Storage

Why your next storage solution may depend on blockchain

Storage has long been centralized and inherently insecure. Blockchain may be able to change that.

storage.jpg
Image: iStockphoto.com/cybrain

In the traditional data center, enterprises managed their storage needs by adding racks of servers. In this world, storage was centralized, making it both inefficient and insecure. If a new open source project has its way, however, storage will be blockchain-based, offering end-to-end encrypted—and highly distributed—storage. Such storage would not be distributed to a few data centers, but rather distributed across hundreds or thousands (or millions?) of nodes, yet still offering superior performance because of its peer-to-peer nature.

Really?

Yes, really. At least, if Storj turns out to work as advertised. In a conversation with John Quinn and James Prestwich of Storj Labs, primary developer of the Storj project, they promised a glorious storage future that costs less, is more secure, and performs better than other data center or cloud storage alternatives. But will anyone use it?

The blockchain is here

Blockchain has become an increasingly hot topic in recent months due to its ability to deliver distributed security. Its impact, however, has mostly evaded storage. The developers at Storj, an open source object store similar to AWS S3 or Microsoft Azure Blob Storage, aim to change that.

SEE: 10 storage trends to watch in 2016

Storj (pronounced like "storage") hopes to make object storage easier to use through intuitive tooling and documentation, a modern API, and an open source, try-before-you-buy approach. But really, much of the magic derives from blockchain. Think of Storj as a distributed cloud storage network, suitable for static content today but with aims to expand far beyond this in the future.

This blockchain-based decentralization allows developers to store data in a secure, performant, and inexpensive way, spreading it across many nodes. As for security, a blockchain approach means that each file is shredded, encrypted with your own key, and spread across the network until you are ready to use it again. On retrieval, files are decrypted and reassembled seamlessly on the fly.

By definition, Storj doesn't need to build or operate data centers so capital expenditures are minimal, resulting in lower storage costs. (Quinn told me that prices are half those from other cloud competitors like Amazon Web Services.) Storj's peer-to-peer architecture also enables parallel downloading of data from these disparate shards, allowing Storj to saturate your download connection rather than relying on the inefficiencies of a single-source download.

That's the good news. The bad news is that Storj is still small, with just 5,000 API users as of August 2016 and seven petabytes of storage capacity. The silver lining, however, is that both of these are up 100% in the last month alone. The growth trajectory for Storj is very positive.

Standing out in the storage crowd

More, however, is needed. As mentioned, the big question is whether Storj Labs can separate itself from the pack. As one storage industry executive told me, while he "loves the Storj tech," there are "so many storage companies out there at the moment" in a "horribly crowded market" that "cool tech" simply won't be enough.

Storj hopes that its combination of ease-of-use and rock-bottom prices will appeal to developers, not to mention the option to sell excess storage capacity on their hardware to earn currency similar to Bitcoin. At the same time, Storj pitches security and performance to the enterprises that employ these developers.

SEE: Why Cisco's data center strategy depends on simplifying storage

Storj Labs, in turn, makes the Storj project much easier to consume by abstracting away the complications of running a node on the Storj Network. The company handles bridging onto the network, payments, audits, contract negotiations, shard locations, etc. so that the user can have a smooth and simple object storage experience. As the Storj Network grows, the company will be able to pull insights from its large data set of performance history of network nodes to optimize network use.

As to whether all of this will be enough, well, that's a big "maybe." The technology approach is cool and the potential to improve cost, security, and performance of storage is even cooler. I suspect we won't see mass adoption, however, until one of the big cloud providers joins the blockchain storage market, perhaps by acquiring Storj.

Also see

About Matt Asay

Matt Asay is a veteran technology columnist who has written for CNET, ReadWrite, and other tech media. Asay has also held a variety of executive roles with leading mobile and big data software companies.

Editor's Picks