Back
Back

How to deliver live low-latency video streaming at scale

How to deliver live low-latency video streaming at scale

Latency is the critical metric for live streaming platforms. Today’s viewers expect exceptional live streaming experiences with near-zero lag or downtime. Failing to meet that expectation has real consequences, damaging platform engagement and monetization.

For live streaming platforms, it means that architecting foundational infrastructures built to minimize delay between video capture and display is critical. With this in mind, we sat down with Oliver Lietz, Founder and CEO of nanocosmos to discuss the key ingredients behind nanocosmos’ success in delivering ultra-low latency live video streaming at scale.

What is latency in live video streaming?

Video latency is a measure of the delay between video capture and playback. It can be categorized in three groups (standard, low, and ultra-low latency). Achieving ultra-low latency is crucial for live streaming services to ensure end-users enjoy live playback in real-time, without delay.

Ultra-low latency video requires robust infrastructure

For live video streaming platforms, maintaining ultra-low latency is central to audience engagement. When video content is transmitted in close to real-time with virtually no delay between video capture and playback, it provides a more interactive streaming experience akin to a real-world interaction.

nanocosmos helps their clients achieve this ultra-low latency through their real-time video solutions like nanoStream cloud, which offer features including adaptive bitrate, stream protection, and advanced analytics. But it’s their underpinning infrastructure that forms the critical foundation.

“We created the whole infrastructure ourselves based on our own technology, content delivery networks, and edge points all around the globe. Then we created the player technology and analytics on top of that so customers can directly implement live video streaming into their web applications. It’s very stable, 100% robust and upheld by our infrastructure partners,” explains Oliver.

Image on the right
“Partners like servers.com enable us to connect in the closest possible way and the most efficient way, with as few problems as possible.”

Secure your CDN foundation

The key foundation for achieving low-latency infrastructure for live streaming is a robust content delivery network (CDN). Building a reliable CDN for live streaming requires thoughtful design across architecture, geographic distribution and redundancy.

  • Architecture: you need a clear architecture that includes origin servers for ingesting encoded streams, globally distributed edge servers for fast content delivery, a control plane for orchestration and monitoring, and smart routing to the best edge locations.
  • Strategic points of presence (PoPs): you should be placing origin servers near your production workflows to minimize ingest latency and edge servers close to audience hubs and growth regions to reduce latency to end users.
  • Multi-vendor approach: leveraging multiple infrastructure providers not only expands your geographic coverage but improves redundancy – essential to maintaining uptime and streaming quality in the event of an unexpected fault or outage.

This multi-vendor approach is key to nanocosmos’ low latency strategy. Oliver and his team partner with multiple providers, including bare metal hosting partners like servers.com, to ensure PoPs are maintained as close to clients’ production sites as possible and that they always have access to reliable, high-performance streaming servers, sufficient infrastructure scalability to meet demand, and support from teams of streaming infrastructure specialists.

“We need the right infrastructure partners so we can provide connections as close as possible to our viewers. It’s essential that we have partners who can help us solve network issues. That way we can focus on the video technology and our customers can focus on their business. Ultimately, we need to have the right back-office partners to work around possible issues – because there are always issues in global network delivery,” said Oliver.  “Partners like servers.com enable us to connect in the closest possible way and the most efficient way, with as few problems as possible.”

Image on the right
“It’s important to have the right points of operation so clients can send us a live stream, have it at the right location as close as possible to production and then deliver it efficiently around the globe.”

Navigating additional infrastructure challenges

Even with such a robust and specialized technology, consistently delivering top quality end viewer experiences doesn’t come without additional challenges. Challenges like compliance, security, keeping up with innovation, and finding reliable infrastructure partnerships, must all be balanced alongside low-latency requirements.   

Compliance

Your streaming platform may be subject to compliance requirements that dictate how and where you deploy your infrastructure. For example, in some regions, data residency mandates require data to be stored and processed within specific regions, driving the need for localized server deployments. You may also need to comply with security standards that influence your system architecture and vendor selection.

For example, as a European based company based out of Berlin, GDPR is a key consideration for nanocosmos.

“We need to comply with GDPR. It’s very important for our corporate customers to have compliance in this area,” shared Oliver.

Ultimately, responsibility for compliance falls to you, but there is support available. As well as seeking guidance from your relevant regional authority, look to partner with infrastructure vendors that have specialist expertise supporting streaming platforms in your region and will work with you as part of your team to design a bespoke, compliant architecture.

Security

Security requirements will also shape how you design and host your live streaming infrastructure. User data protection requires encryption, strict access controls, and monitoring systems. To maintain availability, infrastructure must also defend against DDoS attacks (distributed CDNs, traffic filtering and firewalls all help with this) and continuous monitoring and incident response should always be prioritized.

For Oliver and his team at nanocosmos, it’s no different. “On a global scale we also deal with security threats. Everything from DDoS attacks to attempts to hijack content and re-stream it elsewhere. These are things we must identify together with our clients to protect them against misuse.”

Keeping up with innovation

In an industry like streaming, which is innovating at rapid pace, keeping up with the pace takes significant time, effort, and resource. It’s a challenge that Oliver and his team know intimately:

“We have always been driving change and building new innovations into our cloud platform. It’s a balance between creating new features based on what’s available on the market and effectively implementing this into our production platform to build an additional service.”

Today, innovation is centered around AI adoption and for streaming platforms that is presenting both massive opportunity and significant operational difficulty – especially when deploying out at scale.

“AI is big. It’s everywhere of course. It comes with a lot of opportunities, but when you want to implement that into a live streaming workflow it takes a lot of resources.”

For Oliver, trusted infrastructure partnerships have been key to successful ongoing innovation: “It’s important to have the right server infrastructure and the right partners in the AI space.” It means that despite the obstacles that come with integrating AI into live-streaming workflows, nanocosmos is striving ahead. Upheld by a reliable foundational infrastructure, the team have implemented AI services for live translation and transcription and are implementing AI technologies at the video level to improve picture quality and rendering.

Choosing the right partnerships

Knowing you need reliable, high-availability infrastructure is one thing. Finding the right infrastructure partners to deliver it is another. Finding a vendor or vendors that offer the ideal combination of performance, scalability, support, and geographic coverage isn’t always easy.

“There are a lot of challenges with making the video technology itself reliable. Then to bring that technology into a new production environment on a global scale requires great infrastructure,” said Oliver.  “It’s important to have the right points of operation so clients can send us a live stream, have it at the right location as close as possible to production and then deliver it efficiently around the globe.”

Every live streaming platform will have a unique combination of these needs so there’s no one-size-fits-all solution. But knowing what to look for from prospective vendors is the best place to start. Key criteria include opportunities for customization and control, modern and redundant hardware as standard, efficient provisioning and reconfiguration speeds, a high-availability, redundant network architecture, automation and monitoring capabilities, and 24/7/365 human support.

Image on the right
“AI comes with a lot of opportunities, but when you want to implement that into a live streaming workflow it takes a lot of resources. It’s important to have the right server infrastructure and the right partners in the AI space.”

The future of real-time streaming

Testament to the impact of interactive media on engagement, the global interactive streaming market size is estimated to grow by 23.28% between 2025 and 2032. Use cases for live and interactive video are broadening with industries like auctioning, betting, gaming, enterprise, and broadcast (all of which nanocosmos serve) investing more heavily in live streaming technologies in a bid to engage audiences through interactivity and gamification.

“There are different industries that want to increase audience engagement,” said Oliver. They want to not only increase viewership but to build a community that stays on the channel and stays connected to the producers of the channel to keep engaged in the long run.”

As more platforms embrace real-time video, they will also face the challenge to reduce latency and viewer churn head on. The solution lies in robust design - continuous design and innovation applied not only to the software layer but to the foundational infrastructure that underpins all streaming workflows.   

Thank you to, nanocosmos, for their valued partnership. You can learn more about their work via our extended interview with Oliver.

Author: Frances Buttigieg

Frances Buttigieg, Senior Content Writer

Frances is proficient in taking complex information and turning it into engaging, digestible content that readers can enjoy. Whether it's a detailed report or a point-of-view piece, she loves using language to inform, entertain and provide value to readers.

Related articles