Trading Point Breaks New Ground with AWS

With an increasing need to improve scalability and security, reduce costs, and improve its ability to innovate, global trading platform provider Trading Point realized a rethink of its technology infrastructure was needed.

The company also wanted to expand its operations to the US as efficiently as possible, while still providing a high-quality client experience. Trading Point—which has its headquarters in Cyprus and operates the well-respected xm.com and Trading.com online trading platforms, serving more than five million clients in 30 languages—soon realized a cloud-based approach with Amazon Web Services (AWS) was the best way to address these needs.

Scaling with the existing infrastructure required Trading Point’s IT team to carefully plan and continually, proactively monitor its efforts and still came with the ever-present risk of possible downtime. 

“Our traffic is very spikey because of financial announcements, so being able to scale up and down to meet demand is very critical for us,” says Vasilis Hadjisophocle, deputy chief information officer at XM. 

Furthermore, applications were sharing resources, and that was impacting service reliability. It was crucial to Trading Point’s business that its infrastructure maintained a high-level of availability, meaning reliability was particularly important. 

The provisioning of new resources or changing existing resources was so time consuming it was difficult to develop new services to quickly meet customer requirements. 

Trading Point’s plans to expand the Trading.com brand to the US, meanwhile, needed a more flexible cloud-based infrastructure that would allow it to do so in a scalable way and provide high availability and low latency to customers.

The decision was taken to migrate some key xm.com workloads from the company’s existing hosted collocated server environment to Amazon Web Services (AWS), with migration taking place in September 2020. 

The company now runs data and analytics, databases, and some applications to support the xm.com platform. There are plans to move more of the xm.com workloads via the AWS Migration Acceleration Program (MAP). Back-end logic and servers remain with the existing hosting provider. 

In contrast, the Trading.com platform has been built entirely on AWS. “Because it’s a different approach and client experience, we needed to deploy it on AWS to be cloud native, and to have the flexibility for expansion and scale,” says Chrysanthos Kattimeris, CTA at XM. 

As well as many of the core AWS services—such as Amazon Elastic Compute Cloud (Amazon EC2) and Amazon Relational Database Service (Amazon RDS)—Trading Point uses AWS Elastic Beanstalk for deploying and scaling web applications and Amazon Cognito for user and client authentication. 

Amazon Redshift forms part of the company’s data analytics toolset, while AWS Transit Gateway enables the existing data centers to interconnect with the AWS infrastructure, allowing secure internal communication between the remaining on-premises infrastructure and non-production ready systems based on AWS. 

There are some common services across xm.com—which operates across Europe, Asia, and Latin America—and Trading.com, which is available in the UK, as well as the US. However, customization was needed to meet local regulations.

With fluctuating demand, AWS allows Trading Point to scale up and down on demand. The new infrastructure is more robust and means Trading Point can manage its infrastructure consistently. 

With the trading platforms running on AWS they can handle large amounts of traffic when needed, such as when the Non-Farm Payroll report—which measures the number of jobs added or lost in the US economy—is released on the first Friday of every month. 

“Having an infrastructure that will auto-scale, meet their demands, and is more robust will definitely have an indirect effect on client retention and attracting new clients,” says Kattimeris. 

Moving to AWS also means Trading Point has infrastructure in the locations where it’s needed. This has reduced application latency during peak hours and reduced the resources needed to maintain the environment. 

Trading Point’s 100-plus team of software developers is now able to deploy new functionality more quickly. 

The company uses AWS CloudFormation infrastructure as a code to grow its capabilities faster with less maintenance. This approach also brings greater consistency to the products that are deployed. 

“Infrastructure as a code gave us the ability to deploy faster without additional delays, and we have more control. It also makes us scalable, so we deploy identical environments for testing, staging, and production,” says Hadjisophocle. 

It’s also become quicker to deploy the infrastructure that supports new functionality: “Deploying infrastructure in a dedicated data center takes months to complete. Because AWS is provided as a service, we can provide the network infrastructure, databases, the service cluster, and everything we need, in zero time,” says Kattimeris.