20240028417. EXECUTING FEDERATED WORKFLOWS FROM EDGE TO CORE simplified abstract (HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP)

From WikiPatents
Jump to navigation Jump to search

EXECUTING FEDERATED WORKFLOWS FROM EDGE TO CORE

Organization Name

HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP

Inventor(s)

SATHYANARAYANAN Manamohan of Bangalore (IN)

SATISH KUMAR Mopur of Bangalore (IN)

KRISHNAPRASAD LINGADAHALLI Shastry of Bangalore (IN)

RAVI Sarveswara of Bangalore (IN)

EXECUTING FEDERATED WORKFLOWS FROM EDGE TO CORE - A simplified explanation of the abstract

This abstract first appeared for US patent application 20240028417 titled 'EXECUTING FEDERATED WORKFLOWS FROM EDGE TO CORE

Simplified Explanation

The patent application describes a system and method for orchestrating machine learning workflows across multiple resource partitions or domains. The workflows are composed of multiple tasks and can be executed across different silos. The state of the federated workflow is maintained and shared through a distributed database or ledger, such as a blockchain. Local agents deployed at the silos can orchestrate the workflow within their respective resource domains, accessing the workflow state through the blockchain. The system can operate regardless of the heterogeneity of resources within a silo.

  • The invention enables the orchestration of machine learning workflows across multiple resource partitions or domains.
  • The workflows consist of multiple tasks and can be executed across different silos.
  • The state of the federated workflow is maintained and shared through a distributed database or ledger, such as a blockchain.
  • Local agents deployed at the silos can orchestrate the workflow within their respective resource domains.
  • The agents have access to the workflow state through the blockchain, which acts as a globally visible and consistent state store.
  • The system can operate even if the silos have heterogeneous resources or aspects.

Potential Applications

  • Federated machine learning workflows across multiple domains or organizations.
  • Collaborative machine learning projects involving multiple parties.
  • Distributed machine learning systems with resource partitions.

Problems Solved

  • Orchestration of machine learning workflows across multiple resource partitions or domains.
  • Sharing and maintaining the state of federated workflows.
  • Handling heterogeneity of resources within a silo.

Benefits

  • Improved efficiency and scalability of machine learning workflows.
  • Enhanced collaboration and coordination among multiple parties.
  • Flexibility to operate in heterogeneous resource environments.


Original Abstract Submitted

systems and methods provide for a federated workflow solution to orchestrate entire machine learning (ml) workflows comprising multiple tasks, across silos. in other words, one or more sets/pluralities of tasks making up an ml workflow, can be executed across multiple resource partitions or domains. federated workflow state can be maintained and shared through some form of distributed database/ledger, such as a blockchain. agents that are locally deployed locally at the silos may orchestrate an ml workflow at a particular resource domains, each such agent having access, via the blockchain (acting as a globally visible/consistent state store), to the aforementioned workflow state. such systems are capable of operating regardless of the existence of heterogeneous resources/aspects of a silo.

Retrieved from "http://wikipatents.org/index.php?title=20240028417._EXECUTING_FEDERATED_WORKFLOWS_FROM_EDGE_TO_CORE_simplified_abstract_(HEWLETT_PACKARD_ENTERPRISE_DEVELOPMENT_LP)&oldid=29079"