Advertorial By 2024, the global revenue of fintech will reach €175 billion. 5000 new startups appear in the sphere every year.
Because of this competition, modern technologies are widely utilized in fintech. One of these technologies, app containerization, allows to quickly test and launch new opportunities, grow and modernize services and platforms. But it has a disadvantage as well: the divided structure is harder to manage, which is critical for fintech apps. Luckily, there's a solution, and it's called Kubernetes. In this article, we'll tell you how it allows for faster and simpler development and how it helps to improve and support fintech apps, with specific examples.
Containerization is a method that allows to test and launch separate elements of an app independently, through the core of the operating system. In a more traditional approach, hardware virtualization, the development goes on within virtual machines, each of which needs to emulate virtual hardware and to launch a separate OS. Containers are better for business since they waste less production capabilities and develop and test products faster.
Kubernetes is a software solution that provides automatic management of such containers. It allows creation of deployment templates, launch containers on a high number of hosts simultaneously, and groups them by logic, distributing load between them. Furthermore, another important feature of Kubernetes is error tracking: when one of the containers stops functioning, it is automatically substituted for another, and there is no fault in app operation.
Thanks to its open-source nature, Kubernetes has quickly grown popular among the large organizations and projects. For instance, in 2014, Wikimedia Foundation was one of the first to use this platform instead of an in-house container management system. Currently, Kubernetes is used by over 45% of companies. The convenience, speed and reliability of this solution has been appreciated by finance companies like HSBC, Capital One, Monzo, Starling and ING. Let's take a close look at why large corporations and fintech startups choose Kubernetes.
Security is particularly crucial for fintech companies since they store the most sensitive data of millions of users. Their protection is ensured by a package of measures, an important role which belongs to the distribution of system components in Kubernetes-controlled containers.
Here's an example of the Chinese fintech company Ant Financial, based on one of the world's largest payment services, Alipay. When it needed to transfer data from an outdated infrastructure to a new one, the managers really started thinking about cybersecurity: during peak load, the company processed over 250 thousand transactions per second, and it stored the payment data of 900 million users.
After researching several technologies, the company decided to develop a container-based platform and selected Kubernetes to manage it. This platform simplified the processing of huge volumes of customer data, and it allowed Ant Financial to implement the following security solutions:
"We're witnessing at least a tenfold improvement of operation with this cloud technology,", comments Haojie Hang, Ant Financial product manager. He claims the company isn't even interested in optimizing Kubernetes during the rapid growth stage.
In combination with cloud solutions, Kubernetes provides online business with even more safety and convenience. For example, the G-Core Labs cloud, along with virtual machines and Managed Kubernetes, offers flexible and efficient protection of servers, websites, apps and API from DDoS attacks, as well as a web application firewall. With its help, clients will be able to protect their products from intruder-caused app failures.
The development of fintech apps requires rapid scaling because new clients and partners join them every day. And of course, it's important to ensure user data security in the process, too.
Figo is Europe's first online platform for banking services standardization. In 2019, it partnered with Finreach Solutions to create a SaaS solution: an API platform for financial products. The company's solution supports such operations as financial source analysis, transfers, bank data retrieval, checking the user's account balance, and transaction classification. Figo's partner banks use the platform to automate the provision of services to its clients, like to check their financial accountability.
The platform was not very well-known in the beginning, but gradually many partners joined it. That lead to extra spending on hardware and technical limitations. To optimize its expenses, Figo decided to update the IT infrastructure and move on to cluster architecture, with Kubernetes as the orchestrator.
The platform is now scaled automatically: Kubernetes loads the deployment templates. Meanwhile, the users' data is safely protected since Kubernetes uses secure connections between the nodes.
Fintech apps process thousands of operations per second, including transactions, credit scoring, and user complaints. All these operations have to be performed quickly and precisely. That becomes a real problem when it comes to the daily processing of millions of transactions. However, the American banking company Capital One managed this issue successfully.
The company has already used the platform based on its testing and development containers. For the efficient operation of cloud apps that detect credit fraud and evaluate the borrowers' credit eligibility, the company decided to automate the work with containers. A fault-proof and efficient solution was required - and it was Kubernetes.
"Kubernetes significantly improves the productivity", comments Keith Gasser, Capital One's lead software engineer. "Without this solution, our expenses on cloud services would be three or even four times higher. We also reduced time to market. Now, when a developer team comes to us, we can launch its decision-making service within two weeks, while previously it took at least 3 months."
Financial apps must work in an uninterrupted manner, even when the developers are creating a new deploy. At the same time, the specialists also should be comfortable managing the platform.
Several years ago, the American finance company Northwestern Mutual purchased the fintech startup LearnVest and decided to merge its own developments with the innovational platform of the startup. That proved quite difficult: the existing IT infrastructure was outdated, it took weeks to deploy the updates, and the entire development process had to be paused for 6 hours every Sunday to monitor them. Moreover, faults could arise at that time, which paralyzed the operation of the apps for an even longer term.
"We needed to build a scalable but at the same time much more flexible platform, so that we'd be able to transfer the data to the client website faster, and to make sure our end users got what they expected", comments Brad Williams, Northwestern Mutual's UX design lead.
Kubernetes allowed the team to create a convenient architecture, which became a foundation for the autonomous work of the teams which quickly developed and continuously supplied new solutions. Previously it might have taken weeks to deploy infrastructure, and now it's a matter of minutes. The number of deployments has grown sharply: from 24 per year to 500+ just in the first few months of 2017. Moreover, the company has managed to eliminate the downtime for development monitoring: Kubernetes now shows defective containers automatically.
The convenience of Kubernetes can be experienced with G-Core-Labs: the management of virtual machines, platforms or other cloud products is carried out through a single dashboard or API. This helps the clients to avoid struggle with isolated solutions and instead to use a complex ecosystem of cloud services, ready for efficient - and cost-efficient - operation.
Large fintech apps are developed by several teams of developers at once, and it's important that they should be able to create and test their products independently from one another.
The Chinese financing group CreditEase has grown so much in the last 12 years that it felt an urgent need to modernize its apps. The older solutions couldn't manage the current flow of information and were too complex to scale and develop further.
The company decided to switch to containerization development. It worked, but the tens of thousands of containers in Credit Ease's production environment needed a management system. For that, they utilized Kubernetes - which saved the developers a ton of money.
To simplify updating apps and roll them back to the previous versions, CreditEase started packaging microservices in containers and utilized Kubernetes. Thanks to that, teams got the opportunity to introduce changes into their apps independently. Additionally, Kubernetes eliminated downtime: previously, the entire app could break down because of a small mistake in one part of code, and now the platform automatically substitutes the defective container for another one.
Sponsored by G-Core Labs
A changed RMS appeared at the GNU 40th anniversary event in Switzerland
Friday FOSS Fest Fancy running Windows, Linux and Classic MacOS on your modern x86-64 or Arm64 Mac? Walk this way
Sponsored Feature How DeepBrain made the most of Lenovo's AI Innovators Program
Interview Or how one pesky press release ruined a vacation
Also optimises routes and tames crowds, but can't stop that person who just reclined into your knees
Just in time to get Atlassian's latest cross-team collab bits
OpenAI might respect robots.txt but dunno about the others