Virtual Reality May Test Our Digital Infrastructure

July 3, 2018
Virtual reality content may need up to 20 times more storage space than a today’s high-definition video. Will your data center be ready? Here’s a look at how VR may test our digital infrastructure.

Today we continue “Data Driven,” a series of articles examining the volume of data generated by emerging technologies, with a look at the data impact of virtual reality and augmented reality.

Society is quickly progressing toward a future defined by data. Businesses already depend on data to make educated corporate decisions, and consumers engage with apps, devices and services, most of which collect and transmit data.

Emerging technologies play a role in increasing our reliance on data. One of them is virtual reality (VR).

Virtual Reality Requires More Robust Infrastructure

People who analyze the future needs of VR applications say that storing VR data isn’t as straightforward as dealing with traditional kinds of video. That’s because VR depends on 360-degree views that require the video content to keep running while people turn their heads and interact with VR environments.

As a result, VR content may need up to 20 times more storage space than an HD video. Speed has to factor into storage, too, especially since many VR applications support real-time environmental changes and viewpoints served up to users almost instantly. As such, flash storage could gain popularity for storing VR content that’s speedily available.

Experts say that VR requires substantially more bandwidth than standard kinds of video too, up to 6 gigabytes of bandwidth per second. They observe that the fastest solid state drives available to consumers now only handle about 3.5 gigabytes per second, making them incapable of supporting many VR apps.

Expansive Growth Fuels Storage Needs

Researchers project that the market worth of the VR hardware and software industries may rise to more than $25 billion by 2020. As VR becomes more embedded in the mainstream consciousness, the number of devices generating data will go up. That means data centers have to start planning now for how they will accommodate the influx of data.

Attendees at the Gartner data center conference in Las Vegas sampling a virtual tour of a RagingWire data center.

It’s also possible that clients may use VR to get more acquainted with the companies that provide data storage services. For example, customers could get virtual views of data centers from a distance, right down to the temperature in a given location. They might even take virtual tours of data centers before deciding to sign contracts with particular providers. RagingWire Data Centers and Google are among the providers that have created virtual tours to show off their facilities.

Researchers also believe VR will cause a substantial uptick in traffic levels, which could cause concern for representatives of data centers that aren’t prepared for what’s to come. A paper published by Cisco predicts a twentyfold jump in traffic due to VR and augmented reality (AR) applications by 2021.

Contributors clarify that the change is mostly due to the large size of VR applications and associated content. However, they say if companies introduce VR streaming, traffic will show even more of an upward trend over time.

Vreal is one VR-streaming startup with content available for owners of Vive and Rift headsets, but the company plans to expand its offerings to more devices soon.

There’s also LiveLike, which raised $9.6 million in a Series B investment round. It combines sports-related livestreams with a social aspect, allowing users to watch matches as they happen and enjoy them collaboratively with friends who are fellow VR fans.

The presence of these startups gives people plenty of opportunities if they’re considering careers in VR, and proves that the industry at large offers possibilities people didn’t envision while watching even the most futuristic films. Such forward-thinking companies encourage people to explore what VR offers, and as they do, storage requirements for the associated data will go up.

High-Performance Computing May Become More Widely Used

High-performance computing (HPC) is among the suggested methods to equip data centers for VR and AR in a strategic way. HPC increases the processing power and data performance capabilities without increasing rack space.

However, the investment required to convert traditional data centers to those with HPC components is substantial. Adequate space, cooling technology and processors are some of the characteristics that make those facilities feasible, yet some companies underestimate the effort necessary to upgrade older centers and make them HPC-ready. This is a trend we’ve already begun to see with data centers adapting to artificial intelligence workloads, which feature more processing horsepower and often require liquid cooling.

Experts believe it won’t be long before multi-user VR experiences gain momentum. If they do, it’ll be HPC data centers facilitating the seamless content those people expect as they become immersed in the interactive content with other people from around the world.

VR Beyond Entertainment

People can now purchase VR headsets without spending a significant amount of their savings. That increase in access has caused companies and individuals to explore ways to use VR that aren’t just entertaining — and all of them theoretically involve data collection.

For example, workers who engage in potentially dangerous jobs, including welding and mining, receive VR safety training that gives them skills while removing the risk. As participants move through scenarios, they might receive scores or see indications of completing modules. Supervisors can access that data later, but only if it gets stored somewhere first.

Plus, researchers at the University of Rochester combine psychotherapy sessions with VR. Patients use their smartphone to access a VR app that doesn’t require using a headset. They then go through interactive scenarios with a virtual therapist that include ranking their anxiety levels and getting exposed to personalized audio and images.

The content changes based on choices people make while using the VR application, which means the associated storage solution for the VR content must be able to handle dynamic needs and produce the content quickly to avoid latency.

Plus, the team behind the project wants to investigate ways to sync in-app data with patients’ electronic health records. If that happens, data storage principles used for VR would have to be HIPAA compliant.

A particularly unusual but undeniably useful application of VR involves an HTC Vive Tracker attached to a cat wearing a special jacket. The device then allows people to see their pets and even determine whether they’re standing up or lying down as the humans play their VR games.

The goal is for people to notice the animals as they play VR games, thereby giving them the information they need to avoid tripping over the unsuspecting pets, kicking them during gameplay or otherwise frightening them.

Delays Could Be Devastating for Data-Centric Companies

This overview of how VR fits into a data-propelled future highlights how companies cannot afford to overlook storage necessities when they develop VR apps. Failing to remain mindful of current needs and how those requirements may change as user volume increases could mean the VR apps don’t perform as expected and cause frustration for those who try them.

An example: Pokemon Go, which in 2016 presented an unexpected stress test for one of the world’s largest and most sophisticated infrastructures, the Google Cloud Platform. The game’s launch generated 50 times more traffic than anticipated, leaving many gamers frustrated with login failures and inconsistent uptime.

Almost all apps collect data to some extent, even when users minimize the content they provide to external companies. The increasing popularity of VR apps and the specific storage requirements they have mean companies have to take proactive approaches when figuring out how to best handle their data needs.

Procrastination in that regard could make companies less competitive than other entities in the marketplace, forcing them to struggle to attract consistently loyal user bases.

About the Author

Kayla Matthews

Kayla Matthews is a tech journalist and blogger, whose work has appeared on websites such as VentureBeat, MakeUseOf, VICE’s Motherboard, Gear Diary, Inc.com, The Huffington Post, CloudTweaks, and others. Drawing from her interests in technology and its applications to daily life, Matthews writes about the intersection of technology and productivity.

Courtesy of Stream Data Centers

Ready for AI? Not Until You’ve Overcome These 3 Challenges

Supporting AI deployments is easy to say and hard to do. Stuart Lawrence, VP of Product Innovation & Sustainability at Stream Data Centers and Mike Licitra, VP of Solutions Architecture...

White Papers

Thumbnail2
Thumbnail2
Thumbnail2
Thumbnail2
Thumbnail2

The Data Center Edge of Tomorrow: Five Critical Ways Digital Twins and AI Will Impact the Network

Sept. 21, 2023
In our online world, we persistently rely on our ability to connect. In fact, our ability to connect is critical to our work life balance. To meet our evolving and growing connectivity...