17809079. ACCELERATOR SERVICE CATALOG AND ORCHESTRATION simplified abstract (Dell Products L.P.)

From WikiPatents
Jump to navigation Jump to search

ACCELERATOR SERVICE CATALOG AND ORCHESTRATION

Organization Name

Dell Products L.P.

Inventor(s)

Stephen J. Todd of North Andover MA (US)

Victor Fong of Melrose MA (US)

Benjamin E. Santaus of Somerville MA (US)

Brendan Burns Healy of Whitefish Bay WI (US)

ACCELERATOR SERVICE CATALOG AND ORCHESTRATION - A simplified explanation of the abstract

This abstract first appeared for US patent application 17809079 titled 'ACCELERATOR SERVICE CATALOG AND ORCHESTRATION

Simplified Explanation

The patent application describes a method for using an accelerator service to improve the performance of a workload. Here is a simplified explanation of the abstract:

  • The method involves identifying an accelerator service instance associated with a workload.
  • A service broker is called to obtain information needed to use the accelerator service instance.
  • An accelerator call and job information are received from the workload.
  • A new process is created specifically for the accelerator job.
  • The accelerator job is run using either the accelerator service instance or a locally available accelerator.
  • The data generated by running the accelerator job is returned to the workload.

Potential applications of this technology:

  • High-performance computing: This method can be used to accelerate computationally intensive tasks in fields such as scientific research, data analysis, and simulations.
  • Machine learning: Accelerators can significantly speed up training and inference processes in machine learning models, improving efficiency and reducing training time.
  • Video processing: Accelerators can be utilized to enhance video encoding, decoding, and transcoding processes, enabling faster and higher-quality video processing.

Problems solved by this technology:

  • Performance improvement: By utilizing accelerator services, the method addresses the challenge of improving the performance of workloads that require intensive computational tasks.
  • Resource optimization: The method allows for the efficient utilization of accelerator resources by spinning up dedicated processes for accelerator jobs, ensuring optimal resource allocation.

Benefits of this technology:

  • Faster processing: By leveraging accelerator services, the method enables faster execution of computationally intensive tasks, leading to reduced processing time.
  • Improved efficiency: The method optimizes resource allocation by utilizing dedicated processes for accelerator jobs, resulting in improved overall system efficiency.
  • Scalability: The use of accelerator services allows for scalable performance, enabling the processing of larger workloads without sacrificing speed or efficiency.


Original Abstract Submitted

One example method includes identifying an accelerator service instance associated with a workload, calling a service broker associated with the accelerator service instance to obtain information needed to use the accelerator service instance, receiving an accelerator call, and accelerator job information concerning an accelerator job, from the workload, in response to the accelerator call, spinning up a new process dedicated to the accelerator job, as part of the new process, running the accelerator job using either the accelerator service instance, or a locally available accelerator, and returning data, generated by running the accelerator job, to the workload.