In recent years, the way people consume transportation has begun to look very different as ride-share and car-sharing services multiply.

With this vastly changed mobility landscape, automakers have come to realise that they must change their offerings as well if they are to avoid being relegated to the role of hardware manufacturers. Thus was born the Share Now car-sharing service, a joint venture between Daimler AG and BMW Group (which grew out of the companies’ prior independent services car2go and DriveNow). 

Headquartered in Berlin and offering services in 16 major cities across eight European countries, Share Now employ just one app to put a free-floating 20,000-vehicle fleet at the fingertips of drivers who can rent cars by the minute, hour, or day—insurance, fuel, and maintenance included—before dropping them off on a city street.

Conquering every city

From its roots as an internal project in the Daimler automotive organization, the car-sharing service quickly morphed into a startup company building an IoT service on top of their parent company’s vehicles. Freed to create its own IT infrastructure, Share Now’s primary technical challenge has always been one of scalability.

Explains Share Now head of cloud operations Erik Bak-Mikkelsen, ‘To enter this business, scalability is key because you need to conquer every city before they’re reached by your competitors—which are coming not just from the automotive industry, but also from public transportation and even IT companies like Google. The entire mobility sector—a billion-dollar industry—is being transformed, and everyone wants a share.’

Adopting a cloud-agnostic approach to speed service delivery

To ensure a winning position, Share Now wasted no time moving their IT environment to the cloud. Knowing that to deploy software quickly and efficiently, Share Now needed to consume the best, the cheapest, the fastest, and the highest-capacity cloud services available—depending on a project’s particular needs—the car-sharing company adopted a cloud-agnostic approach. 

Today, that means using primarily Amazon Web Services (AWS) and IBM Cloud for Share Now’s cloud platform, and a Kubernetes container orchestration system to manage, scale, and deploy the approximately 90 microservices that make up the vast majority of their environment. Share Now also use Amazon Elastic Kubernetes Services (EKS) and IBM Kubernetes Service for container orchestration as well as Puppet for configuration management and Terraform for provisioning.

Gaining end-to-end observability

At the heart of all of this is New Relic, providing an observability platform that enables Share Now to oversee the full breadth of their digital business.

Stephan Kaufmann, DevOps Chapter Lead for Share Now explains, ‘In the old days, when you had a single monolithic application, it was doable to monitor it in Java Virtual Machine or elsewhere. In a microservices architecture, however, you need something that looks at how things are working between these services. New Relic provides that, delivering insights that enable us to understand wider application performance. For example, is it the connectivity from Microservice A to B that’s slowing response time, or is it my own database or something in the infrastructure that’s causing the problem? With New Relic, it becomes easy to answer these questions.’

‘In the old days, when you had a single monolithic application, it was doable to monitor it in Java Virtual Machine or elsewhere. In a microservices architecture, however, you need something that looks at how things are working between these services. New Relic provides that, delivering insights that enable us to understand wider application performance.’

Stephan Kaufmann, DevOps Chapter Lead, Share Now

Like many companies these days, Share Now are using metrics, events, logs, and traces to help make decisions and improve its business. This is because with multi-cloud, Kubernetes clusters, customer-facing and backend application environments, and huge volumes of operational data, it’s no longer enough to monitor a bag of metrics from a disconnected set of tools.

With New Relic, Share Now can leverage a single platform to see how all their applications, underlying services, and systems relate. As a result, dependencies across organisational boundaries are easy to observe, and troubleshooting becomes a breeze—pinpointing not just the problem itself but also the underlying issue.

Delivering real-time data analysis

This is critical for a company like Share Now, which process a new rental every second, making real-time data analysis absolutely essential. Says Bak-Mikkelsen, ‘It's important to follow our customers with their smartphones and their need for transportation, so that we can make sure we have vehicles where they’re needed when they’re needed. While our initial use of New Relic was for traditional Web traffic and Web application monitoring, we’re now using it to monitor our telematics, our vehicle connectivity, our footprint in the GSM network, and more. We use it everywhere that we have data in text or number format that we can analyse.’

Obviously, this would not be possible if New Relic were not easy enough to use that just about anyone in the company—not just the technology teams—could access the data it provides. 

‘When we are scaling systems or introducing new services, we use New Relic for both quality assurance and to satisfy our curiosity about how the systems or services are performing,’ says Bak-Mikkelsen. ‘But that’s just the beginning. Oftentimes, departments have dashboards running in the background on monitors, reloading automatically, just to follow development. In these instances, New Relic is almost like a stock market ticker.’

Providing company-wide insights

Because users in teams ranging from business intelligence to accounting can easily create New Relic dashboards to collect and analyse relevant data, Share Now stakeholders are obtaining knowledge that they could previously only dream about, according to Bak-Mikkelsen. 

‘With New Relic, all we have to do is support the platform to make sure it’s running,’ says Bak-Mikkelsen. ‘This is the strongest and most significant paradigm shift from traditional reporting and analysis tools: While those tools can do the job, they’re static rather than active. Using New Relic is like moving from an acoustic guitar to an electric guitar.’

‘With New Relic, all we have to do is support the platform to make sure it’s running. This is the strongest and most significant paradigm shift from traditional reporting and analysis tools: While those tools can do the job, they’re static rather than active. Using New Relic is like moving from an acoustic guitar to an electric guitar.’

Erik Bak-Mikkelsen, Head of Cloud Operations, Share Now 

Delighting customers and optimising operations

Not surprisingly, with issues now being resolved more swiftly, Share Now have been able to exponentially improve customer experience with their application and infrastructure observability since adopting New Relic. As a result, the DevOps team are seeing their own productivity soar as well.

Says Kaufmann, ‘When we can give our developers the functionality to immediately recognise how the customer is experiencing our app in real time, we have an early warning system that enables them to correct bottlenecks before the customer even notices them. They no longer need to solve a whole chain of problems; all they need to do is look into New Relic.’

And that in turn frees up Bak-Mikkelsen’s and Kaufmann’s teams to focus on Share Now’s primary business. Explains Bak-Mikkelsen, ‘Platform monitoring used to be very complex: You had to install and configure agent software; you needed specific knowledge of the component you were monitoring. New Relic, in contrast, is almost like a spray can: It’s ready to go out-of-the-box, and your employees can spend their time developing features and delivering value rather than configuring their monitoring tool.’

Optimising operations

With outages now a rarity, Share Now are increasingly using New Relic to optimise their environment. ‘Now we’re able to analyse the running phase’, says Bak-Mikkelsen. ‘Where can we optimise? Where can we gain speed? Where can we reduce costs? Now, when management asks us how the system performed over the past week, we can provide an immediate answer by simply looking at our New Relic dashboard. This is brilliant.’

In summary, Bak-Mikkelsen says, ‘It's the speed with which you can process data. It's how fast you can gain an overview of complex relationships. And it’s the simplicity with which you can offer this application. These are the benefits of New Relic.’