Presented by AWS for Games
Cloud infrastructure accelerates game development cycles, enhances collaboration and strengthens security without compromising creativity. In this VB Spotlight event, learn how Riot Games is leaning on cloud infrastructure to build award-winning games.
Game development is not vastly different from traditional software development, says Alexey Prohorenko director of software engineering at Riot Games, but it poses unique challenges — specifically, managing a vast library of hefty digital assets throughout a high-speed workflow.
“You will have a lot of audio, a lot of pure video motions and graphics, which have to be a part of the build,” Prohorenko says. “So every single time when engineers and developers want to see the result of their work, and run a quick play test, it’s a much more complicated process. How do you piece those together? For developers, it becomes a problem to solve — how to make it more efficient and how to work more effectively and faster. Because the faster you will be able to write on something, the faster you probably will find your problems, and the faster it will produce some good results.”
Prohorenko joined GamesBeat Lead Writer Dean Takahashi and Kevon Mayers, solutions architect at AWS, to talk about the challenges of the game development pipeline — and how cloud migration is the modern solution for game developers under increased pressure to meet tight deadlines in a crowded market.
“Technology, at its best, is where it gets out of the way and it helps to make it easier for you to ultimately achieve your core goal,” Mayers said. “The benefit of AWS and cloud computing in general is the elasticity and the auto scalability. That gives you rapid iteration and being able to provision an instance in a variety of different geographic locations.”
Where cloud has the most impact
Determining where cloud brings the most value is very much dependent on the company and its needs, Prohorenko says, but there are a few places where it has shown particular promise.
“If you’re trying to go for the most ROI, you have better luck looking at the build step,” he says. “You will be able to make [your game] more cost-efficient and you’ll definitely be able to improve the iteration speed. Next is the step where engineers, developers and creative people are working with digital assets, actual code or any other artifacts.”
In smaller companies, especially startups, where employees have to wear a number of hats, there are a number of places cloud can add major advantages, Mayer says, and let developers focus on core value rather than administrative chores.
“What are the procedures or technologies involved with your game development process, and where does that align with who you have working at your organization – and what makes more sense to offload? What am I going to build from scratch, and what am I going to potentially buy, or some kind of hybrid in between?” he explains. “We’ve seen a lot of startups lean into that midpoint, building on open-source tools or technologies, or quick starts, that can give the best of both worlds.”
Drilling down into specific examples, Prohorenko notes that when multiple games are in development, which need to be scaled across platforms, the AWS cloud has been a game changer. For instance, Mac-build infrastructure was a challenge for a long time. AWS’s EC2 instances for Macs lets them keep reusable build infrastructure in the cloud, which can be accessed by multiple games and multiple teams and scaled effectively.
“You actually have to make a build, which is going to go into play testing, or else you have to do it in some kind of controlled replicable environment,” he says. “For us, it’s easy to make instances in the cloud.”
And that’s also an example of where you can remove some of those extra burdens on developers, Mayers adds.
“You say, I don’t want to add a systems administrator hat to my game studio for something like this — with the scalability that can come with EC2,” he explains.
Autoscaling for large projects and workflows
Traditional infrastructure includes a data center, physical hardware to maintain, and the need for up-to-the-moment capacity planning and management. In other words, calculating how many servers are needed, when they’re needed, what kind of performance is required and what happens when more are needed. And then in the next cycle, that means re-evaluation, buying more servers or offloading servers that weren’t necessary, and so on and so on.
AWS’s auto-scaling functionality is a huge value proposition, Mayers says – it removes having to manage these physical or virtual machines with the ability to always iterate rapidly and provision an instance anywhere in the world it’s needed.
“But beyond that, I can just click a button, scale it up in response to events, like an increase of concurrent players,” he adds. “Definitely a benefit for the unpredictable nature of monthly usage, especially live-service games.”
Advances in cloud security
The same security processes and procedures for on-premises infrastructure apply to the cloud, with some nuances. AWS uses what they call the Shared Security Responsibility Model, which maps what’s under the organization’s control versus security as a user of AWS — and what AWS is responsible for, like the physical security of data centers and security patching.
Developers need to consider issues such as encryption at rest and in transit, identity and access management and the principle of least privilege, how to ensure staff is only using AWS services for business functions, change management, onboarding and offboarding and more. AWS offers features such as Security Hub and Control Tower for multi-account environments, Mayers says.
“Ultimately, we give you the tools to help you operate securely, but you as a user of AWS still have to make sure you’re following best practices,” he explains.
In terms of what a game studio should prioritize, security-wise, Prohorenko says there’s no silver bullet. It depends on things such as whether it’s a heavily distributed organization, where assets are being moved and when, how complicated the build processes are, and when and where they need to scale. But there are a number of areas that often require extra attention.
“From my perspective, I would invest first in the build, followed by version control and virtual workstations,” he says. “But I would say if a large organization is going to fix the build processes by moving partially or fully to the cloud, and also figure out the version control, it’s going to be a huge win. Pretty much any organization will be able to see the results immediately.”
For a deep dive into the benefits of cloud migration, including collaboration, technology upgrades and more, plus an in-depth look at insuring that migration is successful out of the gate, watch this webinar free on demand now!
Agenda
- Learn which parts of your game development pipeline benefit the most from cloud adoption
- How cloud-based build farms and virtual workstations are revolutionizing collaboration
- Why dynamic cloud resources slash costs while increasing productivity
- How to keep your game assets and infrastructure fully secure, start to finish
- The pros and cons of moving, building or buying cloud solutions
Presenters
- Alexey Prohorenko, Director of Software Engineering, Riot Games
- Kevon Mayers, Solutions Architect, AWS
- Dean Takahashi, Lead Writer, GamesBeat (moderator)
Source link