Blockchain technology, a distributed and secure database system, has been making significant advances since it first started gaining popularity a few years ago. It started out in the cryptocurrency world, but is also finding considerable usage in banking and finance, as well as asset settlement. But, can blockchain also be used for other use cases as well, such as data storage?
This may seem counter-intuitive due to a long-held assumption: data needs to be stored centrally to be trusted. But how can an organization keep all its potentially valuable data and make it accessible quickly?
DATAVERSITY® recently sat down with Danny Allan, Veeam’s VP of Product Strategy, to discuss this important question, along with how the growth of cloud technologies also are playing a key role in this equation. Allan began by stating that “blockchain has come on the scene with cryptocurrencies, cross-border payments without banks, and benefits to logistics companies, who can track all their goods and supplies everywhere.” The technology has been growing in so many ways; for example, Singapore recently launched a prototype using blockchain for delivery and payment, and that is just one instance. But, with all these growing use cases and subsequent data volumes, what are enterprises to do?
Allan observed: “More and more data is being collected everywhere, both on premises and in a multi-cloud environment.” Just the sheer size of data needed to fuel the Internet of Things (IoT) is staggering. Raw data exceeds volume increases of over 50 percent per year. Technologies have improved so that physical servers and virtual machines are not the only options. Companies can keep their data off-location and put together hybrid clouds to network data.
While hybrid solutions have increased business productivity and efficiency, two challenges remain. As Allan stated, “latency and bandwidth” continue to slow down accessing data from storage. Data networks can add up to be quite complex. It is one thing to get a few data conduits up and running – for example, to develop a machine learning (ML) algorithm or two for a use case. It is quite another to keep all the data storage networks up and running for operations, especially to maintain the flexibility and deployment advantages of the cloud, while keeping sensitive information on premise.
Data in these different storage options may need to be accessed simultaneously. Keeping all this data flowing as data velocity and bandwidth reach their ceilings poses additional difficulty, said Allan. How can blockchain technologies provide a good solution?
A New Paradigm: Shared Data Storage
Allan remarked that there is an exciting new data storage paradigm that makes use of blockchain that could revolutionize storage systems and data evolution. After experiencing how disk storage has transformed to flash, he has seen benefits in software performance. While some may continue advancing storage system technology to solve this, he sees this approach as limiting. Allan said:
“If you buy a new storage system of a certain size, you limit within that storage area network (SAN) bucket, e.g., the number of storage devices and the architecture that defines it and the backup structure. To expand beyond the SAN capabilities becomes complex and difficult to manage. How do you spread your workloads and do migrations, etc.?”
New ways emerge to store data, overtaking the older technology and architecture around it, mused Allan. The cloud provides another tool in a company’s toolbox, in extending the existing storage environment. But even cloud providers face limits on what they have on premise. A new model is necessary, believes Allan, that does not depend on developing a new technology.
Toward that end, he compared Data Management to a power grid. Allan said, “say everyone in my community has a solar panel. We can share electricity and have a local marketplace for power.” This kind of contribution toward power has gained traction in nations like Rwanda, to improve electricity access for minimal costs. “How does this paradigm play out in the data storage marketplace?” asked Allan. He suggested:
“Rather than sharing power across a community, we can share data storage. Data is exploding, doubling every two years – especially where a single customer needs to copy data. In data protection there is a fundamental rule: three copies of data, two different media types, and one off-site. If we continue to protect all data, then we will make a massive data footprint. Here we leverage community storage grids, distributed storage across a distributed environment.”
Allan sees blockchain as spurring these common data storage shares forward by addressing concerns about data tampering and ownership. As he put it, data owners need to know that “it is [their] data and we are buying and selling excess capacity locally, rather than at a cloud level.” Blockchain is an open ledger, resistant to data manipulation, and a public record promises to address these worries and energize economies trading space in the cloud for data storage.
But this paradigm integrating blockchain, cloud, and data storage is in the early stages. Allan believes it may take a solid ten years to see community grids in the mainstream, just as the cloud hit its stride a decade after a new paradigm was introduced in 2006 when users could access software and files over the web.
Additional problems to solve include the data storage diversity in most companies, “where they end up with ten different storage systems acting and managing data differently and requiring different experts and administrators to run them.” This leads to high complexity and cost. Danny Allan’s vision is seeing fruition through Veeam.
Ready for the Next Data Storage Evolution
Veeam concentrates its data storage product in cloud management, as companies encompass multi-clouds and hybrid clouds in their data storage configurations. Its core product, Veeam Availability Suite, makes a shared data storage network simplified, reliable, and flexible.
Danny Allan has guided Veeam’s solution to be “a single management platform to protect and manage data.” The company works on making its systems run all the time, so it is dependable in delivering services. Lastly, Veeam believes its customers should be able store data in different environments. Allan added, “one size does not fit all. Even within an organization, two or three or four different infrastructures may be needed to deliver services to the company.”
To illustrate: Veeam Availability Suite 9.5 allows users to “tier data to any of the public clouds.” Veeam has 22,500 cloud providers offering its platform. Businesses using Veeam have a great diversity of storage systems. Danny Allan spotlighted Veeam’s ability to:
“Aggregate data systems, whether you have network attached storage or direct attached storage or a network file system. Veeam puts all these into a single scale out storage system leading to a simple transparent way of extending storage in the cloud. With Veeam, users just say ‘this is the data I want, bring it back,’ and it doesn’t matter where it is on any one of ten storage systems or up in the cloud. Veeam instantly recovers it.”
With the maturity of cloud technologies and providers, the cloud has become more secure. Allan noted that “cloud organizations’ lifeblood depends upon protecting data. That has made cloud providers more hyper-vigilant in providing security measures, compliance certificates, and proof points to even exceed those of a typical organization.”
Companies find Veeam’s solution more attractive than increasing the number of on-hand storage blocks – large physical segments that can be assigned to store data. Instead, stored data lives in a wide variety of locations, accessed dependably, with better performance.
Sharing Data Storage is the Present and Future
Businesses face tremendous data storage complexity. As Allan stated, there are so many options combining different types and technologies, a company can easily get buried with work and choices.
Block and physical storage increase expense for upkeep and are confined by technology and performance needs. Hybrid clouds resolve some of this by distributing data, reducing storage administration costs, and enhancing performance.
But with so many different cloud combinations and options, organizations need to simply know where it is and how to retrieve it, in addition to improving their data storage latency. To do this effectively, companies will need to share their data storage across a community network, making a wide area data storage network essential to handle huge, streaming amounts of Big Data.
Image used under license from Shutterstock.com