Multicast ABR & Scalability
Multicast ABR (mABR) is the most efficient and scalable way to handle peak traffic in live streaming. At the “Future TV Network” panel during Connected TV Summit 2025, operators like TIM and BT reported cutting backbone traffic by 80–90% and reducing rebuffering by up to 10×. This article explains why mABR is essential for sustainable streaming, especially as IP delivery replaces broadcast and network costs are predicted to soar.
1. The Peak Is the Problem
Streaming traffic isn’t flattening. It’s surging – especially during live events. FOX reported a peak of 135+ Tbps during the last Super Bowl, driven by 15.5 million simultaneous viewers on Tubi alone. That’s not an outlier. That’s the new normal.
Your network isn’t failing during these peaks because of poor engineering. It’s failing because it was never built to handle 15 million people hitting play at the same time on the same stream.
Live concurrency is growing faster than capacity. Major sports, news, and entertainment events create sudden, massive demand spikes. These spikes crush unicast delivery, and can trigger:
- CDN costs surge
- Rebuffering increase
- Drop in bitrate
- Decrease in viewer satisfaction
These peak loads used to be edge cases. Now they happen more often. That means your streaming delivery strategy needs to change.
2. mABR Is the Proven Answer
Multicast ABR (mABR) offloads traffic from the backbone during peak hours, without hurting quality or flexibility. It uses multicast to deliver a single stream across the network, then turns it back into adaptive unicast at the device or gateway. The benefits aren’t theoretical, they’re operational.
Here’s what operators are seeing:
- TIM (Italy):
- 84% reduction in network traffic during live events
- 10× decrease in buffering
- Viewers get the highest available bitrate more often
“We’ve measured around 84% traffic reduction compared to unicast – users enjoy ten times less buffering. It’s consistent, stable, and it lets us deliver top-quality streams even during peaks.” Diego Gibellino, TIM (During the panel: “The Future TV Network” at Connected TV World Summit, 11 March 2025 in London).
- BT Group (UK):
- mABR preserves bandwidth during spikes
- No compromise on viewer quality
- Transparent to the device
“With multicast ABR, we can manage bandwidth and favor higher quality renditions during peak loads. It’s transparent to the end device, and it gives us a better customer experience .” Brendan Hole, BT (During the panel: “The Future TV Network” at Connected TV World Summit, 11 March 2025 in London).
mABR doesn’t replace entirely unicast. It works alongside it, handling the load during the hardest part of streaming: concurrency. If your streaming strategy doesn’t include mABR, you’re paying for network capacity to accomodate spikes you could avoid.
3. The Ad Insertion Myth
A common objection to multicast ABR is: “You can’t do targeted ads with it.” That’s outdated. It’s been technically solved (Read more HERE) – and already deployed.
Dynamic ad insertion works with mABR. Not as a workaround. As a design feature. Operators are using it right now for both mid-roll and pre-roll ads on live and time-shifted content. It works across devices, including the main screen, and keeps quality intact.
3.1 Unicast Switching Today
Today’s mABR deployments use smart unicast switching to handle ad breaks. It’s simple and efficient:
- During live playback, the main content is delivered via multicast.
- At the ad break, the system dynamically switches to unicast to deliver a personalised ad.
- Once the ad finishes, it switches back to the multicast stream.
This already works in the field. BT and TIM deliver targeted ads on live TV using this approach, without visible glitches or playback disruption. On the viewer side, there’s no difference. The stream continues smoothly. On the operator side, it enables:
- Per-user targeting using ABR manifests
- SSAI (server-side ad insertion) logic fully compatible with industry standards
- Compatibility with all ad decision servers (e.g., VAST, SCTE-35)
There’s no “mABR or ads” trade-off, even when switching to unicast for ad delivery. Yes, targeted ads do require unicast streams. But in live deployments, that switch doesn’t involve every viewer. While ad breaks do trigger a unicast spike, it’s much smaller than it would be without multicast. You still get the capacity savings. In fact, systems with mABR see at least 3× less unicast traffic during ad breaks than those relying solely on unicast.
3.2 Pre-Caching Tomorrow
The next step isn’t about fixing glitches, it’s about scaling smarter.
Pre-loading ad segments in advance reduces backend load and improves scalability during peak viewing.
Here’s how it works:
- Broadpeak’s nanoCDN® software agent in the home gateway pre-fetches upcoming ads before the break.
- It stores them in RAM, ready to be played instantly.
- The manifest still controls what plays and when, based on the ad decision server.
The switch is already seamless for viewers. Pre-caching just makes it more efficient behind the scenes, especially at large scale.
Why this makes sense:
- Ad segments are lighter than live video streams.
- Preloading avoids last-second unicast requests.
- It guarantees smooth transitions and reduces errors during playback.
Pre-caching also helps avoid congestion during ad breaks, when thousands of viewers could otherwise hit the CDN at once. Instead, each device already has the ad ready to go.
4. It’s Not All or Nothing – But It’s Still Essential
One common argument against multicast ABR is: “You’ll need unicast anyway.” That’s true, but it partially misses the point.
Multicast ABR isn’t supposed to replace unicast entirely. It’s supposed to carry the heavy load. It handles the biggest strain on your network: high-concurrency live streams. That’s where it brings the most value.
Here’s how to think about it:
- Use multicast for the big stuff:
National events, live sports, reality finales: anything millions watch at the same time. That’s where unicast breaks. That’s where multicast saves you. - Use unicast for the rest:
Long-tail content, niche channels, one-to-one interactions. It’s fine to keep those on unicast. They don’t overload the network.
Multicast ABR is a selective tool, not a total replacement. You only need to apply it where it matters most.
Partial Coverage Still Pays Off
You don’t need 100% multicast penetration to see benefits:
- Even 20–30% coverage in your network can cut traffic significantly.
- Popular devices like set-top boxes and smart TVs already support mABR.
- You can start with a subset of your customer base and expand gradually.
Multicast ABR works in parallel with unicast. They complement each other. You don’t have to choose one or the other. The real advantage comes from combining them properly.
Multicast ABR offloads what unicast can’t scale. Unicast fills the gaps multicast isn’t meant to cover. Together, they give you a delivery model that’s efficient, resilient and scalable.
That’s why mABR is essential, even if it’s not exclusive.
5. You Already Have the Infrastructure
You don’t need to rebuild your network to deploy multicast ABR. If you’ve already rolled out IPTV, you’ve done most of the work.
Multicast ABR uses the same core multicast capabilities already present in many operator networks. No forklift upgrade. No overhaul. It’s a software-layer upgrade that brings modern ABR streaming into your existing architecture.
It’s Already Working: At Scale
Operators across regions and architectures are using mABR in production:
- TIM (Italy): Using multicast ABR since 2021 for live sports (Tim and OTT), with over 84% traffic offload and major buffering reduction.
- BT Group (UK): Live content delivery at scale without quality drop, thanks to mABR through the MAUD initiative
- Orange: Operating mABR across live Orange and OTT services.
- Claro (Latin America): Using mABR to handle peak loads and scale live distribution.
This isn’t proof of concept. It’s commercial deployment.
You’ve Got the Building Blocks:
If your network already supports multicast (as most IPTV-ready networks do), then:
- You don’t need to rip and replace your infrastructure.
- You can integrate Broadpeak’s nanoCDN® components into your headend and CPE layer.
- You can maintain existing IP multicast distribution, just with modern ABR packaging.
Modern home gateways and STBs are increasingly compatible out of the box. Many already support nanoCDN, or can with a firmware update.
What This Means for You:
- Faster deployment, no hardware overhaul.
- Immediate network offload in high-traffic zones.
- Lower CDN spend with minimal operational risk.
If you’re delivering live TV today, you likely already have what you need to get started with multicast ABR. It’s just a matter of turning it on.
6. The Carbon and Cost Argument
Streaming more video to more people doesn’t just stress networks. It burns energy and budget.
When you deliver one unicast stream per viewer, you multiply the number of sessions and the infrastructure needed to support them. More servers. More interconnect bandwidth. More power.
Multicast ABR avoids that.
- Less backbone traffic means less energy consumption and fewer active servers.
- Offloading peak traffic reduces the need to overprovision for rare high-load events.
- CDN offload lowers costs and limits your exposure to third-party platforms.
Every bit that doesn’t traverse your core network is a bit you didn’t have to pay for, or power.
There’s also a strategic angle. By reducing CDN dependency, you avoid being boxed in by vendor pricing, limitations, or regional gaps in coverage. You keep more control over how your content is delivered.
If you’re looking at both opex and sustainability goals, multicast ABR does two jobs at once.
7. The Broadcast-to-IP Shift Makes mABR Inevitable
Broadcast is winding down. IP is taking over.
The BBC plans to shut down digital terrestrial (DTT) in few years. Other broadcasters are on similar paths. That means millions of households will rely on IP for live TV, and networks will need to scale without breaking budgets.
BT recently estimated it could save up to £1 billion over ten years by using multicast-assisted delivery for linear TV in its own infrastructure, rather than scaling unicast alone [1]. That kind of cost difference makes the case clear: if you’re planning for a full-IP future, you need a delivery model that’s built for scale.
Unicast ABR doesn’t scale economically for national reach. CDNs become bottlenecks. Public internet paths add risk. mABR removes that pressure by sending one stream to many, with modern ABR features built in.
National-scale IP delivery works today, but the cost structure is fragile:
- Unicast ABR gets expensive fast at scale.
- And it’s not just a cost issue, it’s a capacity risk. As major events like the FIFA World Cup shift from broadcast to streaming, they’ll generate unprecedented traffic peaks. Most IP infrastructures aren’t built for that kind of scale.
- CDN costos can also be unpredictable, especially for live. The more viewers you get, the more you pay and the harder it is to guarantee quality.
Multicast ABR gives operators a third path:
- Like broadcast, it sends one stream to many.
- Like ABR, it supports adaptive bitrate, device compatibility, and ad insertion.
- It avoids the scaling cost of unicast and the rigidity of legacy broadcast tech.
As live TV moves fully to IP, you need a delivery method that can scale without spiraling costs or sacrificing quality. That’s what mABR is built for.
Conclusion: mABR Is the Efficiency Layer Streaming Needs
You don’t need to replace your current systems. To become future-proof, you just need to upgrade what they carry.
Multicast ABR isn’t a leap of faith. It’s working now – for TIM, BT, Orange, and others. It cuts costs, lowers latency, and preserves quality during the very moments your viewers care most.
It’s not all-or-nothing. You can start small, apply it where it matters, and grow from there.
But ignoring it comes with real costs:
Higher traffic. Bigger bills. Worse viewer experience. Less control.
Want to assess your network for multicast ABR readiness?
[1] source: https://www.lightreading.com/video-streaming/bt-says-it-can-save-uk-up-to-16-5b-on-streaming-deluge
Written by Xavier Leclercq
As the Vice President of Business Development, Xavier aims to further the position of Broadpeak as the premier provider of streaming technologies.
In his career spanning nearly 20 years, Xavier gained extensive international experience in the industry, specializing in tech consulting and sales to content owners and network operators worldwide. Before joining Broadpeak, he held leadership roles in IP Video at Nokia, Alcatel-Lucent, and Akamai.
Xavier graduated from Institut Mines Telecom Lille-Douai (formerly Telecom Lille) engineering school and holds a Master’s degree in Business from the University of Lille.
