QuantEdge Analytics Reduced API Latency in Asia by More Than 3 times

About the client

QuantEdge Analytics develops a financial analytics platform that collects market data from multiple sources, processes it, and provides the results through APIs and a web interface. The platform serves customers in both Europe and Asia.

Challenge

Originally, almost all of QuantEdge Analytics’ infrastructure was located in Europe. This architecture worked well for European customers, but as the company’s Asian user base grew, network latency became increasingly noticeable. For some users in Asia, round-trip latency to the company’s API reached approximately 180 – 220 ms.

QuantEdge Analytics considered adding another region with a major public cloud provider. However, the company also wanted predictable infrastructure costs and preferred not to manage several unrelated hosting and data-center suppliers.

Solution

Together with Unihost, QuantEdge Analytics distributed its infrastructure between European and Asian locations.

For its primary European environment, the company chose Unihost OVE-X5 dedicated servers. Each server provides an AMD Ryzen 9 9950X with 16 cores, 128 GB DDR5 ECC memory, two 960 GB NVMe drives, and a 1 Gbps unlimited connection. The configuration is offered across several European locations, including Germany, France, the UK, and Poland.

These systems host the company’s primary:

• databases;
• backend services;
• market-data processing workloads;
• internal infrastructure.

For Asia, QuantEdge Analytics deployed Unihost VNF-05 servers in Singapore.

The VNF-05 provides the same 16-core AMD Ryzen 9 9950X platform with 128 GB DDR5 ECC RAM, two 1.92 TB NVMe drives, and a 1 Gbps connection with 30 TB of traffic.

The Singapore nodes host latency-sensitive API services and other components that benefit from being physically closer to Asian users.

Relevant application data is replicated between Europe and Asia, allowing requests to be served from a geographically closer infrastructure location without moving the entire platform away from its primary European environment.

Results

After the Asian infrastructure went live, latency for customers in the region decreased from approximately 180 – 220 ms to 40 – 70 ms. For some users, this represented an improvement of more than three times.

At the same time, QuantEdge Analytics estimates that the resulting infrastructure costs approximately 25% less than the comparable multi-region public-cloud architecture the company had evaluated. Operational complexity was also reduced. Instead of maintaining relationships with separate European and Asian hosting suppliers, the team manages both locations through a single infrastructure provider.

“For a small engineering team, managing five different infrastructure vendors becomes a job of its own. Having multiple geographic locations through one provider makes our operations much simpler.”

What’s next

QuantEdge Analytics is considering a third infrastructure region, a dedicated backup server, and additional DDoS protection for its public-facing APIs.

Results at a glance: 180 – 220 ms → 40 – 70 ms latency, approximately 25% lower cost, Europe and Asia through one provider