May 10, 2022
No image
Completed
DATA PROCESSING PLATFORM
$75,000+
7-12 months
United States
6-9
Service categories
Service Lines
Big Data
Domain focus
Other
Technology
Programming language
JavaScript
Python
Challenge
Creation of several platforms and environments, solving the problem of accessing computing power and research data by data scientists.
Creation of several platforms and environments, solving the problem of accessing computing power and research data by data scientists.
Solution
We have created a platform that has solved the problems of accessing tools, computing power, and the environment, and analyzed data by data scientist experts. Each specialist can now analyze data using our utilities in an isolated environment, located in cloud computing services (AWS and on-prem servers).
We have created a platform that has solved the problems of accessing tools, computing power, and the environment, and analyzed data by data scientist experts. Each specialist can now analyze data using our utilities in an isolated environment, located in cloud computing services (AWS and on-prem servers).
Results
We have analyzed, customized, and tuned the Kubeflow platform, per customer's requirements to provide a more promising and flexible system for data science experts’ use. We have created an automated system for versioning settings for clusters, based on the helm. The system makes archive settings for each installed version, which allows a user to control the environment's predictability and reproducibility.
We have analyzed, customized, and tuned the Kubeflow platform, per customer's requirements to provide a more promising and flexible system for data science experts’ use. We have created an automated system for versioning settings for clusters, based on the helm. The system makes archive settings for each installed version, which allows a user to control the environment's predictability and reproducibility.