AI-driven SD-WAN Demo: WAN Assurance

00:04 welcome to a demonstration of juniper

00:06 mist wan assurance with session smart

00:08 routing

00:10 wan assurance is just the latest in a

00:12 suite of assurance offerings in mist

00:14 built to understand and assure user

00:17 experience completely

00:19 in the journey of a packet from client

00:21 to cloud as it hops onto your

00:23 industry-leading missed wi-fi

00:25 crosses your best-in-class enterprise

00:27 switch fabric

00:29 and traverses your innovative tunnel

00:31 free secure sd-wan the wireless wired

00:35 and wan assurances work together in a

00:37 complete full stack branch solution

00:41 let's hop into our missed user interface

00:43 for a look around

00:46 starting at the highest level view

00:48 designed for msps and the largest

00:50 enterprise network operators you have a

00:52 dashboard showing you the estate at an

00:55 organizational level

00:57 you see your device inventory and get a

00:59 top level indicator of organization

01:01 health drawing your attention to areas

01:04 that may require it

01:06 moving into a particular organization we

01:08 get our first glimpse at a site level

01:10 view

01:11 we see devices and user activity

01:13 happening at the site

01:15 a history of user and device events are

01:18 shown with client events categorized

01:20 into good bad and neutral events

01:23 applications being consumed by users at

01:25 the site are also shown giving a

01:28 comprehensive overview of user activity

01:31 further down key wan edge device

01:34 performance data is plotted in a set of

01:36 charts we see resource utilization

01:39 including cpu forwarding core and memory

01:42 near the bottom of the page we see when

01:44 performance for the network paths

01:46 connecting this wan edge to its peers

01:49 this gives a historical look at path

01:51 behavior including loss latency

01:54 jitter and moss score

01:58 moving back up to our clients list we

02:00 can even get a detailed view of

02:02 applications being used by a particular

02:04 user device at the site

02:07 now being able to see network data with

02:09 your own eyes is great but let's head

02:11 over to marvis and see what your ai

02:13 powered network assistant sees

02:16 these are marvis actions the proactive

02:19 side of marvis

02:20 actions are things that marvis has

02:22 already noticed found and understood

02:25 from taking a proactive look at the data

02:28 if there is an action that can be taken

02:30 to improve the user experience it will

02:32 be brought to the forefront here

02:35 from our wan we can see that marvis has

02:36 identified a persisting lte signal

02:39 quality issue

02:40 from here we can drill into the details

02:42 of the issue and get a better sense of

02:44 the impacts

02:46 looks like i should take some action and

02:47 have the antenna adjusted this is a

02:50 great example of marvis helpfully

02:52 suggesting actions we can take to make

02:54 the user experience better

02:57 marvis also features a conversational

02:59 interface where you can ask it questions

03:01 about the network

03:03 additionally marvis is conveniently

03:05 located at the bottom throughout the

03:07 interface so you can always ask it to

03:09 help you do things like troubleshoot a

03:11 device or access documentation

03:14 heading back to the wan insights page

03:16 let's take a closer look at some of the

03:17 cool stuff that the mist ai engine is

03:20 doing for us

03:21 at the heart of wan assurance are

03:23 something called sles this stands for

03:25 service level experience

03:28 sles break the network down into simple

03:31 units of measure

03:32 for the user experience called a user

03:34 minute

03:35 which is then used to score the various

03:38 things which may have an effect on user

03:40 experience

03:41 behind these seemingly simple

03:43 measurements are the complex and

03:44 powerful ai models of the miss cloud fed

03:47 by the rich telemetry from the session

03:49 smart network

03:51 this is where we can be assured that our

03:53 user experience was good and if it

03:55 wasn't quickly get to the root cause

03:58 here we see that our user experiences

04:00 have not been too affected by things

04:02 happening within the wanage device

04:04 itself or by issues on the wan links

04:08 but even still mist has noticed that

04:10 some application experiences are being

04:12 impacted

04:13 what could be going on

04:15 let's hop into our application health

04:17 sle to see

04:19 each sle contains a set of classifiers

04:21 that break down the root causes of poor

04:23 user experience in this case 98 percent

04:26 of the time that user experience has

04:29 been poor the cause has been actual

04:31 application server responsiveness issues

04:34 going even further we can see that

04:36 applications themselves have been slow

04:38 to respond

04:39 even while all the rest of the wan has

04:41 been performing perfectly

04:43 as we look at affected items by this

04:45 issue it shows us the users and

04:48 applications that have been impacted

04:51 so mist is telling me that my network is

04:53 fine

04:54 but that issues out on the internet are

04:56 still causing poor experiences

05:00 let's check the news here to see if

05:02 there's public reports of these issues

05:05 sure enough the same day of this

05:07 recording there was an amazon server

05:09 outage

05:10 applications reported as impacted are

05:12 the same ones that mist told us about

05:16 this is when assurance in action showing

05:18 us exactly what is happening in the

05:20 network

05:21 thanks for watching this quick

05:22 demonstration of missed wan assurance

05:24 with session smart sd-wan

05:27 hopefully you got a glimpse of how mist

05:29 assures user experiences on the lan and

05:32 helps you find and resolve impacting

05:34 issues faster than ever before