LiveAction: Gaining Deep Visibility Into The WAN Overlay
Operation teams need to detect and correlate application problems with the real-time events happening on, both, the overlay and underlay networks. They also need capability for retrospective visualization into such problems. SD-WAN introduces new requirements for such advanced monitoring capabilities.
LiveAction’s LiveNX™ is a robust visualization and analytics engine that integrates the network-generated metadata from Viptela SD-WAN technology to monitor application traffic on an end-to-end basis in real time. The retrospective visualization capability complements the intelligence provided by Viptela’s management solution and thus simplifies SD-WAN planning, monitoring and troubleshooting.
This webinar will cover the major integration capabilities between LiveAction’s LiveNX and Viptela’s SD-WAN and share existing best practices to effectively manage the entire lifecycle of an SD-WAN deployment. These include:
- Planning: Selecting pilot sites when planning SD-WAN deployments, identifying unsanctioned applications and end users, and establishing network utilization baselines which are used for right-sizing
- Deployment: Using LiveNX real-time visualization to verify Viptela SD-WAN policies and assure services are not impacted in the event of brownouts or abnormal events
- Management: Monitoring bandwidth and visualizing path control so network administrators can ensure application performance and mitigate problems before they affect users. Reporting on business applications to manage and optimize voice and video communications
Manan: Good afternoon, good morning, and good evening for folks who are joining from all across the globe. My name is Manan Shah. I’m director of product management here at Viptela. I’m joined with David, who is the principal engineer in the technical marketing group of LiveAction. Just a couple of notes: There are various attachments on this webinar, as well as a couple of poll questions we’re going to go through. As with other webinars, please feel free to ask us any questions that you might have. Let’s continue the conversation on Twitter.
Toddy’s topic is LiveAction and how LiveAction can complement Viptela’s to provide deep visibility into the Viptela WAN overlay. So in terms of the challenges today, if you look at the traditional WAN network, most of the traffic – almost all of the traffic – used to go from the branch networks to the data center, where the application is used to decide; and all of the traffic would get serviced through those applications. Very limited – five to 10 percent of the traffic would go outside of the data center to the internet.
But with applications moving outside of the data center to public clouds, like AWS and Microsoft [unintelligible 00:01:39] as well as applications being consumed as [unintelligible] service, most of the traffic these days is destined toward the data center, but is going on the internet and toward these applications residing in various locations. The traffic dynamics have changed; and as a result, the challenge has become bigger in terms of getting the end to end visibility into this network from a user perspective, from an applications perspective, and from a network traffic perspective. This is what customers tell us is one of the biggest challenges in deploying the [ST band] network today.
So with that, I’m going to start with our first poll question. The first poll question is, as applications move to the IaaS and SaaS clouds, how do you see the need of additional network visibility in your environment? Do you have the visibility that you need for the IaaS and SaaS clouds, or do you need visibility for IaaS cloud, SaaS cloud, or both clouds? We’re going to keep the poll open, so please respond to the poll.
Then, in order for those networks to be deployed in a seamless manner across this variety of different environments, I’m going to walk you through Viptela’s SD-WAN solution and how it facilitates this deployment, from visibility and the solution of WAN connectivity, across the different cloud environments.
First and foremost, Viptela SD-WAN solution is built on this transport agnostic fabric. What that means is, you can continue to use your MPLS circuits that you have. You can augment that with broadband circuits as well as LTE connectivity; and with all of that, we built that transport agnostic fabric; and on top of that transport agnostic fabric, there is a robust infrastructure capability, starting with security. The minute customers move from MPLS to a hybrid transport model, security becomes of utmost importance to them. So first and foremost, every packet that goes out of the Viptela SD-WAN solutions authenticated, authorized, and encrypted.
Viptela solution provides routing and segmentation capabilities across these multiple different types of circuits in the transport independent fabric. In addition to that, customers no longer need to get a different VRF from the MPLS providers, because in the whole network, the Viptela customers can create all of these VPN segments and segment the traffic across different types of traffic they might have from guest Wi-Fi to PCI traffic to other business applications.
We’re going to close the poll now. We’ve got a lot of responses, so thank you for responding. We have 50 percent of respondents who said that they need visibility on both IaaS and SaaS clouds. So that’s very good to hear, and that’s very relevant to the topic that we’re going to discuss today.
Moving forward with the SD-WAN solution: Viptela’s SD-WAN solution, in addition to the security, routing and segmentation capability, also provides a host of different infrastructure related capabilities like service chaining and service insertion, as well as multicast functionality. So once we have this transport independent fabric and robust infrastructure services, on top of that, we can layer down what we call the application policies. This allows customer like yourselves to decide which applications are important to you and how you want those applications to traverse through the network.
You no longer need to bind the application to an MPLS transport or a broadband transport. You can just set your applications and map them to certain SLAs. I can say that my voice traffic has to go over any path that’s meeting 75 milliseconds. I no longer have to say that the application has to go over MPLS or broadband. Those are some of the advantages. Then, of course, connectivity to the cloud – we talked about how the traffic patterns are changing. So it becomes extremely important to have efficient access to cloud services and the applications residing in the cloud.
One of the major hindrances of moving towards this SD-WAN is to get visibility and monitoring and analytics capabilities in the solution. So Viptela’s SD-WAN provides you a rich set of capabilities as well as management interfaces to manage the overlay SD-WAN network, and provides that from a single [unintelligible 00:07:17] called vManage. Let’s talk about how these different management interfaces are provided with the Viptela SD-WAN solution.
So in terms of the solution itself, what you see at the bottom of the slide are called vEdge routers. They come in software as well as the hardware factor. They form a secure connection with the vSmart controllers. vSmart controllers are the brains of the solution, which enforce all of the policies and allow different vEdges to talk to each other, depending on the policies set by the users.
Then on top of the vSmart controllers, is also vManage. This vManage provides the management interface, as well as different APIs that allow you to control the entire overlay network. In terms of management interfaces, there are various interfaces. All of the industry standard interfaces that you would expect from any next-gen network are available in the solution. All of the interfaces that traditional networking users expect are also available in this network, like the command line access, SNMP, syslog, and NETCONF. All of the flows are exported using IPFIX; which is, again, a standards based technology.
All of this is also available using the RESTful API, which is what customers want who want to automate the network. They can have used it to automate provisioning as well as monitoring on this network. With that I’m going to let David talk about how LiveAction is leveraging all of these management interfaces to provide the best visibility into our Viptela WAN solution. David.
David: Thank you for that nice introduction, Manan. It’s really a pleasure being able to present at the SD-WAN summit. A little bit about LiveAction: Really our goal is to simplify the network, and we do that means of our visual network analytics. So, a little bit more about the company: We’re based out of Palo Alto, California. That’s where we’re headquartered. We also have an engineering office in Honolulu, Hawaii – that’s actually where I’m from – as well as sales offices around the world. We’ve had our series A, as well as series B, financing. We’re continuing to grow our customer base as quickly as possible as well.
So it’s really appropriate that LiveAction is talking about the WAN. We have a strong heritage for the WAN, going back to 2006. LiveAction’s product, LiveNX, was actually conceptualized as Department of Defense research project. We’re working with the Office of Naval Research, and we’re tasked to provide a four times improvement in network operator efficiency on WAN edge routers. So what does that basically mean?
What that meant was, we were tasked to help very young, inexperienced marines – maybe 19, 20 years old – to be able to do very complex tasks on the WAN routers themselves. We provided NetFlow to be able to provide that end to end visibility across the topology. That’s what we’re really well known for.
So LiveNX, like I mentioned, is a visual analytics tool to really be able to help our end users improve the overall quality of experience for their users themselves. Now, the network – when you think about it – is built to serve up the applications. When you think about critical applications like voice as well as video, it could really expose the network weaknesses that are found within the system. So being able to use LiveNX to provide and make sure that the applications are performing well is critical to LiveNX itself.
Then the ability to troubleshoot quickly is very important also; being able to focus on finding, as well as fixing, the problem. LiveNX allows for that with a heterogeneous environment. We support many different vendors in the environment. Then, as Manan was mentioning, so many of our customers are moving toward cloud based applications. You think about Office 365, Salesforce.com, WebEx, GoToMeeting, various other types of applications. So many users are relying on that.
LiveNX provides that visibility to understand how those applications are being utilized within your network. When you also think about different lines of business, at times the network team, the IT organization, may not even know about all the different applications that are being deployed on the network by these various lines of business. Using LiveNX, you’ll get that visibility.
Of course we’re talking about SDN. We’re talking about SD-WAN. Many customers are embracing these brand new technologies and use LiveNX to be able to manage them as well.
So today, especially in our live demonstration, we’re going to be focusing on three areas. First off, being able to baseline and plan for your rollout of SD-WAN: When you think about that, we need to be able to understand which applications are being used within the environment. Then, being able to also identify and pick your pilot sites: Which sites are really good candidates to use in your proof of concept? Then, moving on to deployment phase: Being able to really understand how the Viptela system is doing its application routing; verifying that your applications that you’re putting into your policy – to take, maybe, the MPLS path or the internet path – ensuring that those applications are taking the path that you designed the network to do; and being able to use LiveNX with that end to end visibility to verify that, and understanding your site to site traffic as well.
Finally, moving into that next phase, is day 2 operations. That is really, after SD-WAN is rolled out, being able to understand how your applications are performing; performing your capacity planning; understanding how your cloud applications are being used as well.
So let’s talk a little bit about the value that LiveAction, LiveNX, brings to SD-WAN. On the left-hand side is really those values: Being able to make quicker decisions, being able to deploy SD-WAN quicker. Then on the right-hand side, we have a bunch of case studies in which we’ve been working with various customers on SD-WAN. A couple that I’d like to speak to: Regional Financial Bank. They initially deployed with 125 sites, and then they quickly expanded to 825 sites. With LiveNX in the mix, they were able to reduce the SD-WAN deployment project from nine months all the way down to three months. Then that Worldwide Logistics Company: About 2000 devices. The thought that they had from the network team is, LiveNX is an absolutely must-have tool for the SD-WAN.
So that’s what we’re hearing from our customers, and we look forward to working with all of you, too.
So let’s dive into a live demonstration at this time and kick the tire, so to speak; and see how LiveNX can help you with SD-WAN. This is LiveAction’s LiveNX. Let me help you get your bearings to understand what you’re seeing.
The topology is made up of multiple Viptela network elements. On the left-hand side are the Viptela management systems. So we see vBond, vSmart, vManage as well. Then on the right-hand side you’ll see a bunch of the various sites that we’re managing within LiveNX. So here in the upper left-hand corner, this is a vEdge router out in New York. We have a site out in Seattle in the upper right, as well as Los Angeles down in the lower section of the topology map. It’s all interconnected by multiple service providers.
So in this case we have the MPLS network that’s providing a lot of bandwidth for your critical applications. You have a lot of traffic going over that network as well as the public internet, being able to leverage that as well as a transport for your enterprise.
Now, as you can see, LiveNX is very dynamic in nature. You’re able to move the vEdge devices, place them wherever you’d like. Our goal is really to recreate the mental model of the network engineer. You can think of this topology map kind of like a very large dynamic whiteboard. Typically when you think about the network, you draw a bunch of circles; and then that represents the routers themselves. So this is that vEdge router out in New York. I’m just scrolling in and out of the topology using the scroll button on your mouse.
Now, that little circle inside: That’s representative of an interface. So in this case, this is Gigabit Ethernet 0/1. It’s connected off to the internet service provider. The top portion of that circle represents ingress bandwidth; the bottom portion, egress bandwidth. So very quickly and easily, you’re able to understand exactly how your network is laid out.
Now, what about all of those pretty colored lines that you see on the map? Well, that’s indicative of all of the flows, all the conversations, that are running throughout the network. You can click on these flows, and it will highlight in black. You’ll see how that flow is traversing the network. Here it’s originating from Los Angeles down below. It’s going through the MPLS network up to a site in New York, that vEdge router; and then going off to the internet at that particular point.
So, really being able to understand all of your network traffic is key to LiveNX. By double clicking on our router, that will automatically bring you into the real-time device view. It’s a real-time flow report of sorts. So here, you’re seeing all of those flows, all of those conversations, going in and out of this vEdge router in New York. You can interact with it by asking questions. You can type in the search bar for things like IP addresses or application names, to really understand what’s happening.
But we realize that you’re not always going to be at the console all the time, watching all these flows going in and out of the system. Click on “Playback,” and that will automatically bring up our historical playback capabilities. You can think of this kind of like a NetFlow DVT of sorts. You can select the date and time that you’d like to go back to, maybe a couple of days ago. You can show how often you want to update the flows, maybe every five minutes. Simply click “Play,” and you can sit back and watch all of those flows going in and out of the router, this vEdge device, as if you were sitting, watching all those flows live; being able to go back in time and find that issue that you’re troubleshooting, take advantage of these historical playbacks, to really understand what happened at that point in time.
So let’s talk a little bit further about those three steps that we talked through on the slide deck. First off, being able to help understand your baselining and your planning, before you set up a proof of concept with Viptela SD-WAN: The way that you’re going to do that is, we’re going to go to our dashboard, our WAN dashboard, here; and you can think of this WAN dashboard and read it kind of like a page in a book, from top down and from left to right. Now, when you think about the WAN, you typically think about three things. You think about sites. You think about applications, and you think about service providers.
That’s what we’re showing on this dashboard, intercorrelating all those three dimensions together. Now, when you’re picking your pilot sites, trying to understand and baseline your network, which sites are going to be a good candidate for a POC? You’re going to use this widget here in the middle top section: Site Utilization by App Group. Notice that New York is a good candidate. It’s using a good amount of bandwidth. The application mix is really critical applications, some [unintelligible 00:20:22] traffic as well. You can use, also, this middle widget for App Group Bandwidth by Site. Again, we see critical applications that are being used by New York, Los Angeles, and Seattle.
So take advantage of these views to be able to understand, baseline your network, pick your pilot sites. You can simply click on these widget headings, and they’ll automatically bring up a report; or you can double click on the bar, and it will automatically zoom in to that particular parameter – in this case, New York in that search term – and here is a time series chart of all the applications traffic that’s going outbound from that particular site.
So now we know which sites we’re going to use for our proof of concept: New York, Los Angeles, Seattle – those are good candidates we’re going to be able to use. So what about during the proof of concept phase? You’re going to be setting your policies, and you want to really understand how application routing is taking place. You can go back to the topology view. Let’s go ahead and do that.
So again, here’s that topology view that we were just describing. You can interact with this topology view in other ways as well. Perhaps during the proof of concept, you want to check out all of the traffic ingressing and egressing out of site New York. LiveNX knows about your sites and your service providers. Simply type in, “site = New York.” We looked at our backend database, and there is all the traffic going in and out of site New York, taking the MPLS as well as the internet path.
Maybe another example: Maybe you want to also troubleshoot and verify all the traffic that’s going over the MPLS network. Simply type in SP = MPLS. We’re going to look at that backend database, and we see all of that traffic going over the MPLS circuit. We see that there’s a good amount of your critical applications being marked AF21. That’s showing up as that dark green color that you see there.
Likewise, as you’re looking for specific applications, you can type in flow.DSCP = AF21. We’re going to look specifically at those critical applications. That’s the marking that we’re using by design. We see that AF21 traffic, those critical applications like Office 365, Salesforce.com, are being load balanced in this case. We’re verifying that. All that traffic, those green lines, is going over both the internet path as well as the MPLS circuit.
So during that proof of concept phase, take advantage of this topology view to really verify your application routing. You can even go back in time. So I you’re also working with your vManage, you can see when there might be certain alerts. Go back in time. Use LiveNX kind of like a network time machine. You can pick that date, pick that time, press “Okay,” and we’re going to be able to show you and paint the pic of exactly what your network looked like at that particular point in time.
So also during the proof of concept phase, what you can do is also take advantage of our brand new web UIs as well. Here’s an example of our site to site analysis story. Being able to understand all the traffic going from Seattle to New York: You can drill down in the court, and that will automatically bring up this page where it shows all of the application traffic going from site to site. You can then drill down even further, maybe on streaming video in this case, and then we’re going to give you a profile: Where we’re seeing streaming video, various reports based off of application bandwidth, DSCP values, the conversations, the IP addresses, and so forth. So really use LiveNX to help you during that proof of concept phase.
Now, next, we’re going to be moving into Day 2 Operations. It’s really all about, you rolled out SD-WAN. You now want to verify and understand how your applications are performing, how your sites are performing. Take advantage of our NOC dashboards as well. This is fully customizable. You can lay these widgets out how you’d like. You can see where your sites are, what the statuses of your sites are, on this geo topology map; what your top sites are, what your top interfaces are, and so forth. There are many other widgets that you can add to this palette.
But also important with Day 2 Operations is understanding your capacity planning; really, month over month, how much utilization am I taking advantage of for my various service providers that I have; do I need to add more bandwidth to my provider links at my various sites. Within LiveNX we have the concept of stories. You can think of this like guided workflows. There is the site to site analysis story that we talked about earlier, as well as this WAN interface capacity story. By clicking on that, it will bring up this page, where we’re going to show you your top interfaces at your various sites and how much utilization they’re using based off of the bandwidth that’s available at that site.
Then we’re also going to show you, month over month, what percentage of changes, month over month, at these various sites and the connections to your various service provider links. So notice that New York, at the vEdge router, on Gigabit Ethernet 0/0, there’s a good amount of utilization. You can drill down further by clicking on that interface name, and then we’re going to bring up a bunch of reports for you to really help you understand exactly what’s happening over time at that site connected to that provider link.
So in this case, interface bandwidth, interface errors, utilization, and so forth. We’re going to zoom down to applications as well, inbound and outbound of that particular site. Also, DSCP values; what your handoffs are to your service providers, maybe the MPLS network; what are the DSCP values inbound, as well as outbound; and then the conversations, the top source IPs, as well as the top destination IPs.
Further, use LiveNX in Day 2 Operations by running reports, scheduling them. In this case, here’s an example of an application report at our various sites. In this case, this is the New York site for your WAN interfaces; all the application that your users are taking advantage of in site New York; maybe even running conversation reports as well, where your conversations are going, where your users are going to, whether in the data center or off to the internet. We have a whole host of various reports that you can use. Go ahead and run them ad hoc or even schedule them.
So this is just really a high level overview of being able to take advantage of LiveNX to understand how your applications are being used throughout the network, track the end to end flows, use it for baselining and planning, for deploying the SD-WAN in proof of concept and of course during Day 2 Operations to really understand how your applications are being used by those that are on the network.
So now we’re going to head back over to the slide deck. We have a few concluding slides at this point. At this point we have another poll question for you. How do you see the need for network visibility changing as a result of SD-WAN and cloud deployments? A, do you need additional visibility; B, do your needs remain the same; or C, less visibility because of network simplification? We’d love to hear from you. We’ll keep the poll open for a bit of time. We look forward to hearing all your answers.
Manan: Great. Thank you, David, for a great demo. I’m sure the audience really like the demo and the additional visibility that everyone can get from this LiveAction, LiveNX too, and how that can complement with Viptela’s SD-WAN solution.
One of the questions that a lot of customers often ask is, how do I take this of breed solution right Viptela, from a WAN perspective, building and deployment, and work with LiveAction to get best of breed visibility into this WAN. That question, a lot of times, comes in the early part of the deployment; but as customers deploy the WAN and leverage both of the products together, they quickly realize that Viptela is very good at site activation, [unintelligible 00:29:53] provisioning capabilities, how you bring up different sites and how easy it is to templatize everything and deploy it in bulk and, on top of that, get visibility in performance monitoring like your WAN links with [unintelligible], latency, [unintelligible], MTU changes, with app route and SLA capabilities of what applications are flowing over which channels and how those channels are doing; along with the troubleshooting aspects where you can quickly do some more troubleshooting using vManage and identify where the problem could be; leverage the insight that you got from LiveAction and leverage vManage to troubleshoot the network.
In addition to that, Viptela’s vManage also provides the full lifecycle of your WAN and branch [unintelligible 00:30:48] solution. Then with LiveAction, once it’s deployed and monitoring, the Viptela’s WAN solution provides a lot of different traffic visibility as we showed you in the demo, from an apparent perspective; from a prioritization perspective; from a site perspective; from a provider perspective; as well as from a replay and troubleshooting and historical visibility perspective.
So there’s a lot of complementary functionality between Viptela and LiveAction. So it was natural that our customers wanted to leverage these products together and get the best of breed solution. In addition to the traffic visibility, LiveAction also provides a lot of application analysis; so you can look at a particular application and see how that application has been perform and/or which parts of it have been going along with the capacity planning information. LiveAction does provide multiwindow network visibility as well, so that’s an additional benefit that customers would get.
At this point I do want to go back to David and ask him: For customer who are looking at SD-WAN, David, how can they get the benefit of SD-WAN and leverage some of the capabilities? How do they get started? But before we talk about that, let me just close the poll. Again, we got a lot of good voting; and 99 percent of the users have said they need additional visibility as a result of deploying SD-WAN in the cloud application. So that’s close to 100 percent, and that’s what we’ve heard from a lot of customers. It’s very consistent with the questions that we have been getting on the poll as well.
So with that, I’m going to hand it back to David to talk about how we get you started with troubleshooting or [unintelligible 00:32:56] use case for deploying Viptela WAN and LiveAction.
David: Okay, perfect. Thanks for that, Manan. So really, why LiveNX? We saw some interesting things on the demo – being able to get that visibility. That’s really what it is. LiveNX has built-in, intuitive visual analytics. Being able to understand all of the flows, all the conversations going on in the network, is so key to our network teams being able to do their job. LiveNX in the back end has, really, a big data backend. So all of the flows that you have: Just sent it LiveNX’s way.
We have best of breed solution to be able to collect 1 million flows per second per node. So per LiveAction node, we have a hierarchical architecture, a distributed architecture. We can put nodes, collector nodes, in various regions. Each one of those nodes can collect up to a million flows per second, all of those conversations. Just send it LiveNX’s way, and we’re going to be able to help you solve those complex problems.
Also, we talked a little bit about the historical playback; things like the NetFlow DVR capabilities where you can sit back, press “Play,” interact with a device, as if it was real time; but it may be replaying flows from maybe a week ago, a month ago, to really identify where the problems are. Go back to the topology view, run reports; use it like a network time machine to see exactly what the traffic scenario looked like, end to end, across the topology; really being able to provide you that visibility, so you know how your applications are performing; giving you that business intelligence.
Then, also, from the very beginning of LiveAction and LiveNX, from 2006, we have a strong heritage for the WAN. Hundreds of our customers rely on LiveNX every day to manage their WAN, and we realize that our customers are going to have a whole mix of various different types of devices, including the Viptela vEdge devices. So use LiveNX to give you that visibility, run the reports, so you really understand exactly what’s happening on your WAN from an application usage perspective or even capacity planning and so forth.
Next, a call to action for all of you folks that were able to join us on the webinar today: Please go ahead and download a copy of LiveNX. We have a free 14-day trial. You can go to liveaction.com/download. Reach out to us, also, at email@example.com as well, because we could be able to understand exactly how you can use LiveNX in proof of concept. If you need a longer trial, our sales team will easily be able to help you, also.
We also have a bunch of links that you can download. Those links are available in the webinar also, so go ahead and take advantage of that.
Manan: Thanks, David. So with that, I just wanted to highlight a few other sessions that we strongly recommend you to attend: What to Ask Your Vendor on Enterprise SD-WAN Capabilities. That’s the next session, starting in less than half an hour. You can also watch all of the recordings and the replay of all the sessions, including the three sessions that have been highlighted here. They were all done earlier in the week, and the recording is available right now on the FutureWan website at viptela.com/futurewan-sd-wan-virtual-summit. The link is also available on the slide, as you can see.
With that, we really want to thank you for your time today. Thanks for joining and listening to us, as well as looking at this wonderful demo from David on LiveAction and how it can provide deep visibility into Viptela’s SD-WAN solution. Thank you.