18307728. SCHEDULING JOBS ON GRAPHICAL PROCESSING UNITS simplified abstract (HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP)
Contents
- 1 SCHEDULING JOBS ON GRAPHICAL PROCESSING UNITS
- 1.1 Organization Name
- 1.2 Inventor(s)
- 1.3 SCHEDULING JOBS ON GRAPHICAL PROCESSING UNITS - A simplified explanation of the abstract
- 1.4 Simplified Explanation
- 1.5 Potential Applications
- 1.6 Problems Solved
- 1.7 Benefits
- 1.8 Potential Commercial Applications
- 1.9 Possible Prior Art
- 1.10 Original Abstract Submitted
SCHEDULING JOBS ON GRAPHICAL PROCESSING UNITS
Organization Name
HEWLETT PACKARD ENTERPRISE DEVELOPMENT LP
Inventor(s)
Diman Zad Tootaghaj of Milpitas CA (US)
Junguk Cho of Milpitas CA (US)
Puneet Sharma of Milpitas CA (US)
SCHEDULING JOBS ON GRAPHICAL PROCESSING UNITS - A simplified explanation of the abstract
This abstract first appeared for US patent application 18307728 titled 'SCHEDULING JOBS ON GRAPHICAL PROCESSING UNITS
Simplified Explanation
The patent application relates to scheduling jobs for multiple GPUs to provide concurrent processing by virtual GPUs. The computing system receives requests to schedule new jobs, allocates them to virtual GPUs, updates existing job allocations, and minimizes operational and migration costs.
- Efficient scheduling of jobs for multiple GPUs
- Concurrent processing by virtual GPUs
- Minimization of operational and migration costs
- Updating existing job allocations on virtual GPUs
Potential Applications
The technology can be applied in various fields such as:
- High-performance computing
- Data processing and analytics
- Machine learning and AI training
Problems Solved
The technology addresses the following issues:
- Efficient utilization of multiple GPUs
- Minimization of operational and migration costs
- Optimal scheduling of jobs for improved performance
Benefits
The technology offers the following benefits:
- Increased processing power and speed
- Cost-effective operation of multiple GPUs
- Enhanced job scheduling and allocation efficiency
Potential Commercial Applications
The technology can be utilized in industries such as:
- Cloud computing services
- Video rendering and editing
- Scientific research and simulations
Possible Prior Art
One possible prior art could be the use of job scheduling algorithms in parallel computing systems to optimize resource utilization and performance.
Unanswered Questions
How does the technology handle job prioritization among multiple GPUs?
The article does not delve into the specifics of how job prioritization is managed among the virtual GPUs.
What impact does the technology have on overall system scalability?
The scalability implications of implementing this technology in large-scale computing systems are not discussed in the article.
Original Abstract Submitted
Example implementations relate to scheduling of jobs for a plurality of graphics processing units (GPUs) providing concurrent processing by a plurality of virtual GPUs. According to an example, a computing system including one or more GPUs receives a request to schedule a new job to be executed by the computing system. The new job is allocated to one or more vGPUs. Allocations of existing jobs are updated to one or more vGPUs. Operational cost of operating the one or more GPUs and migration cost of allocating the new job are minimized and allocations of the existing jobs on the one or more vGPUs is updated. The new job and the existing jobs are processed by the one or more GPUs in the computing system.