3 Questions Your IT Department Must Answer Before a Live Stream

One of the great parts of working for a company like Haivision is that if you have a question about enterprise live streaming, there’s always someone who has the answer.

We rely heavily on our IT team, as well as our many experts to help us create great video experiences both internally and externally, which allow us to help others to have the same great experiences. We use all of our products internally, which helps us to make sure that all of our customers are getting our best.

I had a chance to talk to JF Noel, our Director of Information Technology, to find out what he and his team prepare for when setting up internal and external live streams.

Let’s take a look at how the Haivision IT team handles live streams here, and what they do to make sure that every broadcast can be viewed in high-quality video by any employee who wants to watch, no matter where they are. Let’s start with the three questions they ask before every live stream.

1. Who is the talent?

Why is this question important? Hasn’t your Video Producer already answered that and planned this part of it out?

The reason this question is important is because the answer often involves more than one person. And should those people be in different locations, now you have some more questions to answer.

2. Where are the streams originating?

There are a couple of different ways in which this could happen, and even more ways to handle it. But let’s just look at two instances that our IT team has dealt with.

For the first example, let’s look at our monthly all hands meetings in which our CEO, Mirko Wicha, interviews various people from around the company. In these meetings, we get to find out more about what people in other departments and in other offices are working on, and they’re a really valuable part of our culture here.

For one of these live events, Mirko broadcasted from our Austin, Texas office. For these remote events, our IT team needs to know where the stream was coming from, where the production will happen, and where the live stream that goes out to viewers will originate.

In this case, the live stream came from Austin, but the production was all done here at our corporate headquarters in Montreal. IT needs to be ready to handle a stream from another location, and ensure that output from the headquarters is reaching everyone. In this case, they also sent Makito X low latency encoders to the Austin office that were set up to be able to send the stream directly to our head office, where that stream was handled by the video producer, and then broadcast live to all of our employees.

In the second case, we had a public webinar that featured a live presenter, Dan Swiney of LinkedIn, who joined presenters at our head office for a live broadcast.

The IT team just had to send an encoder to LinkedIn with the settings already entered. On their end, their great IT team plugged it in, sent us tone bars and audio as a test, and we decoded it on our end, then sent it to the production console. All that needed to be done was to make sure that we understood what ports needed to be accessed, and check the right boxes in the pre-set for the encoders.

This allowed us to have an interactive, interview-style broadcast that went out in an HD signal.. .which was the next piece of the puzzle for IT.

3. How many people are we trying to reach?

JF doesn’t care if the streams are going to salespeople or QA staff, he just needs to know where they are, and how many of them there are. In the case of an internal live stream, he’ll need to know how much bandwidth needs to be available at each location that the stream is headed to.

In the case of either an external webinar or an internal broadcast, the IT team needs to know how much output they’ll need from the source.

If the stream is going to three offices, the team will need to assess all three and make sure that the video networking workflows  can handle the streams without choking all of the bandwidth. And then there are more questions.

Bonus questions to make sure live streams are working correctly

Before an all hands live stream, the IT team goes through a laundry list of items that help them to ensure that everything will run smoothly.

In our case, one of the team will open up the interface for the Haivision Media Platform (HMP) and look to see if all of the gateways (in this case, we’re using Haivision Media Gateways) and confirm that all of them show up as online and working.

If that’s a yes, then they go to the next step, which is to check and make sure that the stream is viewable everywhere. There is a setting in HMP that allows the administrator to push a stream, but not make it live to everyone. This way, the admin can validate that the stream is operating correctly before pushing it live to the audience.

Once the gateways are confirmed as operational, and the source video is confirmed as valid, then there is a final check to validate that each gateway is receiving the source. The way this is done is by using a remote desktop in each location, whether it be our Atlanta office, our Chicago office, or any of the others.

All that needs to be done is to open the stream from the location that needs to be confirmed as receiving the stream, and see that it’s working.

Once these steps are done, the IT team can be very confident that the setup is good. What JF would add here, is that this is not just done once. The first check happens three days in advance. That gives the team lots of time to work out any potential kinks.

They’ll also work with the video producer to make sure that they’re getting everything that they need from his production tools and that the source is originating as a single stream.

The next check happens the day before. They go through exactly the same process, checking each of the above points once again. And then the last check happens the day of, within about an hour of the time when the stream will go live to viewers.

At that point, any potential problems should be easy to troubleshoot, given that they’ve so extensively and exhaustively tested the system already twice prior to then. With an hour, if something is broken, there is time to fix it.

This is the process from a very high level. But when things to go wrong, the team will be looking into things like:

  • Confirming that the ports the stream is being sent from and received by are available
  • The bandwidth is optimized for the stream
  • Checking all connections for potential errors
  • Investigating encoders and checking that settings are correct

And during the all hands stream, they’ll be checking to make sure that we’re not peaking any bandwidth in specific areas (like if too many people are watching via one wireless access point), and they’ll also monitor any tickets that might come in through our support system from viewers who may advise them of issues that are happening in specific viewing areas.

And if everything runs smoothly, it’s time to kick back and enjoy the live stream!

Get a deeper understanding of the workflows required for all hand streaming

There are a lot of different ways to handle the situations that come up for IT when setting up an enterprise live stream, and a lot of different ways to handle them. In the end, it comes down to ensuring that there is enough bandwidth the whole way through to ensure that the stream can be viewed in high quality, and that no one is missing out on the broadcast.

In our webinar, Broadcasting All Hands Live Streams from Anywhere, we explore a few different workflows for this process. And we’re also joined by Dan Swiney, who let us in on how they handle things like remote production for live streams at LinkedIn.

Click the link below to watch the recorded version of the webinar in high-quality streaming video!

Share this post