Blog article
See all stories »

Stream If You Quant To Go Faster

Below, I outline the four key capabilities that should be present in a modern quant research analytics platform. Relevant for both business users, seeking better trading outcomes for themselves and their customers, and quant teams looking to develop innovative new data-centric analyses to create an edge for their business.

1: Single platform data store

Data is often fragmented around an organization meaning quant teams can spend more time managing data than analyzing it. By choosing a streaming analytics platform that can store deep historical and real-time data in one platform, applications can access data much faster to make quicker, better-informed decisions. Also, a single platform reduces the need to create custom data pipelines – leading to better data integrity – and it’s generally much easier to manage.

2: Common data-science language capabilities

It used to be the case that specialist technical tools were required to have real-time access to data for quant research, but not anymore. If you’re using a platform where you can’t bring your own tools to work, such as Python and SQL, you’re making it harder to develop the applications and trading strategies needed to stay ahead of the competition.

3: Real-time access to data for faster insights

Processing and managing large volumes of data can be slow and time-consuming. But, whether it’s for algo trading, signal generation, pre or post-trade analytics, backtesting or model development, the faster the time to insight the faster the time to value. Look for a platform that has the flexibility to scale storage and compute without compromising on speed.

4: Complex Event Processing (CEP) streaming

Data pipelines are still often batch-oriented rather than streaming. This makes it hard to analyze data in-flight without running multiple queries which adds latency and affects performance. CEP streaming means teams of quants and data scientists can ingest massive amounts of ticking, streaming data in-flight, embed models into streaming data workflows and have that data immediately available. For client-facing businesses, I would consider this a critical capability.




Comments: (0)

James Corcoran

James Corcoran

VP Engineering


Member since

23 Mar 2021



Blog posts


This post is from a series of posts in the group:

Big Data

this Group Focusing Big Data Area and Data Engineering /Data Warehouse /Data Analytics

See all

Now hiring