Why should you monitor your workflow
“In a perfect world, everything is orchestrated and works without a problem; Devices don’t break down, configuration is error free, network is private and all you need to do is to drink a glass of lemonade and enjoy life at the beach, so why are you still coming to work?”
A decade ago, it was simple; we had point to point cables, all the equipment based on dedicated hardware and resided in one location. Now we moved to COTS equipment based on Software and GPU, IP networks that need special attention, operating Cloud services, spread and remote locations.
Yes, it may be cheaper to design and easy to adapt, but now it is up to you to babysit this new operation.
Why companies do spend $$$ on monitoring:
- The software running on COTS may run in to trouble
- The network must be rearranged with any new service and application, network services may go down
- More and more bandwidth requirements, Network devices may be stressed to a limit
- Cloud is provided as a service, who do companies call in case of a trouble?
- Resources are unknown, you only know what was booked but not if you received the right configuration
- External network is provided as service, how do you report a problem? What do you report?
- How do you know if the network can meet your service?
- How do you pin point the root cause of a problem; is it service or network?
The answer is – you have to be on top of all of the above and spend money on resources; manpower, training, field testing, contractors and vendor SLA.
Just like the broadcast industry suggested to use tools like ETR-290 compliance monitoring and Multiviewers to assist organization to be on top of the media workflow. The IT industry has produced tools, training and techniques to monitor and network availability and equipment health in general.
No attention was given to the network ability to provide and sustain the delivery of your video service. Network Architects are great in drawing a straight line from your location to the cloud and then between cloud services. This gives you the fuzzy feeling that its simple and easy. You may be instructed to add some kind of Reliable delivery protocol ( FEC, SRT, RIST, SMPTE2022-7 ), but how do you know how to make it work for you? Which to select? Will it overcome the network challenges? What are the challenges and are they changing?
So, what is the problem?
Companies need to run a very sensitive media data on top of an IP network. The operation and IT teams need to be aware of what can cause this data to break down; jitter, Latency, packet loss, re order and more. This data is not alone and runs with other services on the same network. Just like your commute to work, you share the road with others and they may miss behave and cause problems. You must pay attention to your car and other road users in the same time.
Your job is to be on top; identify limitation before you start the service, select the best network provider, identify the problem as they happen and not from the company call center, promptly provide the information to the people that job is to deal with the problem.
Let’s think on how many resources does your company spend on a service issue?
- How much time does it take before spinning a new service?
- How much time and material a team spends to evaluate the service end to end?
- How do you detect a problem like connectivity, bandwidth, Latency, buffer configuration?
- How much money does the company pay for SLA?
- How much time and resources the company spends to travel to a remote site?
- Try to remember the last time you called your equipment vendor for assistance?
- How much time did it take to find the root cause and be back in service?
The bottom line of these question is that you probably spent a lot of time and resources that you never faced in the old days. Time and resource are operational cost that we try to reduce. Monitoring is not a new technology in the block, it has been around for many years; IT and network, equipment configuration and health, QOE and more. But the Delivery network was left aside as there was no true solution to evaluate and report it in real time. Companies used other techniques like Speedtest/Iperf, Traceroute, network status/availability/overall bandwidth but that does not give any insight on the packet by packet relationship and how it affected the traffic.
If we only could shorten the time to give you the right tools and insight of the problem – it will save you 95% of your time and effort allowing, making your management happy, your customer with continuous service and allow you to focus on the main company objectives.
Alvalinks CloudRider observability service comes to answer many on these challenges head on:
- Provide you with tools to test your delivery network before and during streaming, find those pesky limitations in a fraction of the time.
- Provide live and proactive network testing with statistics on Packet loss, Jitter, Latency, RTT and possible route changes
- Provide dynamic route/hop/RTT information to alert you of the path your media is using, allows you to prefer one over the other.
- Identify a slowdown and congestion of a network hop
- Buffer estimation to guide you on the best buffer setting for your de-jitter and reliable transport buffer setting.
- Be on top on services running in parallel to you service or those which can cause harm for avoidance
- Setup alarms to detect trends, events that may cause your service to break down.
- Collect information in IT language for easy data collaboration
How much does it cost? almost nothing. How much you win from it? everthing.
Want to know more? book a demo!