The Tech Talks Daily Podcast is a fantastic daily listen, where veteran tech writer Neil C. Hughes interviews leaders and experts on how technology is transforming their business. Back in April, our very own CEO, Ash Munshi, joined Neil to talk about analytics stack performance and the Pepperdata solution.
The episode, titled Removing the Blindfold to Control Cloud Spend, explored how effective analytics stack performance plays a key role in cloud cost management and prevents enterprise cloud bills from spiraling out of control. In the first half of the episode, Ash shared how Pepperdata approaches two key concepts: observability and automated tuning.
On observability, Ash explained: “Big data is different from your standard computing paradigm, where you know a single machine, or a small set of machines, solves a single problem. Here, you have hundreds or thousands of machines solving a single problem. So the problem complexity is very, very high.”
“The amount of resources and the cost on this is very, very large,” Ash went on, “both in terms of compute resource storage, as well as in terms of IO as well, because there’s so much data. So, whereas these problems get very complex, when you have bugs, these can eat up a lot of resources, which translated means, eats up a lot of money very, very fast. And I’m talking about millions of dollars.”
“So the ability to be able to understand what’s going wrong, and pinpoint this very quickly, and be able to get to the root cause so that you can fix it quickly, translates to a massive amount of savings in terms of the actual cost to the end customer.”
Ash then went on to cover automated tuning: “What happens is a lot of these systems have applications written by developers who assume that they’re going to get a lot of resources, or need a lot of resources, for their applications to run. But in reality, what happens on these machines as all these applications are running, is that they may use a lot fewer resources than originally anticipated at any given point in time.”
“As a result, the machines have idle resources available. What we do is help the schedulers on these machines understand their idle resources, and then utilize them so that the machines are utilized at maximum efficiency. That translates to millions of dollars of savings.”
Observability and automated tuning, Ash explained, translate to efficient cloud cost management, “massive ROI, and as a result, we have some of the largest companies in the world as customers of ours, including several companies in the Fortune 10.” Ash and Neil touched on how the pressures of big data are only set to increase as more and more companies move toward a hybrid multi-cloud strategy.
Neil asked Ash if he had any examples of customer success stories that he could mention. In response, Ash summarized how Citibank and Comcast have seen success with analytics stack performance solutions provided by Pepperdata:
“We are globally deployed at Citibank. Citibank uses us to manage, and to perform, all of their big data infrastructure. We also help Comcast get more capacity out of their infrastructure. So all of the Comcast boxes that you see out there, for doing your cable TV and your internet, are all monitored. All that data goes back to a central set of servers, and then they analyze those servers to figure out whether there are any patterns to failures, issues that are happening, and things of that sort. All of those servers use our technology to be able to maximize throughput, and Comcast has gotten a 30 to 50% improvement in performance as a result of that.”
On the future of Pepperdata, Ash gave a powerful description of how the company is looking to continue to improve: “What we’re doing is we’re listening to our customers. We’re very, very careful about understanding what it is that they need, whether it’s across systems like Snowflake and RedShift and BigQuery, or streaming systems like Kafka, or other technologies that are coming out. Our customers are looking to us to help them figure out how to get those technologies embedded in their architecture, and then be able to make sure that they’re using those technologies optimally. We’re lockstep with our customers; we listen, we learn, and we move forward with them.”
Besides analytics stack performance, Ash and Neil discussed a number of other interesting topics, including Ash’s personal history, how Pepperdata and Gartner have interacted, and more. Listen to the full episode here.