Access years of the highest quality historical market data, built-in analytics, enterprise-grade features and scalable compute in one place. Built for research, analytics, and data science
Designed by quants. Trusted by the market.
Every challenge, from messy datasets to scalable workflows, is solved with BMLL Data Lab - built by quants for research, execution analysis, backtesting and market surveillance.
Full-depth, harmonised market data
Access full-depth historical market data built direct from source, that preserves all information from the exchange, ready for research, back-testing, and analysis. View the data in the way you need it - either with the raw fields preserved or in a normalised format.
Built for scale
Process petabyte scale historical datasets without infrastructure bottlenecks or performance constraints in a completely managed service, leveraging the latest compute and technologies.
Easy to use libraries
Fully customisable environment containing easy-to-use APIs and built in analytics.
Automation as standard
Move from ad-hoc experiments to daily pipelines using automated scheduled jobs and version controlled notebooks. Seamlessly connect to your own cloud storage and SFTP servers.
World class support
Utilise dedicated expert quant support, pre-configured notebooks and robust documentation on every feature of the BMLL Data Lab to begin deriving insights immediately.
Cloud-native, reproducible environment
Enterprise-grade solution with built in version control, SSO and security features.
Back-test strategies without weeks of data cleaning
Testing and refining trading strategies depends on clean, full-depth historical data, but poor normalisation, changing protocols and inaccurate data mean teams can spend weeks pre-processing before they can run a single back-test.
Work with full-depth historical market data built direct from source, available raw or normalised
Run back-tests on petabyte-scale datasets in a fully managed service, with no infrastructure to maintain
Build and refine models using easy-to-use APIs and built-in analytics
Spend your time testing and refining strategies instead of cleaning data, and move from experiment to repeatable back-test with confidence.
Prove execution quality across venues
Analysing execution performance and meeting Best Execution obligations requires consistent data across venues, but inconsistent schemas and manual data wrangling make reproducible, defensible analysis hard.
Analyse execution across venues using harmonised data that is consistent across markets
Build reproducible TCA and Best Execution workflows with programmatic access
Measure execution against real market conditions using full-depth order book data
Produce consistent, reproducible execution and TCA analysis you can stand behind, without rebuilding the data pipeline each time.
Study liquidity and price formation at scale
Studying liquidity, order book dynamics and price formation means working across enormous historical datasets, processing petabyte-scale data normally demands significant infrastructure and dedicated resources.
Study order book dynamics and price formation with full-depth, harmonised historical data
Process petabyte-scale datasets without infrastructure bottlenecks or performance constraints
Explore the data raw or normalised, whichever your research requires
Work in a customisable environment with easy-to-use APIs and built-in analytics
Research liquidity and market structure across markets and history at scale, without building or maintaining the underlying infrastructure.
Optimise routing with normalised cross-market data
Improving smart order router performance depends on high-quality data normalised across many markets. Assembling and maintaining that data, and the computer to model on it, is a major engineering burden.
Use high-quality historical data normalised across multiple markets to identify trends and test routing logic
Back-test SOR models in a scalable, fully managed compute environment
Model against realistic conditions using full-depth order book data direct from source
Identify market trends and back-test SOR models on normalised cross-market data, all within one scalable environment.
Explore new markets without the engineering cost
Evaluating new venues and asset classes usually means buying, sourcing and engineering new datasets before you can even begin, a slow, expensive commitment made before you know whether the market is worth it.
Explore new venues and asset classes without buying, sourcing or engineering the data yourself
Work with harmonised historical data that is consistent across markets from day one
Scale analysis up or down in a fully managed environment with no infrastructure to maintain
Assess unfamiliar markets and asset classes quickly and cost-effectively, without committing engineering resources up front.
Explore the technical capabilities of the BMLL Data Lab — sign up for a trial and see what’s available before moving to the full version.
|
Trial
Explore the environment |
Enterprise
For Scaling teams
Full production capabilities |
|
|---|---|---|
| Instance Sizes |
Up to 16 GB |
Up to 1.5TB |
| Datasets |
All available* |
Custom selection |
| Scheduling |
|
|
|
Shared Areas
Secure, partitioned workspaces for cross-functional quant and research teams to collaborate on code and datasets.
|
Multiple shared directories allowed |
|
|
Clusters
Elastic MapReduce (EMR) allows you to run distributed data processing frameworks like Apache Spark for massive-scale historical analytics.
|
Run large scale EMR clusters (64 cores per node) |
|
| Support |
Standard |
Dedicated |
| GPUs |
Multiple available for selection |
|
| SFTP | ||
| Cloud Connection |
|
|
|
* Some markets not available in trial due to exchange licensing. |
||
Discover how BMLL helps trading desks optimise execution and market insight
