In Live Streaming, Redundancy Does Not Mean Resiliency

He goes on to explain that "redundancy can be your best friend if you test it," and describes the way he teaches "simple redundancy" to his staff. "You can have redundant servers, you can have redundant regions, but then you need an operator that understands, 'This trip just went off, where am I going next? What do I need to do next?' So many people think this is automatic and it's just not."

"Simple redundancy" is as simple, he argues, as making checklists and adhering to them. "It's really a series of checklists you create for yourself, and you try to go through as many of those as you can that your client's going to pay for so that when you hit that mark, it's all smooth. But in order to get there, we have to be very, very proactive."

Avoiding Redundancy Theater

Upadhyaya turns to LinkedIn's Packard and introduces the notion of "redundancy theater," or poor, untested failover design that masks the lack of a real plan and doesn't deliver the type of resiliency one might expect. "When you're reviewing a customer's workflow," he says, "they're coming to you and they're saying, 'Here's our plan.' What are some examples of what you see as redundancy that "looks like it's going to be in place, but doesn't give you the operational protection that you think it does?"

"It feels like I've been doing this for long enough to recognize when I'm working with somebody who is unfamiliar [with live streaming], or comes from more of a video production background as opposed to a live background," Packard says. "If we get to a point where we feel that the show is at risk, we try to get out ahead of it as soon as we can so that the client knows that their vendor is not where things need to be. Ultimately, we want the show and the stream to go well. That's what we're about."

Packard says what's often needed here is to have the "people skills" required to know "where you can apply that pressure. Sometimes, going up escalation paths into the sales pods can be your best friend in that process."

Upadhyaya agrees that streaming ops teams often walk a difficult tightrope in these scenarios where they're managing multiple roles, competing budgetary line items, and stakeholders, and "balancing technical redundancy with the business reputational risks for your customers. You're making sure that they recognize what is the reputational or financial risk of this event going wrong versus the amount of money they want to spend on redundancy, resiliency, and testing ahead of time as well."

Upadhyaya agrees that streaming ops teams often walk a difficult tightrope in these scenarios where they're managing multiple roles, competing budgetary line items, and stakeholders, and "balancing technical redundancy with the business reputational risks for your customers. You're making sure that they recognize what is the reputational or financial risk of this event going wrong versus the amount of money they want to spend on redundancy, resiliency, and testing ahead of time as well."

Join us November 9-11, 2026 for more thought leadership, actionable insights, and lively debate at Streaming Media Connect 2026! Registration is open!