New Podcast
Managing Traffic During Peak Demand; Plus, Microsoft, Akamai Outages

Customer Stories

How Orange Business Services Ensures Business Continuity During COVID-19

By Ian Waters
| | 27 min read

Summary


How Orange Business Services Ensures Business Continuity During COVID-19

In this video presentation, Sonia Gameiro, Head of Application Performance Management Consulting Practice - Europe at Orange Business Services, shares how Orange Business Services supported its customers’ business continuity needs during COVID-19 lockdowns. During the early period of the pandemic, Orange Business Services needed to increase its global network capacity by 25%, to cope with an increase of 35% in Internet traffic from their users. At the same time, they also observed that the number of users remotely connecting to their company's networks increased by 700%. Gameiro discusses how her team leverages ThousandEyes end-to-end visibility to proactively monitor the availability and performance of IT resources—such as VPN gateway and Wi-Fi connectivity, SaaS applications and ISP performance—both for its external customers and internally.

Sonia Gameiro of Orange Business Services speaks at ThousandEyes Connect.

Follow Along with the Transcript

Sonia Gameiro:
So it's a pleasure for me to be here today and to have the opportunity to share on one side a testimonial of what COVID-19 disruption meant for Orange Business Services. And on the other side, how do we see the monitoring space as a key component for companies to manage business continuity.

Sonia Gameiro:
So to introduce the topic I've put together in a slide what the Orange Business Continuity Plan meant in figures. So this is really to illustrate what we have all been, one Internet transit network. We have registered an increase in Internet traffic volume of 35%. So this is really quite substantial. And in order to support this, we had to actually increase our capacity by 25% globally. So our backbone capacity, and by 50% in the trans-Atlantic connectivity, which is also an interesting data. So secondly, and in order to support the remote connectivity due to the lockdown policies across the world, we have observed that the number of users remotely connecting to their company's networks increased by 700%. So at Orange Business Services, we have reached around 95% of the employees working from home. And probably most of your organizations have gone through the same. The usage of remote collaboration solutions we mentioned Zoom increase that Ian mentioned and others like Webex Teams, etc.. They have increased also by 20 to 100% depending on the solutions, which is also, of course, substantial. And we also observed that audio conferencing, simultaneous calls increased by 250%. On the cybersecurity side, where Orange is also very, very active. We have observed in a three week period of time, our cybersecurity teams identified 8,000 new suspected domain name identifiers and sites. And all of them registered with the terms, COVID and Coronavirus. So most of them, of course, being used for fraudulent purposes. In France alone, our teams managed up to 130 additional customer operations or interventions per day around networks, data centers and all of the services that Orange is managing. So a really substantial increase also in our activity. And all in all, nearly 3,500 Orange Business Services employees were mobilized to manage our customers' business-critical activities. So this can give you an idea through these figures on how hard we've been working over the last two months to support our customer business continuity plans, quite substantial.

Sonia Gameiro:
So, in order to illustrate the reactivity that we had to do to use in order to support our customer's business continuity plans. I would like to share with you this customer use case, concrete customer use case. So this is a story of a customer, which is a large industrial manufacturing company with activities in several sectors: energy, health care, industry, etc., and with operations in over 150 countries. So the situation, of course, was the lockdown policies that had employees to be confined at home. And which had to work remotely. So and this really almost from one day to the other. So how did we help? In only five days, our teams had graded the customer's remote access capacity from 25,000, which was a nominal utilization, up to 72,000 remote users across all regions around the world. And we've done that actually in 2 steps. First, by increasing the capacity of the existing physical infrastructure up to the maximum of its capacity. And then this was still not enough. So we had to implement 4 additional virtual remote access gateways in Microsoft Azure. And then working with all of these gateways as a cluster to really reach the level of capacity of the customer was needing. And all of this was done in 5 days. So this indeed illustrates the kind of urgent request that we had to handle. And the scale of these requests. So this was, this is a bit closing the topic on how did we handle the COVID-19 disruption. And I would like now to introduce the topic, which is dear to all of us, here in this call, which is that the visibility and the and the monitoring topic. And I will do that by starting with a statement that actually applies at all times, not only during this disruption, which is that you can only manage what you can see. So proper monitoring is really the only way for you to be able to manage what you are mandated to manage. If you don't have that visibility, you will be blind to do so and you will probably not have the means to do that properly. So today I'd like to show you how we see the ThousandEyes visibility, help you better manage business continuity during a global disruption like the one we're living.

Sonia Gameiro:
So the lockdown and the massive working from home policies brought in new IT challenges to the IT organizations. These are multiple. Here, I'm only focusing on 2 of them. The first one is around the availability of specific IT resources. And we've mentioned them before. They are typically VPN, for example, remote access or collaboration applications. And we have seen the scale of how these have been increasing in use because of these massive working from home policies. And the other topic is around performance, of course. So how is the productivity of your users being impacted when they're working from home? And the challenge, as well, is that IT organizations don't control home connectivity. So what happens, as Ian mentioned, when the children are doing remote education or when someone else in the family is watching a film on Netflix? So how are the home connectivity conditions so that the employees can continue to be productive? This is a big challenge for IT organizations. Also, how is the application performance doing from home? Because it can differ from when these applications are accessed from the office. So we need to have proper visibility on all of this in order to manage and to make the needed actions in order to improve the user experience as much as we can, as IT organizations. So how does the ThousandEyes monitoring help going more into concrete examples? Bottom line, by providing actionable data which enables IT to better manage the disruption.

Sonia Gameiro:
So this actionable data is really important because it's interesting to have the data. But what really matters at the end of the day is what can you do with it? How will that help you enable a better user experience and the business continuity at the end of the day? So we find again, our two main IT challenges that I would like to focus on today. The first around the availability of specific IT resources and the second round performance while working from home. So I have split, let's say, the two columns in the first one you will see. What are the questions that the ThousandEyes monitoring solution allows to answer to? And on the right, you will see what are the enablers or the actionable, let's say, benefits that it allows IT organizations. So on the availability of specific IT resources, we have again the VPN usage and failures. What is that per region? How can we monitor that? How can we react quickly when there are issues, collaboration apps, as we mentioned as well, or virtual desktop infrastructure, so also very important because many organizations, they didn't have really laptops to provide to all of their employees and they have chosen to allow their employees to connect to virtual desktops to be able to continue working from home. So that's also a key, a key point.

Sonia Gameiro:
So on the enablers, we can do, of course, better VPN capacity management and then react on application availability issues quickly. At the end of the day, we're also very interested in going from a reactive incident management approach to a proactive incident management approach in order to minimize disruption for the users and for the business. On the performance space, the questions are what is the performance per employee and per application? So here we're more focused on indeed on the endpoint visibility where we need to have this granularity per user because it will be sitting at home and there is no other way actually of measuring their experience besides actually equipping their endpoints to see what's going on there. And where the performance issues come from? So the path from an endpoint and a user up to the application server and the databases behind can be quite long and complex. So when we have performance issues, where do these performance issues come from? How many people are impacted? And then we can, of course, try to restore normal operations as quickly as possible. Also, the comparison between the application performance when connecting through VPN, when being on a site or through split tunneling when we're at home with VPN and then exiting directly the Internet without going back to the corporate network.

Sonia Gameiro:
So in terms of the enablement, it's crucial for the IT support organizations first to do diagnosis of the issues and also to prioritize the incidents depending on how many people are impacted. As an IT support organization, you will want first to treat the incidents, which impact hundreds and thousands of employees rather than incidents that impact one single employee. Secondly, regarding the performance of the on-prem applications, we can also, as IT organizations try to optimize that performance for remote access precisely because typically latency can be higher or the bandwidth conditions can be very variable. So we can probably optimize this content so that it reaches the users in the most effective way. Also very important during crisis management. Having the needed data in order to communicate relevant guidelines to the users. And I'm pretty sure you have all been involved in this communication to the users. Of course, that could be much more relevant. When do you know what they're doing and when you know what they're experiencing.

Sonia Gameiro:
So that's it on the, a little bit on the overview on how a ThousandEyes can help. And I thought of illustrating this with some dashboard examples. And of course, Steve already showed very valuable disperse that he produced with an IT group. Maybe this is also as a compliment and it's focussing only on endpoints dashboard. But here, for example, I also added a dashboard, the one on the left, which comes from Cloud Agents. So typically on the VPN availability and performance, on the left, we see the status of the VPN gateways throughout the world. And we see actually how available they are with a very simple color code: green, yellow, red, etc.. And this is coming from Cloud Agents monitoring. On the right, we already have an endpoint view with the latency towards the VPN in which the user is connecting into, and we can have that visibility on a per-region level—so North America, Europe, Asia-Pacific, etc—and per-user level, if we want, we can also group the users and have a per country proceeding as we wanted to show it.

Sonia Gameiro:
So this is an example of a dashboard. Another example, again, around collaboration applications, availability, and performance. So here on the left, probably a bit small. I hope you can still see it. But on the left, overall, we have a Team's dashboard showing the performance and availability towards the front-end web server. And we have actually split it per country. But here again, we can actually group the data in the way we want. And on the right, we have some network information towards the Teams at transport relay. So we can have views on packet loss, latency, jitter from the user to the Microsoft team's transport relay. So very useful information where we want to troubleshoot collaboration application issues in terms of both availability and performance.

Sonia Gameiro:
The next dashboard is about remote workforce experience score. So here, again, it is very important to understand how the users experience application performance. So on the left, we have a worldwide map showing the end-user experience score, again with a simple color code so that you can see per region how your users are experiencing application performance. And on the right, we have a card view of the same data per region and per, in this case, per SaaS application. We are monitoring specific SaaS applications—Google, Salesforce, SharePoint Online—whatever you want to monitor. And then we will also see a color code per application. So this is to manage overall the remote for workforce experience. And Steve also mentioned that at IG Group they have equipped all of the workforce. This is something that you can actually extract the maximum potential indeed when you have a global deployment so that you can immediately pinpoint where the issues come from, from a regional perspective and geographical perspective.

Sonia Gameiro:
And the last dashboard that I would like to show you today is around the home connectivity performance, which is really a challenge for IT support organizations with a ThousandEyes Endpoint Agent, you can see a sort of simplified diagram between the user's computer and whatever visited site the user is accessing and you will see the intermediate hop connectivity element. So the Wi-Fi with the signal. So if you have a specific poor signal, you will be able to tell the user you have a poor Wi-Fi signal. You need to get closer or remove something that it's a has been put in front or between the Wi-Fi router and where you're standing. And the gateway here, again, very important because when you have several people at home, let's say fighting for the same bandwidth, as education, as was said before, remote education or Netflix or any other application, you will see that translating into an increase of packet loss and latency at the level of the gateway, which is actually the home router. So it's also something that you can see through the Endpoint Agent analytics. And when you're solving incidents, when people are calling saying, I can't access this, you will be able to see this and tell the user to try to improve the situation of his home connectivity.

Sonia Gameiro:
So that was it's on the dashboard examples. And now I would like to end up with a couple of slides around the Orange and the ThousandEyes partnership. So we are really proud to have started our partnership with ThousandEyes in 2018. And back then, Orange actually became a customer of a ThousandEyes for its own internal monitoring needs. And then through 2019, as we realized that the solution was really relevant and they had great potential to answer our customer challenges, we signed a partner agreement to be able to resell their solution to our customers. And here the vision of our top management was actually to help our customers expand their visibility into SaaS applications, which are the new applications. Let's say expansion's worlds, public cloud hyperscalers and the remote workforce as well. And of course, the end-to-end Internet underlay connectivity, which, as Ian very well said, still is a quite unpredictable environment. So the story of this partnership can actually be very well illustrated by the use cases that we're addressing with ThousandEyes as a company. So we can structure them in two types. On one side, we have internal use cases and on another side, we have our customer use cases. So on the internal use cases, this is how our partnership actually started back in 2018, with our open transit Internet monitoring needs. So this is our Tier 1 global transit network, Internet network, and we use ThousandEyes to monitor the backbone and also from different affiliates towards applications to make sure that when using our transit network, the performance was as expected.

Sonia Gameiro:
Then in 2019, we became, well, we expanded our internal monitoring needs into our certified ISP monitoring needs. So this is the way that we are certifying ISPs. We're doing it by measuring the latency from these ISPs throughout the world, in all the countries of the world towards our next generation hubs, which are gateways between the Internet and our backbone. So through that, through a ThousandEyes, we can tell our customers that we can guarantee a specific SLA in terms of latency between their site connecting through this specific ISP and the closest next-generation hub. And this allows us to build an automatic Internet weather map with a different color code, depending on the latency to the code to the closest next-generation hub. So this was the second use case in which we became a customer of ThousandEyes. And now we are working in equipping our operational teams to better manage the services that we're providing to customers. So typically, when we're providing Internet links to our customers, whether we're using SD-WAN or not, we are using ThousandEyes in order to monitor that underlay, end to end.

Sonia Gameiro:
And then on the customer use cases. So we have the digital experience monitoring to cloud service providers from the customer sites. Very important, this is really the use case that most of our customers are interested in today, making sure that when you're doing local breakouts through the Internet from a branch that end-to-end path towards your cloud service provider, including the hop-by-hop details and the cloud service provider performance in itself. This is really what most of our customers are looking for today, and we're helping them in this journey. Also around the reward for workforce monitoring as was presented today and then to finalize the consulting engagements. So we are using ThousandEyes for a number of engagements around which our short-term engagements are delivered by consulting my team. In order to help them really activate the value of the data, which can be extractives from ThousandEyes and a very good example of that is what we're doing currently with Philips. And I will let Divya tell you everything about it right after this session. So that's it from my side. Matt, back over to you.

Matt:
Sonia, thank you for a very informative session. We do have a couple of questions which I'd like to ask you. We've got time for this. The first question that's come in is does OBS provide ThousandEyes as a service for your customers? Or are you using it for in-house only?

Sonia Gameiro:
So, yes, we are providing as a service for customers. We are starting with consulting services and we are at this very moment building the fully managed service, including the operational support typically to handle the incidents and the alerts that will result from ThousandEyes data, ThousandEyes alerts. But today we are already delivering consulting services which can cover perfectly well the build and the run of a ThousandEyes deployment. And we have, of course, developed through the last year, a lot of expertise in that.

Subscribe to the ThousandEyes Blog

Stay connected with blog updates and outage reports delivered while they're still fresh.

Upgrade your browser to view our website properly.

Please download the latest version of Chrome, Firefox or Microsoft Edge.

More detail