MWS Data Lakehouse

The main articles are:

• Big Data
• Data mining Data mining

MWS Data Lakehouse is a Cloud Native platform that allows you to independently scale storage and computing layers, which makes it possible to maximize the available resources and place petabytes of data in it. Various solutions can be used as an infrastructure for deploying the platform, including Kubernetes, object S3 compatible storage, which allows companies to reduce costs by 40% compared to hosting in classic DWH storage. The platform also has built-in tools for centralizing access control, audit and encryption, and dynamic masking of sensitive data that fully meet modern information security requirements.

2025: MWS Data Lakehouse platform launch

MWS Cloud, part of MTS Web Services, July 21, 2025 announced the launch of the data storage and processing platform - MWS Data Lakehouse. The platform allows you to work with any type of data - structured, unstructured and vector, which makes it possible to create a single environment for performing various types of tasks: from building operational reports to training and inference of ML models and LLM. The platform allows you to use storage 40% more economically by eliminating duplication of data between different systems, 2.5 times increase the efficiency of service personnel by centralizing and automating administration functions and 2 times reduce the time for calculating analytical storefronts.

The platform allows you to store and process any type of data, such as information sales and orders, demand for goods and, logistics information about payments and services provided. On top of the platform, you can run tools for training and inference ML and. For example, large language models deploy a model to predict demand for goods based on accumulated sales information, or predict the return percentage, or credits find the most optimal supply route. The company can deploy both its services to work with and AI use ready-made MWS MLOps tools.

The platform is compatible with Greenplum and Postgres. This enables customers to retain their historical assets, shorten the migration cycle, and reduce the costs and risks associated with migrating to a new solution.

The platform has support for a large range of data, including open source Apache Parquet and Iceberg formats. The system is optimized for a variety of workloads: from SQL queries to machine learning tasks. To improve efficiency, it is possible to run several computing clusters in parallel for different product commands, applications and types of requests without duplication of data and additional replication.

The service is administered through a single interface that allows centralized management of users, clusters and scaling. Dynamically changing compute cluster resources is supported, ensuring continuity and no downtime while scaling.

Tens of thousands of Western companies are already using the Big Data solution. MWS Data Lakehouse enables companies to store data 40% more efficiently than DWH and Data Lake, reduce the calculation time of analytical storefronts by 2 times, and save IT time by 2.5 times by centralizing and automating administration functions. The platform architecture allows companies to work with both structured and unstructured data, which is especially important for training and infering artificial intelligence models. This makes the introduction of AI into the company's business processes almost impossible without using the platform, "said Igor Zarubinsky, CEO of MWS Cloud, Executive Director of MTS Web Services.

The platform is part of the MWS data services complex - MWS Data. In total, as of July 2025, it includes more than 25 products for storing, processing and transforming data, as well as business intelligence services and AI agents for working with data.