BETA
This is a BETA experience. You may opt-out by clicking here

More From Forbes

Edit Story

Asigra Plays Forward In Cloud Backup Market

Following
This article is more than 8 years old.

Storage technology is boring, at least that’s what people think. Actually it’s kind of not, but it can be hard to see why. Without storage we would have no cloud. Without cloud storage at the backend, your tablet, ‘phablet’ and smartphone wouldn’t handle a whole lot. Without cloud-based storage, backup, disaster recovery and archiving with intelligent caching your computer probably wouldn’t get any of the data it needs at the right time for the right app in the right place.

As cloud becomes even more prevalent and ubiquitous we will see potential bottlenecks of data service loom up more frequently – and these are not bottlenecks caused by bandwidth issues, so what functions and intelligence is running in the server room to keep us all online?

Toronto-based Asigra is a company on a mission to become better known in the cloud storage, backup and recovery market. Its direct competitors range from Barracuda Networks to Commvault – and then onward (at a higher level) to IBM and Symantec etc. In basic terms, service providers use white-label versions of Asigra’s software to sell online backup services that protect data in the datacenter on physical or virtual servers, enterprise databases and applications.

“The Asigra Converged Data Protection Appliance family disintermediates storage software stacks from expensive hardware, removing the cost and complexity of secondary storage systems,” said Eran Farajun, executive vice president, Asigra. His comments coming on the back of the launch of the Asigra Converged Data Protection Appliance for Managed Service Providers (MSPs) -- developed to enable managed service providers/cloud service providers to enter the cloud backup market with cloud backup/DR services to achieve monthly recurring revenue streams.

“We have engineered-out the complications of storage system integration, troubleshooting, fixes, patches, installation, monitoring, maintenance and the requirements for skilled hardware and operating system staff. This is a backup/DR industry first that tackles the inflexible designs and costs of traditional storage systems, introducing a new approach to highly scalable and performant data protection storage. The approach enables service providers who want to offer cloud backup services for the first time a quick and easy way to get to market in days.”

How far do we need cloud backup?

The kind of backup services we are looking at here will protect users’ data on:

• desktops and laptops,

• smartphones and tablets and phablets,

• cloud SaaS-based applications like Microsoft Office 365, Google Apps, Salesforce.com,

• also on IaaS-based platforms like AWS or Microsoft Azure.

Powered by Asigra, trademark

There is great opportunity here i.e. backup infrastructure costs continue to rise, putting significant financial pressure on IT service providers seeking to grow their operations. So technology that brings this kind of competency forward will surely do well, right?

Yes and no. Asigra is playing in a difficult market. Its brand name is not so well known, a fact that the firm openly recognizes (but it is putting a lot of effort into rectifying). Secondly, the white-labelling of its product leaves it even deeper down as an unknown ‘ingredient’ technology and nobody would pretend that Asigra is the next NutraSweet or Intel Inside processor i.e. people don’t buy it just because it is there internally as a constituent component. Farajun has commented that he would like to see more partners label their backups as “powered by Asigra” to create end user engagement awareness, but it’s still early days.

Asigra’s technology is smart and benefits from being able to optimize the way backup data is handled. To explain – a lot of backup technology moves entire file chunks left and right to backup the entire system in usage. Asigra’s software is capable of detecting only those blocks of data that have been changed or updated and therefore need backing up. When bandwidth becomes limited (and cloud bandwidth is always under pressure), this is the kind of thing that could decide whether data moves to serve your app works in milliseconds or an ‘unacceptable’ several seconds.

Oracle partner position

Bob Handlin, strategic director for Oracle ZS storage spoke at the recent Asigra Partner Summit 2015 event in Toronto. His firm sits in an OEM partnership relationship with Asigra and he says that we now have the chance to use cloud storage technology to change the way our systems operate.

“People talk about all these ‘topline’ changes going on with social and mobile, but it still comes down to the technologists [and the stuff that we make and work with] to make it all happen and connect the knee bone to the ankle bone,” said Handlin.

We have been used to building cloud systems that keep compute and storage relatively tightly coupled…this means that bottlenecks can result if one disk is getting blitzed with writes and reads at the same time. Storage ends up bottlenecking on a single drive. This is close to noisy neighbour problems where one ‘chatty machine’ makes the others near it go slow.

“These traditional storage appliances probably work just fine still, for now,” said Handlin, but he suggests that times are about to get more demanding.

I/O Blender effect

If you’ve found cloud storage interesting enough to get this far, then use this as your chance to learn about the so-called I/O Blender effect. TechTarget defines this as a phenomenon in virtualized environments that degrades storage performance.  The I/O blender effect occurs when multiple virtual machines (VMs) all send their input/output (I/O) streams to a hypervisor for processing.

But in the face of danger from the I/O Blender, we’re now looking to disaggregate compute from storage and virtualize the top layer connections above the disk in the cloud itself. The idea is that all roads lead to one disk so that the traffic to the cloud is better. Caching algorithms often decide which data is ‘hot’ and keep it more available in terms of disk location. The idea is to be able to put cold data on less expensive storage. Being able to put analytics down on storage is important… being able to splice and dice the storage (right down to the filename) and find out where the bottlenecks is even more important.

Cloud storage is going through a heady (almost sexy?) revolution but unfortunately it’s only the geeks that operate the mechanics that know about it. We will all benefit from better backend services, so it’s time to hug a server if you do pass one this week.

Editorial disclosure: Asigra paid for a proportion of Adrian Bridgwater's travel expenses to attend its partner summit event in Toronto this June.

Follow me on Twitter or LinkedIn