“80% of disk drives worldwide are only 20% full,” says Ben Gloub, CEO of innovative cloud provider Storj.
This fact is the very reason Storj exists.
“Storj taps into this existing storage capacity from tens of thousands of nodes worldwide, eliminating the need to build new drives and data centres,” he tells Capacity.
And alongside the obvious sustainability benefits this provides, it allows Storj to offer cloud storage for a tenth of the price of legacy providers. Golub says that the carbons emissions saved by using this model can reach up to 83%.
Storing data securely across tens of thousands of nodes worldwide eliminates a single point of failure, and because of parallelism, reduces congestion. This enables Storj to deliver upload and download speeds that are consistently high compared to players like AWS, Golub claims. The distributed model is also able to meet requirements that cloud users desire, such as speed, reliability, security, global accessibility and low cost.
“However, industries that handle highly sensitive data like healthcare and finance need to meet certain compliance requirements that limit where data can be stored,” Golub says.
This has inspired Storj to launch Storj Select, a new enterprise feature that delivers customisation on which of the thousands of nodes specific data is stored. With Storj Select, customer data is only stored on points of presence which meet customer specified qualifications.
Cloudwave, a provider of healthcare data security solutions, was revealed yesterday as one of Storj Select’s first customers.
“The more than 280 hospitals that we work with across six countries are focused on protecting patients. Our role is to protect their patient data from cyber attacks,” CloudWave said.
“Storj Select's uniquely customisable compliance features, speed and security is a key component of our cloud strategy.”
The idea for Storj Select came directly as a result of feedback from enterprises who were interested in switching to its cloud storage platform, but couldn’t because they needed to meet specific compliance requirements.
The new feature allows Storj to store data only on the storage nodes that meet those qualifications, such as SOC2, a a voluntary compliance standard that governs how data should be protected.
“Storj Select opens the door for us to serve more customers like CloudWave which is now a Storj customer because of the existence of Storj Select,” Golub says.
“As a disruptive startup, we have to be agile to meet our customers’ needs,” he continues. “There are several aspects of our business model and architecture that allow us to bring this type of service online very quickly.”
Firstly, because Storj use existing capacity that’s already spun, powered and cooled in data centres, it doesn’t face the same supply chain challenges of other providers when it comes to sourcing the physical hardware to bring a multi-petabyte storage service online.
Storj were also able to leverage globally distributed facilities that already had SOC2 certifications in place, further reducing the time to market.
“It was a coordinated effort to build the product capabilities, source the supply, and onboard our first enterprise customer, but 90% of what we needed was already in place,” Golub says.
Golub also says that the product capabilities Storj built enables it to be much more dynamic in how it selects and restricts the nodes that will store a particular data set.
“This enabled us to bring a tier of storage online that stores data only in SOC2 Type2 data centres in the US very rapidly,”
How does it work?
“Instead of storing data on a few centralised data centres that are vulnerable to outages and cyber attacks, Storj encrypts and splits data into small pieces including redundancy that are then stored across its global network,” Golub explains.
When a download request is made, only a small subset of the fastest data segments are needed to reconstitute the file. This distributed architecture is designed to protect against outages, ransomware, and data compromise.
Storj Select takes this one step further and enhances data security by ensuring that customer data is only stored on nodes that meet the specific compliance requirements of that customer.
“The distribution of data on the network is critical to performance, durability and security of data stored on the network,” Golub continues.
All of the processes for selecting nodes for storing a particular object, distributing the encrypted pieces of erasure encoded data to nodes, auditing the nodes and repairing data is fully automated.
“Those processes are highly scalable, both horizontally by adding more nodes and resources, as well as vertically by scaling up the resources of a particular node,” he explains.
The Storj Select service does require an initial manual process to set placement rules or geofenced regions and choose where the data in particular buckets or projects will be stored .
But once a placement rule is created and applied, everything after that is fully automated.
“That automation ensures we achieve our eleven 9’s of durability with billions of objects stored on the Storj network in tens of billions of pieces distributed over 10’s of thousands of nodes in over 100 countries,” Golub says.
What’s next for Storj?
“We’re continuing to innovate and adapt Storj to better meet the needs of our growing customer base. Additionally, we have some exciting new integrations, customers, partnerships and sustainability initiatives coming down the pipeline that we’ll be sure to keep you apprised of!” Golub concludes.
