Xignite delivers real-time financial data at scale, while realizing a 50% cost savings

YugabyteDB is the open source, high-performance distributed SQL database for global, internet-scale apps. Schedule a demo, or learn more at yugabyte.com

Xignite is the leader in market data cloud solutions, providing real-time financial data to more than 700 financial services and fintech customers.
With a growing dataset and customer base, SQL Server on AWS became expensive, and scaling out proved to be very challenging.
With YugabyteDB and Yugabyte Platform, Xignite moved to a cost effective database solution, while achieving the high scalability and performance it needs.

Xignite is transforming market data infrastructure with cloud-native financial data distribution and market data management solutions.

Xignite provides financial market data, including real-time and reference market data, to its customers through cloud-based APIs.

The Xignite platform runs on AWS and is available in two different service offerings: a data-as-a-service (DaaS) solution that provides customers access to Xignite’s real-time and reference datasets, and a market data management-as-a-service (MDMaaS) solution that financial companies can use to integrate their own real-time consolidated feeds and proprietary datasets.

Xignite’s platform currently supports more than 250 different data sources and 12 billion API calls per day for more than 700 customers, including BlackRock, Robinhood, SoFi, Investopedia, Betterment, and other financial innovators.

Challenges

  • Running SQL Server on AWS with Amazon RDS became expensive
  • Scaling SQL Server is very difficult, especially with a large and growing dataset and customer base
  • Trying out MySQL reduced cost compared to SQL Server, but didn’t improve scalability
  • Achieving consistency with a Redis cache in front of SQL Server or MySQL to scale read queries slowed down development velocity
  • Long running scans of data would affect performance of lookups by overwhelming the cache and thereby affecting query latency SLAs

Key Database Requirements

  • Cloud native and ability to run on AWS in Xignite’s VPC
  • Low TCO in production
  • High scalability, especially as Xignite’s datasets grow
  • Available as a managed service to reduce the burden on
    the ops team, while providing enterprise support
YugabyteDB gave us the ability to store more data than we could store on SQL Server. This means that YugabyteDB made some of our new use cases possible that we could not have done with our previous database solution.
Qin Yu
Senior Vice President of
Engineering and Operations,
Xignite
It’s been a very great experience working with Yugabyte. We highly value the relationship, as well as the support we get from the Yugabyte team. With other vendors, sometimes the response time or level of help is just not quite there. But working with Yugabyte, we get support in real time or near real time, and the support level is great.

YugabyteDB Solution

4 fully managed clusters
240 total cores and
11+ TB of data
21 total nodes in single region, multi-zone deployments
Easily deployed and managed in Xignite’s
AWS VPC

Technical Results

  • Easy to scale out, especially as dataset grows
  • High performance reads and writes for a large amount of data
  • Scan-resistant Least Recently Used (LRU) cache prevents large
    scans from affecting low-latency queries
  • Flexible upgrades, without downtime

Business Results

  • Cost effective database solution, with overall cost savings of approximately 50% compared to SQL Server
  • Simplified and streamlined operations with Database as a Service
  • Flexibility to support multiple workloads, including real-time data and reference/fundamental data use cases, across both Xignite’s DaaS and MDMaaS product lines
Stephane Dubois
CEO, Xignite
The ability to scale and rebalance very large datasets without downtime or performance bottlenecks is critical to our customers and our business. YugabyteDB helps us focus on growing our business instead of maintaining a complex caching and relational database architecture.