Hybrid connectivity and SD WAN are subject to performance degradations from a myriad of sources, not all of which are within IT and network operations’ direct control. However, with actionable evidence they can work with their vendors, cloud providers and internal teams to proactively optimize performance.
Transcript:
DNS is a particularly common source of latency! You are trying to resolve multiple host names to reach a SaaS, a cloud, a CASB that applies to everything. And on that host name, you get an IP address and that IP address is going to change through time that I address is going to change depending on your ISP.
It’s going to change depending on your location. And that’s going to be dynamic. So the first thing you need to understand is whether that redirection texts you to a place which is narrow enough to provide the good latency or far from where you are. So if I take examples, I remember, people having trouble with their CASB depending on the ISP redirected, not to the local node of the CASB, but being redirected to west coast and then trying to make their way to Europe.
There’s no network problem in that. Actually, the network works as expected. It’s just that the past is very long. Yeah. And the resulting experience is dreadful. So that’s number one. The second one is the ISP performance and the BGP routing. So it’s all about, what is the quality of the connectivity that MYP is going to provide you to that destination?
What is the what is the route that I’m going to be teching? And that’s driven by BGP, the protocol that drives, how internet flows are routed through public networks. And here you can find plenty of different consequences. Either the past can be I ly wrong or non-optimal you might find that some devices on the way are not familiar as expected, and that can result in.
A bad pass some additional, latency or had it packet loss on the way which is going to capitalize a, the application going through that. Okay. The third one is CASB and cloud proxies. You’ve got to think of CASBES as an additional node on the past, but CASBS have all the the bad characteristics of any cloud or sales platform.
They regarded to a place that place might be the right one, might be the wrong one and they have a processing time. So they might have delay on the way. And it’s important when you choose a CASB solution to understand whether. Their capability. So how distributed there are offers the right service for all of your users, depending on where they are located, which ISP use and so on.
And I guess it’s very important to measure that. Because you might find that some SASEs do not have a good coverage of certain region actually have a longer latency from your users to downloads or simply have a much slower processing. And all three reasons were had very. Very big latency additional impact , or had it latency.
The fourth one is SD WAN delay pass changes. When you use SD WAN, okay. You can switch from one later to another, that has an impact, but each delay’s got his own behavior and can change pass on the work and that can drive two packet loss that can drive. That does.
And the same thing applies to CDN sales platforms that you use are going to use CDM networks to in principle, to match, to make the content close to your users. But sometimes it just doesn’t work as expected and you might end up, having a long pass to CDN and that will have an impact.
The sixth one, We’ve got to remember that. Although we go for the internet, local networks also have a huge impact and have their own issues. And wifi is one of them. And we find that sometimes the wifi coverage is just not right. Sometimes it just doesn’t work as expected. And it’s important to identify that this might be the cause of bar experience for your users.
And the final one is cloud network past length. So again, a cloud is just not one hope I’m in the cloud. Next hope I reach my application. You might find that, sometimes you’ve got 10 nodes on the way, and sometimes you may just go across the pond to through the cloud network.
