Your Browser is Out of Date

Nytro.ai uses technology that works best in other browsers.
For a full experience use one of the browsers below
Accelerating ML and DL Pipelines with Intel® Granulate™ and Dell PowerEdge Servers on Cloudera Data Platform
Introduction Technology overview
Granulate architecture
Validation overview Workload and results Configuration details Conclusion References Intel notices and disclaimers

Granulate architecture

Granulate architecture

  • Figure 1 shows a SaaS-based solution architecture of Granulate. The Granulate agent collects profiling metrics and performance telemetries before sending to the Granulate Cloud. The data is encrypted at transit and rest, and is logically separated between clients. An Extract, Transform, Load (ETL) process continuously runs against this profiling data to build an optimization configuration based off the client data. This configuration once created is stored within Granulate's S3 bucket for the agent to pull from to define where to optimize and what optimizations to load.

    Diagram of a High-level SaaS solution architecture

    Figure 1.   High-level SaaS solution architecture