s flexibility that truly drives SDS deployments. You can accommodate use cases like Hadoop, Docker, sensor and IoT networks, operational analytics where traditional storage was never an option. What it really comes down to is a simple question: “Am I looking to cut costs by 60% or adopt new cloud applications?” If the answer is yes to either of these, then SDS makes sense.
Data&StorageAsean: What is unique about your own SDS offering?
Avinash: The Hedvig Distributed Storage Platform is the only software-defined storage solution built on a true distributed system. This means organizations of any size can get started and invest in a storage solution that will scale indefinitely. Additional, Hedvig has three technical differentiations:
1. Support for any compute environment. Today, Hedvig supports any hypervisor solution (VMware vSphere, Microsoft Hyper-V, KVM, Xen), any container solution (Docker, CoreOS), and bare metal operating system (Linux, Microsoft Windows), and cloud platforms like OpenStack, Mesos, and Kubernetes. What this means is any organization can be assured they can use Hedvig as their single storage platform regardless of the compute architecture above.
2. Application-specific storage services. Hedvig also provides the ability to have each application set 12 unique storage policies. IT ops and DevOps can tailor the storage for each application and workload. This includes features like deduplication, compression, snapshots, cloning, tiering, and caching. Furthermore, Hedvig provides a full set of RESTful APIs so that all application-specific storage policies can be automated in any cloud or orchestration framework that customer chooses.
3. Hybrid cloud ready. Hedvig not only provides the ability to run in both private and public clouds, but also provides the ability to set unique data location policies across a hybrid cloud architecture. Coupled with the application-specific storage services, this means each app can have it’s own unique high availability architecture. This also helps solve geolocation and geofencing challenges for companies governed by strict data sovereignty laws