SVG Sit-Down: Spectra Logic’s Hossein ZiaShakeri

Last fall, Spectra Logic unveiled a tape-based deep-storage system that the company believes will have a dramatic impact on the storage and archive market. The ultra-low-cost, highly scalable Black Pearl Deep Storage Appliance relies on Spectra Logic’s RESTful DS3 interface and embedded tape-management software, allowing organizations to store all their data forever for, the company claims, costs as low as pennies per gigabyte.

SVG caught up with Spectra Logic SVP Hossein ZiaShakeri to address how Black Pearl will affect the market, the initial response from sports clients, and where he believes the market as a whole is headed.

Why did it make sense for Spectra Logic to introduce Black Pearl to the market at this time?
We’ve been hearing from the end users about life cycle. They capture the video; then they do something with it for a while when it goes through that project. After that, it is going to go into a larger [archive], and it stays there. Every now and again, you may have to touch it to revalidate it, or, rarely, you may want to bring it in to do something with it, but, for the most part, that data asset is going to sit for a long time. The longevity of the matter is [based on] access to the matter.

So we thought that we could make tape more available to the masses for this [near-line] application, bring it from something that used to be specialized storage, and make it available now to the end user. We wanted to make it easy for the end users to access it. We wanted to simplify the tape solutions. With that said, we wanted to have cost under control [and] support model under control. We wanted to have an environment where it was portable and open.

The platform marks a fairly significant departure from traditional archival workflows. Has it been difficult to convince end users that Black Pearl is the way to go?
The whole issue of migration has always been that nobody wants change. So, when you think about our long-time archive, the first thing that comes to mind is the medium, but it’s really not that. It’s really the application. It’s the stability of the format. We want to touch on all those aspects. There is nothing better than tape that actually controls cost or gives you better cost. We wanted the portability.

We wanted an environment where we actually migrate the data over time but, at the same time, verify the viability and the security of the data. So we thought that, if we put this whole thing together, deep storage makes sense.

What has been the response from the market, and what are some of the elements that that have received positive feedback from customers?
People love the simplicity of it and the fact that they could just access it directly. The cost model, they also like, since it is a lot lower cost. As for the simplicity, the way [end users] look at it is that I give you an object and I’ll get it back when I need it. So tomorrow, if it is optical or holographic or genetic sequencing or whatever, it is all back here completely abstracted.

You mention the lower-cost model. How has the market reacted to that?
I’ll give you a particular case where a post house [was] talking about storage needs for shooting 8K and it was 3D. It was at 30 frames per second, and they were talking about how many terabytes fly by to capture that 25 minutes [of footage]. It is just huge, and cost matters. Whatever solution, it’s going to have to bring cost under control. By simplifying this sentiment, we can have 10¢ a gig or so for deep storage and all that. That really makes it appealing.

And, with 4K and even 8K on the horizon, cost-efficient storage is going to become an even bigger issue, right?
The number of cameras [being used for live sports productions] is also increasing. And there’s all these second-screen outlets you need to feed. You didn’t care about getting every single feed from all 16 cameras at a single football game [in the past], but now you do.

How have you seen the sports market — and the broadcast sector as a whole — warm to the idea of cloud-based storage in recent years?
Everyone that we talked to likes the idea of cloud, but they really didn’t feel comfortable putting their asset in a public cloud. That was another aspect that we were thinking about. If we could provide them with a storage solution that takes care of all that and, at the same time, gives them the flexibilities that cloud-based solutions give them, we thought that that would be something that people would respond favorably to.

The idea of cloud is really not new; cloud is just a different term for it. It’s now in its third or fourth generation, but what people like about the whole idea of cloud is that they can just make it somebody else’s [responsibility] and you just have to pay for it. But as you peel the layers, you can see where the cost lies, and it becomes a whole different issue.

There are cloud providers that use Spectra Logic to take data to the backend. So one way or another, tape plays a role.

While people like the idea of cloud, they don’t necessarily like a couple of aspects of it. One is the cost model. Although public perception is that, because cost is low, if you try to retrieve it, it’s pretty expensive.

But, more important than that, they don’t necessarily feel comfortable putting in their assets, especially the media. But they like the idea that it is easily accessible by the application.

So a lot of the folks are looking at this as a perfect [compromise]. With deep storage, you can actually have it centralized and shared among multiple people because it is connected over the Web. But it is also safe and extremely cost-efficient.