17958216. SHARED LOCAL REGISTERS FOR THREAD TEAM PROCESSING simplified abstract (Intel Corporation)

From WikiPatents
Jump to navigation Jump to search

SHARED LOCAL REGISTERS FOR THREAD TEAM PROCESSING

Organization Name

Intel Corporation

Inventor(s)

Biju George of Folsom CA (US)

Fangwen Fu of Folsom CA (US)

Supratim Pal of Folsom CA (US)

Jorge Parra of El Dorado Hills CA (US)

Chunhui Mei of San Diego CA (US)

Maxim Kazakov of San Diego CA (US)

Joydeep Ray of Folsom CA (US)

SHARED LOCAL REGISTERS FOR THREAD TEAM PROCESSING - A simplified explanation of the abstract

This abstract first appeared for US patent application 17958216 titled 'SHARED LOCAL REGISTERS FOR THREAD TEAM PROCESSING

Simplified Explanation

The abstract describes an apparatus for shared local registers for thread team processing, including a graphic processor with multiple processing resources and memory for data storage. The processor allocates thread teams to processing resources, provides a shared local register space accessible only to the thread team, and allocates individual register spaces to each thread.

  • Graphic processor with multiple processing resources
  • Allocation of thread teams to processing resources
  • Shared local register space directly referenced in ISA instructions
  • Individual register spaces allocated to each thread
  • Accessible shared local register space for thread team
  • Inaccessible shared local register space for threads outside of the team

Potential Applications

The technology could be applied in:

  • High-performance computing
  • Graphics processing
  • Artificial intelligence

Problems Solved

This technology addresses issues such as:

  • Efficient thread team processing
  • Resource allocation optimization
  • Data access control

Benefits

The benefits of this technology include:

  • Improved performance in multi-threaded applications
  • Enhanced data security within thread teams
  • Simplified programming for shared resources

Potential Commercial Applications

Potential commercial applications include:

  • Data centers
  • Gaming consoles
  • Supercomputers

Possible Prior Art

One possible prior art could be the use of shared registers in multi-core processors for parallel processing.

Unanswered Questions

How does this technology impact power consumption in comparison to traditional processing methods?

The article does not address the potential impact on power consumption of using shared local registers for thread team processing.

Are there any limitations to the number of thread teams that can be allocated using this technology?

The article does not specify if there are any limitations to the number of thread teams that can be allocated to processing resources.


Original Abstract Submitted

Shared local registers for thread team processing is described. An example of an apparatus includes one or more processors including a graphic processor having multiple processing resources; and memory for storage of data, the graphics processor to allocate a first thread team to a first processing resource, the first thread team including hardware threads to be executed solely by the first processing resource; allocate a shared local register (SLR) space that may be directly reference in the ISA instructions to the first processing resource, the SLR space being accessible to the threads of the thread team and being inaccessible to threads outside of the thread team; and allocate individual register spaces to the thread team, each of the individual register spaces being accessible to a respective thread of the thread team.