We have brand new Cisco Cat 9500 and 9300 MGig switches.
With Netskope installed on our computer have run a speed test and getting 100Mbps connected to the Internet with Netskope enabled. With Netskope disabled we are getting speeds of up to 700Mbps connected to the internet.
Is there a reason why this is so slow, has it got to do with deep packet inspection?
Can anything be done to speed this up?
I am new to Netskope and our security team are in charge of this.
Page 1 / 1
Hello @dazmacmain,
What tool are you using to test the speed with? There are some tools that will show greater degradation based on how they handle HTTP threading and other details. Further, are you using the Netskope steering client, IPSEC/GRE tunnels or another method to steer the traffic?
the idea is the same, speed tests are not overly relevant when gauging performance. Lots of factors come into play here, some include SSL Decryption. We are fighting this same question internally since historically gauging performance has always been “just run a Speedtest”, “oh, you are not getting line speed there must be a problem”. Most users try to diagnose a problem that doesn’t exist. For example, we don’t steer X application through Netskope. When a user has an issue with X application they run a speed test and assume that’s the problem. You’ll get different results on Speedtest.net, fast.com and speed.cloudflare.com. You’ll further get different results if decrypting Speedtest traffic.
What we look at instead is user experience. In just about all cases where someone reports an issue to us, performance isn’t the issue. It might be “zoom is crashing and hey look, my speed tests are junk. However everything else runs good”. We are pivoting to DEM based tools to help tell the story. We’ve been telling this story through a Kadiska deployment, where every application being accessed is showing to be faster and less problematic going through Netskope than without (likely due to their strong peering).
the idea is the same, speed tests are not overly relevant when gauging performance. Lots of factors come into play here, some include SSL Decryption. We are fighting this same question internally since historically gauging performance has always been “just run a Speedtest”, “oh, you are not getting line speed there must be a problem”. Most users try to diagnose a problem that doesn’t exist. For example, we don’t steer X application through Netskope. When a user has an issue with X application they run a speed test and assume that’s the problem. You’ll get different results on Speedtest.net, fast.com and speed.cloudflare.com. You’ll further get different results if decrypting Speedtest traffic.
What we look at instead is user experience. In just about all cases where someone reports an issue to us, performance isn’t the issue. It might be “zoom is crashing and hey look, my speed tests are junk. However everything else runs good”. We are pivoting to DEM based tools to help tell the story. We’ve been telling this story through a Kadiska deployment, where every application being accessed is showing to be faster and less problematic going through Netskope than without (likely due to their strong peering).
Please, dear Netskope Team, provide an official page discussing the topic of speedtests and provide your own speedtest-page which can be accessed by the users using their browser.
You could provide the information from https://notskope.com/ and also the functionalities you already have in the client:
If you haven't already registered, now is a good time to do so. After you register, you can post to the community, receive email notifications, and lots more. It's quick and it's free! Create an account.
