Landbay successfully migrates to AWS and goes cloud-native with Dynatrace
The value we get from Dynatrace is on a day-to-day basis, proactively preventing outages from occurring in the first place. Dynatrace just gives us instant answers.
Migrating to AWS in a cloud-native architecture
Originally launched with the support of a third-party developer in 2014, Landbay wanted to rebuild its peer-to-peer lending platform from the ground up, to increase agility and gain full control of its underlying ecosystem. As part of this shift, Landbay migrated all systems to Amazon Web Services (AWS) and rebuilt the platform in a cloud-native architecture, based on a growing number of microservices – currently over 30 – running in containers.
The success of this new cloud architecture would be largely determined by its ability to guarantee an excellent user experience for every site visitor. Chris Burrell, Head of Tech at Landbay explains; “Customer transactions in the peer-to-peer lending market are typically low volume, high value. That means that every visitor to our site must be treated like a VIP. Having a fast site, that is easy to use and always available is key to conversions and repeat transactions.”
Optimizing performance in a hyper-dynamic ecosystem
Landbay realized that, as part of its move to the new platform, integrated performance monitoring would be vital for guaranteeing the excellent user experience it required. It therefore needed a solution that could provide full stack application monitoring and that integrated easily with AWS.
“We required visibility across our entire IT stack and we needed to monitor and understand the experience of every visitor to our site. We looked at a number of options, but the alternatives were unsuitable as they were not cloud-native or comprehensive. Dynatrace made sense. Not only does it cover the full stack and provide the deep code-level analysis that we require, but it’s also built in a cloud-native architecture, so inherently supports cloud-native applications,” adds Burrell.
“The Dynatrace dashboard enables me to have a real-time view of the health of our technology estate, as well as providing me with visibility into the activity of our customers and the quality of the experience we’re providing for them,” says Julian Cork, Chief Operating Officer at Landbay.
Another major advantage Landbay saw in Dynatrace was the simplicity with which it can be installed to monitor a hyper-dynamic cloud native infrastructure. Other solutions needed manual configuration from one host to the next, but Dynatrace’s AI capabilities mean it can be launched with a single agent. The single agent auto-discovers all components and dependencies in the application landscape, and auto-instruments performance monitoring across the whole IT stack.
Maintaining control of rapid innovation
Landbay releases two to three code deployments a day to continually deliver new features and functionality to its platform. However, such rapid release cycles increase the risk of performance problems when new code is deployed. To reduce this risk, Landbay has implemented a shift-left paradigm to ensure that extensive testing occurs during multiple stages of development. Dynatrace plays a key role in this process, running extra monitoring once code has been deployed into production. This enables Landbay to instantly identify increases in failure rates and performance degradations and ensure its Engineering team is alerted rapidly, before customers are impacted.
“Making the jump to continuous deployment is only possible when you have confidence in your team’s ability to identify issues before your customers do. You also need confidence in their ability to diagnose the cause of those issues and understand their impact both quickly and accurately. Our partnership with Dynatrace has been invaluable to us in this respect,” explained Jays Shortt, Head of Product at Landbay.
“Dynatrace’s AI capabilities mean that when tests fail, we don’t waste crucial time trawling through logs to identify the root-cause; it just gives us instant answers. In one instance, this insight enabled us to roll back from a deployment that caused our registration journey to fail within just 15 minutes of it going live; limiting its impact to just two or three customers. However, we’ve only experienced two or three significant outages in the past 18 months. Most of the value we get from Dynatrace is on a day-to-day basis, proactively preventing those outages from occurring in the first place,” adds Burrell.
Building a performance culture
Performance management and the process of proactively preventing any service failures has now become so ingrained in Landbay’s culture that Dynatrace is the first thing the Engineering team checks when they arrive at work. Each morning, a designated support team member checks the event log for any performance alerts or notifications of failure rate increases that have been raised overnight and identifies any immediate actions to be addressed.
Looking to the future, Landbay will continue to make improvements to its mobile-first portals to optimize the user experience for customers across all platforms. For example, Landbay can adopt the latest development frameworks within just days of their release, safe in the knowledge that Dynatrace will provide the visibility and support needed to ensure the transition is made successfully.
Summing up the experience of using Dynatrace, Burrell concludes; “We’ve been really impressed by Dynatrace on so many levels. We have a brilliant account manager who is incredibly responsive to our needs and ensures that the customer service we receive is second to none. From a technical standpoint, Dynatrace is also an extremely comprehensive solution that is very cost-effective. Its rapid release cycles mean that it regularly brings new features and functionality to its users.”
The value we get from Dynatrace is on a day-to-day basis, proactively preventing outages from occurring in the first place. Dynatrace's AI capabilities mean we don't have to waste crucial time trawling through logs to identify the root cause when tests fail; it just gives us instant answers.