i3D.net logo in blue
i3D.net logo in blue

A 50% cost drop and a faster Anycast backbone: Inside Acronis’ North America network transformation

Acronis has spent more than two decades helping the world protect its data. Founded in 2003 and headquartered in Switzerland, the company has grown from a backup software provider into one of the industry’s leading cyber protection platforms. Acronis delivers a simple promise: to keep data safe, accessible, authentic, and secure, no matter where it lives.

Delivering on that promise, however, requires a global technical foundation capable of moving vast amounts of data quickly and predictably. And in North America, that foundation was becoming increasingly complex to operate.

The challenge: An Anycast backbone with heavy operational weight

For years, Acronis ran its own North American Anycast backbone. This included a dozen Points of Presence (PoPs) interconnected by redundant routers, multiple ISP connections, internet exchanges, direct peerings, and a custom‑engineered, jumbo‑frame‑enabled backbone built from third‑party wavelengths. It was a high‑performance Anycast design, and for a long time it worked exactly as intended.

But complexity has a gravity of its own.

For Ilya Sivkov, Lead Network Engineer in Acronis’s Data Center Operations Network Team, maintaining that backbone meant additional operational coordination  A typical week might include handling fiber cuts on the West Coast, tracking ISP contract expirations, coordinating with colocation smart‑hands staff to swap transceivers that had broken down, or juggling IX port issues and related traffic shifts.

“Running that backbone meant nonstop capacity planning, hardware swaps, contract tracking, and dealing with outages. This could sometimes be an optic cut in Seattle, other times a transceiver failure or an IX port going down. It required significant operational coordination to maintain the level of performance we needed.”

Ilya Sivkov, Lead Network Engineer, Acronis Data Center Operations Network Team

The solution for that operational strain was a cluster of Acronis edge proxy servers — specialized nodes that terminate customer-facing TCP sessions as close as possible to the traffic source. These proxies are essential for high-throughput backup performance. When a business in the Midwest backed up critical workloads to a storage destination in Phoenix, , the edge proxy split the TCP session into two legs. A short hop from the end customer to the Acronis edge proxy allowed the TCP window to scale quickly, while a longer hop across Acronis’ jumbo‑frame capable backbone carried data efficiently the rest of the way.

It was smart. It was necessary. And it required Acronis to maintain both edge compute and backbone capacity in all the right places. As requirements evolved, the engineering team began evaluating a model that could preserve performance while reducing both direct network cost as well as easing operational complexity

That search ultimately led them to i3D.net’s Backup and Disaster Recovery Infrastructure.

When your partner becomes a co‑builder of your architecture

Acronis approached the evaluation like engineers: methodically, skeptically, and with clear criteria. They needed a partner with a global, highly optimized Anycast-capable backbone. They needed the ability to deploy bare metal servers close to Acronis’ end customers as a standard well-maintained product. And they needed flexibility; someone willing to adapt their footprint to Acronis’ architectural needs.

What stood out immediately about i3D.net was that it could offer both the backbone network and the edge compute. But what sealed the decision was something more unusual: when Acronis explained that Phoenix was one of its most important data destinations, i3D.net agreed to build a brand‑new Point of Presence there specifically to support the project.

“You didn’t have a PoP in Phoenix before,” Sivkov explains. “You opened a whole new deployment just to accommodate our needs. That was a major deciding factor.”

Validating the approach through real-world testing

The partnership began with a proof of concept in Chicago, where Acronis tested how a customer backing up to Phoenix performed across three different paths. Over public internet, speeds reached around 100 Mbps. Using Acronis’ existing backbone with its edge proxy bumped that up to roughly 440 Mbps. But the test over i3D.net’s backbone network, with only a single 10‑gig transit link and a test server, achieved slightly better results at around 460 Mbps.

It was a small difference, but it mattered. It meant that Acronis could preserve or improve customer performance while dramatically simplifying its architecture.

With i3D.net’s Server BGP Session product, Acronis edge proxy servers participate directly in the routing by Anycasting their IP space into i3D.net’s backbone. This approach allows the end user to hit the Acronis edge proxy closest to them, while also giving Acronis self-service control over which nodes actively take part in handling customer traffic — a convenient tool to have at your disposal for maintenance activities.

“Your Anycast backbone was simply better optimized for this kind of workload. It showed that the customer experience could stay the same or get better, while reducing our operational footprint.”

Ilya Sivkov, Lead Network Engineer, Acronis Data Center Operations Network Team

With the POC outcome clear, the teams moved into migration.

Migration & onboarding: A partnership that shifted gears

As the migration plan began, Acronis initially expected a long transition where both backbones would run side by side for a period of time. Some edge PoPs managed by Acronis were scheduled to remain online deep into 2027 due to contracted colocation or wavelength termination dates. But as Acronis and i3D.net mapped out the costs, the picture shifted.

“Once we assessed the early termination fees, it became clear that accelerating the transition was the more efficient path. Maintaining a hybrid architecture for several more months would have added unnecessary operational complexity, so we decided to complete the migration earlier.”

The pivot dramatically accelerated the migration timeline.

What made this possible was the unusually tight collaboration between the teams. Instead of support queues and escalations, Acronis worked directly with senior engineers and architects at i3D.net.

“From day one, we were in tightly connected discussions with the key stakeholders. It was one of the smoothest onboarding experiences I’ve ever had.”

Ilya Sivkov, Lead Network Engineer, Acronis Data Center Operations Network Team

Some locations, particularly Phoenix, required intense coordination. Acronis had to secure connectivity from a data center without a dense carrier ecosystem. Dozens of back‑and‑forth exchanges with third‑party providers were needed. But what stood out was i3D.net’s speed.

 “We know how difficult it can be to get the right connectivity in the Phoenix area, and the team managed it in record time,” said Sivkov.

i3D.net navigated hardware shipping challenges, tight third-party carrier service delivery timelines, and North America–specific wavelength provisioning constraints, often by leveraging pre‑stocked hardware or creating temporary interconnections elsewhere, such as Dallas, to guarantee operational continuity regardless of any risks to Phoenix’s timeline.

By April 2026, only two private Acronis backbone links remained: geo‑diverse paths between Ashburn and Phoenix used exclusively for data‑center‑to‑data‑center backup replication.

An Anycast backbone that is leaner, faster, and ready for growth

Acronis went from managing twenty-four backbone routers to just six, from more than twenty leased long-haul backbone waves to only two, and from a more resource-intensive operating model to a stable, predictable, managed backbone.

Financially, the shift delivered more than a 50% reduction in annual operational costs.

“This solution gives our customers the same experience as before, but at a fraction of the operational burden and financial costs."

Ilya Sivkov, Lead Network Engineer, Acronis Data Center Operations Network Team

Turning a network overhaul into a strategic advantage

Today, Acronis’ North American traffic flows seamlessly across i3D.net’s backbone. Edge nodes sit closer to customers without Acronis needing to operate the underlying infrastructure.  Latency‑sensitive workloads perform as intended. And the company now has a clearer, more flexible path to future data center expansions.

What began as a performance challenge became an opportunity: a chance to simplify, improve, and reclaim the engineering hours once spent chasing fiber cuts and replacing failed optics.

In the end, Acronis didn’t just rebuild a backbone. They found a smarter, more sustainable way to run it. If your team is facing similar network complexity challenges and wants to explore what a different approach could look like, we’re always happy to talk.

Managed Game Hosting Services header visual

Contact our team to find out more about our Backup and Disaster Recovery Infrastructure!

Contact us
View all Customer Stories
Acronis logo
Company Description
Acronis is a global cyber protection company providing integrated backup, disaster recovery, cybersecurity, and data management solutions.
Industry
Enterprise
Products
Solutions
➜ Contact us