cPacket Service Assurance: Realtime Video Production
Real-time video environments demand precision and speed. Troubleshooting can’t wait for decoding or downstream analysis. In this session, cPacket explored how packet-level observability enables immediate detection of transport-layer issues like encoder faults, fiber/switch errors, and edge-to-cloud latency disruptions. They demonstrated how their observability solution, with real-time alerts, dynamic dashboards, and ServiceNow integration, empowers proactive monitoring and MTTR (Mean Time To Resolution) reduction across complex, long-path video delivery networks. Erik Rudin, Field CTO, introduced the scenario of live video streaming, emphasizing the critical importance of video quality for businesses. Ron Nevo, CTO, further detailed the intricate environment of live streaming, involving multiple cameras, production vans, cloud processing, transcoding, and distribution, all of which can introduce potential points of failure.
The core of cPacket’s approach is to deploy monitoring points throughout the video delivery path to quickly determine if an issue is network-related. For real-time video, the presence of even minimal packet loss is a clear indicator of a problem. cPacket’s solution continuously analyzes RTP (Real-time Transport Protocol) streams, triggering real-time alerts (e.g., via Slack) when packet loss increases. These alerts provide direct links to detailed analytics, allowing operators to pinpoint the exact location and nature of the fault, whether it’s a physical cable issue, a video machine problem, or a cloud link disruption. Furthermore, the system automatically creates tickets in existing IT service management tools like ServiceNow, ensuring that identified issues are integrated into the customer’s operational workflows for prompt resolution.
This use case exemplifies cPacket’s broader strategy for service assurance, focusing on delivering actionable insights rather than just raw data. By acquiring and contextualizing packet data at line rate, integrating it into existing ecosystems, and leveraging AI for anomaly detection, cPacket aims to proactively identify and prevent service degradations. The emphasis is on improving the triage process and providing measurable outcomes, such as reduced MTTR and improved customer experience. The session underscored that AI serves as an augmentation to existing analytics, enhancing the ability to predict and prevent outages by identifying subtle patterns like under/overutilized links and their correlation to service degradation or security concerns.
Presented by Ron Nevo, CTO, and Erik Rudin, Field CTO. Recorded live at Networking Field Day 38 in Silicon Valley on July 10, 2025. Watch the entire presentation at https://techfieldday.com/appearance/cpacket-presents-at-networking-field-day-38/ or visit https://techfieldday.com/event/nfd38/ or https://cPacket.com for more information.
Transcript
Good afternoon. I'm Mayor Kine. I'm Field CTO, uh, for c Packett.
And today we're gonna talk about our first use case, which is around service assurance, around an interesting use case, real time video production. So, uh, here we are streaming our data alive on the internet to you receiving this information at home. So I think this is a great example.
I've seen network cables all over the place. We're pushing packets through the internet and ult ultimately, you can hear what I'm saying, and you can, uh, you can get the context of, of the shot. So I think, you know, we've, we've put together a, a, a short video to kind of show what this looks like.
And, you know, I've, if you've been watching streaming services, uh, online, have you ever seen this before? This is when something happens. And then, ugh, you missed the shot, right?
And so there has been a couple events in the news where data, the quality of the video service is really, uh, a big issue for these businesses. Um, so we're gonna walk you through a scenario today that shows how video production works and the different components along the path, but more importantly, show that there's a complexity and there's a real time element that needs to be troubleshot, and we need to have that data in and provided to the right operators at the right time. And so Ron is gonna walk through, um, what that looks like.
Thank you, Eric. Yeah, I think, uh, we didn't, we didn't, uh, we didn't, uh, create the incident that we had, but, uh, and I was actually thinking maybe we don't need this, uh, use case example, because actually what we want to do with, uh, this use case is kind of demonstrate the, I would say, pre AI and how things work when things don't work, right? So the idea is when you have a live streaming event like this one, uh, just think about, you know, a hundred millions and millions of, uh, uh, people watching and, uh, you know, many, many cameras around the stadium and you want to get it done and you know, you want to do it in real time.
Uh, it's a pretty complex environment, right? You have the video cameras going to the, uh, van, that is the, where the production sits. It goes up many times to the cloud.
In the cloud, it can be, uh, during the processing, then you have to do transcoding. Sometimes it's coding is done in the cloud, sometime it's done in their main studio, and then it goes to distribution. And if something doesn't work, then who's to blame, right?
Is it the, uh, video machine? Is it the, uh, transcoder? Is it some, the link to the, to the cloud?
Is it just somebody, uh, kick the, uh, cable of the camera, right? So in order to address that, what you're looking for is really being able to, uh, put monitoring points everywhere and at least get to a very quickly to the MTTI, right? Was it the network issue or was it something else?
Because again, there are a lot of systems that are operating there. And the idea here is really be able to deploy that, get the information, uh, to a dashboard and set out of the dashboard an API that allows you to create an alert in real time and from a lot of, uh, uh, consideration. That's a relatively simple thing, right?
Because really when you're doing video, you pretty much don't want packet loss, right? So you don't need to be very smart about how you set a a baseline to know that you have a problem. So we have the ability to analyze RTP or video, uh, video streaming all the time.
And, and what you're seeing here is essentially we set up an alert that says, okay, if you are seeing any, uh, increase in video packet loss, send me an alert. I'm getting it in a real time alert into Slack, and I'm clicking the link and 'cause the internet is working, I can see exactly what it happened, which stream is impacting, this is a replay that we have in the lab, and I'm able to, uh, go and resolve the location of that, right? Was it someone hitting the, uh, fiber on the camera?
Was it the video machine? Or if it's not a video machine, was it the AWS uh, link? Was it anything else?
Right? So the idea here, again, because it's relatively simple to know that you have a problem, you can respond in relatively real time and you can get a Slack message, uh, you can, you know, I think at the same time we, okay, plug me out. We created also a ticket in ServiceNow or Datadog or any of these other tools that you have, right?
So it's really about, if I know that something, if it's very easy to identify based on the metrics that we have that something is wrong, I can connect it to your workflow to get the alert in real time. Go figure out where the problem is and either reroute or if it's a video problem called a problem, You'll have to believe me that it's there because I'm not a ServiceNow. Oh, here it is, right?
So we basically, we got the ticket in ServiceNow, right? So the idea is really to show that it's not just about, um, you know, again, hoarding the data, getting the data, it's really about identifying things in real time, sending the event to whatever tool that you're using, and then you are able to go and figure things out from there. So if all works well, you are actually gonna see the shot.
And maybe, and this, this happened here, we, we came back and I was, uh, I think it's, uh, this section I was arguing if we should take it out. Uh, but I guess it has, uh, it was, uh, in, in, in the context of today. It worked well.
Okay. I'll just let Eric just make sure you close it. Okay.
Just to go back to our maturity model, or sorry, that we talked about with the customer journey, and this is just to help apply it and get everyone thinking about this from a, from an outcome perspective, and how can they consume that? So what Ron talked about with speed is, is essential, right? Real time video is real time.
Um, time. It take, there's complexity with the transcoding and the processing, um, where in the path it could happen. And so if you're deploying CPAC along those monitoring points, um, we're getting the data into the system, then we're integrating it into the ecosystem.
We're having the dy uh, dynamic dashboards, but also we're improving the triage process, right? And that's really essential. If you have the right data at the right time, um, that's gonna help you, uh, get that insight.
And then, you know what, we know how to validate this. And so this is, we were talking to the, the business executives that are concerned about the services. We wanna maintain, you know, great customer, uh, experiences, you know, you want to be able to get the shot, you want to be able to, uh, prove that we are able to do this better.
We can do that by looking at meantime to resolve, um, also meantime to innocence. And then we also can use the a AI on top of this to help accelerate our existing processes.
