Tengri Data Platform

Tengri Data is an enterprise analytics platform.

Challenges

While time-tested tools such as PostgreSQL exist for the operational activities of large companies, choosing the best solution for data analytics is still a major challenge.

Traditional database management systems can handle up to 1TB of analysed data and support 10-20 analysts. However, further growth in analytics needs, the introduction of AI and ML, and an increase in the number of users and data domains can severely increase the timeline and reduce the reliability of results.

For more information on the 3V big data issues — Volume, Velocity, Variety

Traditionally, when working with big data, a list of the 3 V problems has been identified:

1. Volume

The data to be analysed is getting bigger and bigger, we are talking about tens and hundreds of TB.

2. Velocity

There are increasing demands for speed of response, the number of analysts and models analysing the data is growing. The velocity of data growth is also increasing.

3. Variety

The number and variety of data source formats is increasing.

Later, other V issues were added to this list of basic problems, including:

  1. Veracity — reliability of data

  2. Variability — variability of data

  3. Visualisation — visualisation of data

  4. Value — value of data

  5. Venue — where the data is stored and used

  6. Vocabulary — the need for a common glossary for working with data

  7. Vagueness — uncertainty of some data types

All these limitations and challenges point to the need for a fundamentally new approach when building corporate systems for analysing big data.

What Tengri offers

Separation

The separation of Compute & Storage principle enables data storage and compute resources to scale flexibly and independently without multiples of infrastructure costs and adaptively fit into today’s hybrid enterprise installations.

diagram 1
Each request has its own resource: analytics, reporting or machine learning services do not have to compete with each other
— Nikolay Golov
Tengri Data Product Director

Convenience

Using familiar language SQL and standard connections ODBC/JDBC, analysts can get answers in seconds — whether the data occupies 1 GB or 1 PB, and whether there are competing queries or not. And open data formats provide maximum flexibility and scalability.

Analysts can focus on solving business problems without being distracted by the technical nuances of the platform.
— Mikhail Tyurin
Tengri Data Department Director

Result

The success of this approach in the global market is confirmed by the commercial achievements of Snowflake and Databricks, whose solutions are not available to companies in Russia. Tengri provides similar and sometimes superior capabilities using an even more advanced technology stack.

We’re building a platform that opens the door to truly unlimited analytics, removing barriers to unlocking the full potential of data.
— Nikolay Golov
Tengri Data Product Director