Case study
2,000 News Items a Second for a Global Proprietary Trading Firm
How SquareShift replaced two custom-built programs with a Logstash pipeline sustaining 2,000 news items a second and a Kibana search screen, running on Elastic Stack on Kubernetes.
Book a session →
Client
A global proprietary trading firm that trades across major financial markets using real-time information and data-driven strategies.
Industry
Financial Services
Services
Stack
Elastic Stack · Logstash · Kibana · Kubernetes · Lucene
How SquareShift delivered it.
The challenge
The firm depended on real-time news for its trading decisions, but ran two custom-built programs to handle it — one bringing in thousands of items a second, one letting traders search them. Both were hard to maintain and had reached their limit for further scaling.
What we delivered
SquareShift replaced the ingestion program with a Logstash pipeline built on persistent queues and retry logic, sustaining 2,000 news items a second without losing an item. The search screen was rebuilt in Kibana, replicating the Lucene query support, highlighting, and filtering traders relied on in the original interface.
The outcome
Legacy apps were decommissioned in favor of Elastic Stack running on fully managed, Kubernetes-based services, without losing the throughput or the advanced search features traders depend on. Moving custom-built systems onto Elastic Stack is part of our Elasticsearch migration services.
Retiring a custom-built system only works if the replacement is invisible to the people who rely on it. The measure of success here wasn't the new pipeline — it was that traders never noticed the switch.
Platform & Software Engineering Practice Lead, SquareShift
