With contribution from:
Mary Meduski: President and CFO of TierPoint
At SxSW this year, connectivity is more important than ever before as AI, IoT, and edge computing play a bigger role in delivering high-quality video, music, and gaming to consumers each day. With this in mind, film, music and interactive content stand to suffer if connectivity is threatened.
In the panel on March 9 at 2pm CT, How Connectivity Will Control Everything We Know, Mary Meduski, President and CFO of TierPoint, will offer her three decades’ worth of experience in the media, telecom and technology industries and point to how the industries have changed in that time due to how important connectivity has become. In the meantime, however, I sat down with Mary to discuss the role connectivity plays in business today, what it means from one company to the next, and how her work at TierPoint positions her as an expert in her field.
1. Looking back on telecommunications and technology from the early 2000s to today, what developments have truly surprised you?
Mary Meduski: One of the most surprising developments of the last two decades has been the insatiable consumer demand for more and richer content. Consumer preferences have evolved quickly: from text to photo to video – from standard definition video to high-def to 4K. Meanwhile, online behaviors have evolved from the point at which downloading content was a primary focus, to today, where uploading content (often, user-generated content) is an increasingly ubiquitous activity. And that need for uploading – for sharing content between end users – has, in turn, been driven by the proliferation of online socialization tools and platforms.
A related trend, which I believe surprised many in the industry, was the rapid evolution of wireless connectivity. This, too, was an outgrowth of consumer demand, to be always-on, always-connected – from any device, in any place, at any time.
Altogether, this evolving, multi-layered, multi-directional consumer demand – perhaps even more so than business demand – has supported a range of technological advances; from bigger and faster internet pipes, including more robust last-mile connections; to more sophisticated and intelligent platforms for content sharing. And yet, every single time the technology advances, consumers find a way to use it up and clamor for more.
2. Edge computing is a major focus right now in the data center industry, as a way to cope with the incredible amount of data. How do you think it will evolve?
Meduski: The “edge” means different things for different companies and different applications.
Historically, the primary IT needs were storing and, when requested, serving data or content. These were largely one-way transactions, and the edge didn’t have to be very close for them to work. Even today, that’s still true for certain businesses, including popular video streaming services. For such companies, the model of massive regional data centers in major markets may be sufficient for years to come.
But a different model is required for applications that depend on super-low latency and/or more robust computing capabilities at the edge.
Research suggests that on a variety of highly interactive applications – such as financial trading, gaming services, or social media platforms like Facebook – if consumers are required to wait more than 30 milliseconds for content to load, they’ll get impatient and go elsewhere. And that’s why such businesses are seeking to limit, if not reduce, latency by moving the edge closer and closer to their end consumers, to keep them engaged for extended periods of time.
Next, think about applications that do more than simply store and serve content; they also require significant compute capabilities at the edge. Here, I’m talking about artificial intelligence (AI) and augmented or virtual reality (AR/VR).
Consider, for instance, the compute side of the driverless car. Granted, some compute capability will be housed in the car itself – but not all of it. The driverless car will also require real-time information about traffic patterns and nearby vehicles, coming from multiple sources. This complex, dynamic, interwoven and multi-layered data flow will need to be evaluated to accurately inform the driverless car’s actions. And some of that evaluation or compute capability could reside outside the car, in edge data centers.
3. What are the implications for data center/colocation and cloud service providers? How does the architecture of data centers need to evolve to support edge computing?
Meduski: Data centers in secondary markets will need to serve a more diverse range of customers than they serve today.
We expect mid-market enterprises with headquarters in a certain city will continue seeking out local or otherwise proximate data centers, just as they do today. But there will be new customers, too. Social media, virtual and augmented reality, gaming companies – all of them will look for more computing capabilities, closer to the edge. And this combination of customers could lead to larger, modular data centers. Today’s 15,000 square-foot edge data centers may become 30 to 40,000 square feet to meet the combined demand of legacy and new users.
Edge computing also means these new data centers will need the right carriers, including connectivity from incumbent cable, telco, and other trusted, last-mile providers. They’ll also need to invest in more connectivity to carrier hotels, where top providers converge. And software-defined wide area networks (SD-WAN) – with flexible connectivity on demand – will become more important, allowing data centers and their customers to scale up and scale down as needed.
And that’s the unique space where a company like ours can be of value: Offering facilities that are appropriately sized, secured, and connected.
4. How will TierPoint serve customers with edge compute requirements?
Meduski: TierPoint has invested in a strategic team that understands the requirements of businesses moving to the edge and has built trusted-advisor relationships with decision makers at social, content, OTT and other companies at the forefront of edge deployments.
We’re also rolling out enhanced connectivity strategies – into both our data centers and carrier hotels – to make sure we have the right carriers, including last-mile providers.
We’re also revisiting our design protocols, to make sure we meet the expectations of edge compute customers. As noted, the edge data centers of the near future may need to be bigger than they are today and built more quickly and efficiently, which means working more closely with expert partners, such as Schneider Electric.
The post How Connectivity Will Control Everything We Know: Part 3 appeared first on Schneider Electric Blog.