Team Internet
<svg xmlns="http://www.w3.org/2000/svg" xmlns:xlink="http://www.w3.org/1999/xlink" x="0px" y="0px" viewBox="0 0 200 150" style="enable-background:new 0 0 200 150;" xml:space="preserve"> <path d="M90.3,23.9l-25,5.8v21.1h-9.7v25.7H65c0,0-0.3,26.6-0.3,34.1s3.9,18.3,22.5,19.1s23,0,23,0v-25.2l-12.5-0.1 c-3.9-0.1-7.1-3.1-7.4-7l0.4-21.4l21,0.1c4,0.2,7.1,3.4,7.3,7.4l-0.4,46.6h25.2c0,0,0.6-53.9,0-62.5c-0.4-6.4-3.5-16.4-19.6-16.4 s-33.4-0.5-33.4-0.5L90.3,23.9z" fill="currentColor"/> <circle cx="130.3" cy="33.9" r="14.1" fill="currentColor"/> </svg>
Founded in 2010 and based in Munich, Germany, Team Internet serves more than 2 billion visitors per month with its two products available via its self-service platforms: ParkingCrew and TONIC.
Team Internet partnered with Catchpoint to:
- Ensure digital availability and reachability.
- Optimize operations and drive strategic growth.
- Observe what is happening beyond their network and across the Internet.
Catchpoint Network Insights gives us confidence about what is happening with both our internal systems and on external servers, so we can proactively ensure availability and reachability.
Problem
Using ParkingCrew, the market leader in the domain parking space, clients can monetize their domain portfolios. To support more than 5,000 page-requests per second, ParkingCrew runs its own name server as well as a specialized web server in two data centers, and processes data at Amazon Web Services (AWS).
Through TONIC., clients and partners can monetize their native, social, search, display and organic website traffic with the best search feed in the industry. As one of the biggest marketplaces for buying and selling traffic, TONIC. processes 30,000–40,000 requests per second.
One of the companyʼs guiding principles is to automate as much as possible. Thatʼs how it hit $100 million revenue in 2017 with just 30 employees.
Expand monitoring reach
As Andreas Lunz, Head of Technology for Team Internet explains, the availability of the company’s services is essential to containing costs and recognizing revenue. Team Internet must answer all requests as fast as possible otherwise its partners’ automated systems start to move traffic to other providers. “If our services are unavailable, we try our best to reduce the impact on all of our partners. Being down a single day easily leads to a six-figure loss for us.”
To try to mitigate such events, Team Internet deployed a second data center in Canada as a failover to complement its main data center in Munich. It uses AWS as a large-scale load balancer between the two data centers. It also uses ServerGuard24 for basic monitoring of server status.
Since integrating its second data center, Team Internet realized it needed more robust monitoring. This became especially apparent when it began seeing a larger number of Distributed Denial of Service (DDoS) attacks. It began using a DDoS mitigation service from Neustar, which took over Team Internet’s BGP announcements. However, there were times it lost traffic for 10-20-minute stretches.
“We wanted to understand what was going on and knew we needed high-grade monitoring that gave better insight into issues beyond our network, including DDoS attacks and other potential issues,” says Lunz
Solution
With Datadog as a partner for internal monitoring and the decision not to build a solution in house and distract its lean team from the core business, Team Internet evaluated Catchpoint’s Network Observability solution. Lunz quickly discovered that this out-of-the-box solution would address a major visibility gap for Team Internet by combining Border Gateway Protocol (BGP) monitoring with Domain Name System (DNS), and Traceroute.
With Catchpoint, Team Internet can ensure that its widely distributed architectural components – ISPs, wireless networks, cloud providers, and BGP routes – all work together seamlessly to ensure its servers are available and that its services are available and reachable at all times.
Ensuring availability and reachability
Now the company has a nearly immediate and clear view into situations that it was once blind to, such as providers being unable to reach its platform. Specific incidents include a UK provider being unable to reach Team Internet because of an issue with a BGP announcement. More recently, traffic from Russia was routed unexpectedly.
“We can’t see these occurrences solely using ServerGuard24 because it operates in just a few locations. With Catchpoint, we can check from so many more locations around the world,” says Lunz.
“We were never able to monitor reachability in this detail before using the Catchpoint solution. Now we can figure out whether an issue is related to a BGP announcement, routing or something else instead of blindly trusting our partners that everything is working as it should,” he says.
Gaining monitoring flexibility
According to Lunz, Catchpoint stands out because of its ability to check intervals down at a granular level.
“We need the flexibility to check network conditions every 5 or 10 seconds so we can zero in on specific issues, and most solutions only check one time per minute.”
Appreciating support for SSL tests
Team Internet must synchronize all SSL to its data centers, even those where it houses its own servers. With SSL tests enabled via Network Insights, it can terminate SSL and mitigate CPU calculations on data center servers, even those hosted by AWS.
“This will allow us to monitor how long it takes to synchronize our certificates in all data centers so we can take measures if the synchronization will be too slow,” Lunz continues.
Streamlining on-call responses
Integration between Catchpoint and OpsGenie, Team Internet’s incident management tool, ensures that on-call staff is only woken up for critical issues. Plus, because Catchpoint’s BGP data is delivered in real time, Team Internet can be alerted to route anomalies as they happen so they can detect and remediate them quickly.
“The staff on on-call rotation are the same people developing our solutions so seeing what is happening beyond our network and across the Internet is very valuable,” says Lunz.
Results
Optimizing operations and driving strategic growth
Deeper monitoring and better insights have empowered Team Internet to increase the availability of its services.
“We can’t fix issues in networks beyond our control but we can see them and tell our partners so they proactively plan accordingly,” Lunz explains.
In fact, Team Internet is now empowered with a level of foresight that enables it to operate at a new level.
“We can recognize and locate issues and the networks where they are occurring and keep our partners up to date so they can optimize their traffic purchases. We can even spot potential issues in advance, such as by seeing rising latency in a region,” he continues.
In the past, Team Internet’s partners and the sales team often notified Lunz’s team about availability issues.
“Now we are aware of the issues before anyone else. This will help us deliver a new level of service and prevent huge losses due to issues that happen outside of our network. Our IT Operations Lead exports graphs and BGP announcements from Catchpoint so our data center teams are empowered when reaching out to peering and upstream partners,” Lunz explains.
In fact, Lunz and his team are making use of the insights from Catchpoint to strategically plan for Team Internet’s expansion.
“With Catchpoint, we can detect latency issues and make informed decisions about setting up data centers in new regions. We are already using the insights to optimize our name servers and increase DNS responses. We’ll do the same with our edge servers to move even closer to users,” Lunz concludes.
Empowering the business intelligence team
Lunz anticipates that Team Internet will get additional value from Catchpoint Network Observability solution going forward. Currently, the company writes about 1 TByte of data every day but the Business Intelligence (BI) team wants to focus on areas that can positively impact the business. With Catchpoint's network observability solution, Lunz can provide the BI team with more detailed information so the team can pinpoint potential issues ahead of time.
“This will also empower them to make smart decisions about where we should deploy new data centers to expand revenue,” he explains.