I last spoke with Chief Technology Officer, Cameron Bahar, a few weeks ago. (Check out our previous conversation here.) When we met up again more recently, we continued our discussion on data protection, especially as it relates to scale-out and scale-up data storage. Listen to our full conversation in Episode 43: CTO Series: The Veritas Advantage in Data Protection.
Scale-Out vs. Scale-Up
I asked Cameron which approach is better for the purpose of data protection: scale-out or scale-up? He explained that the answer depends on what you’re trying to achieve. “If you’re doing analytics or running applications on data that’s easily partitionable, then scale-out architectures tend to do better than scale-up systems,” he says.
But at the same time, scale-out systems are often more complex to operate. They also tend to be more expensive. “With a scale-out system, you’re adding CPU and memory as you go, which comes with a cost,” he notes. “The higher cost and complexity is worth it if you’re going to actively mine and run analytics on that data. But if not, you’re leaving money on the table.”
“I would argue that mid-range scale-up systems are adequate for the majority of customers if the secondary storage is mostly acting as a backup insurance policy,” he says. “It would also be more cost-effective.”
Why Many Vendors Focus Only on Scale-Out
Today, many vendors are heavily marketing scale-out systems, even for secondary storage use cases. I asked Cameron why all the focus on scale-out solutions if they’re not the best fit for many users. “Many vendors have one product they're trying to ship and stabilize,” Cameron says. “So they apply that one solution to every use case scenario.”
This is partially driven by the vendors’ desire to compete for larger volume clients. “They can’t win larger customers with a small system,” Cameron notes. “So they’re forced to build large scale-out systems, but is that the right strategy for every customer? Not necessarily.”
“We Don’t have an Attachment to a Specific Architecture”
At Veritas, we have deep roots in both cloud and distributed systems. “We can already scale to hundreds of thousands of servers. We do it everyday,” says Cameron. “This isn’t something that’s technically new for us. Our scale-out platforms have been around for 20 years; we have a lot of DNA in this space.”
As a result, we can choose from a wide variety of solutions, depending on the size of the customer and the workloads involved. “We don't have an attachment to one particular architecture,” says Cameron. “We can offer a flavor of offerings and solutions to our customers that will meet their needs—as opposed to a one-size-fits-all solution.”
Our Diverse Portfolio of Scalable Solutions
As a result of our decades of experience, we have a diverse portfolio of scalable solutions including:
Bring Us Your Use Case!
No matter what your data protection challenge, we have a way to solve for it. As Cameron explains: “We’ll optimize the solutions that make sense for your business, whether your priorities are reducing costs, improving scalability, and/or improving performance. It could be a distributed system or in the cloud — it doesn't matter. Whatever your needs, we have service offerings to help you satisfy them.”
He continues, “We look forward to hearing from our customers about their unique needs, because we want to be customer-driven and responsive.”
Learn More about Veritas Scalable Solutions
The best way to learn more is to visit veritas.com. I also recommend checking out our online community, Veritas Open eXchange. There’s lots of great information there about our products and solutions including blogs, community forums, and much more.
Listen to the Podcast
Visit iTunes to listen to this ‘Voice of Veritas’ podcast, “Episode 43: CTO Series: The Veritas Advantage in Data Protection” for more of our discussion.
Subscribe to the Voice of Veritas podcast to get new episodes as they’re released.
You must be a registered user to add a comment. If you've already registered, sign in. Otherwise, register and sign in.