Increasing maximum number of Logical Processors in a Processor Group greater than the default (64)

Question

Can the maximum number of Logical Processors allowed in a Processor Group be increased by Bcdedit command or registry key to a number greater than the default (64) to allow non-group-aware applications to seamlessly scale via multithreading across more cores
MSDN documents methods to reduce the max number of LP's in a Processor Group to 1 or any power of 2 less than 64 in order to increase the number of Processor Groups.  But there is no discussion of how to increase the maximum number of Logical Processors
allowed in a Processor Group to a number greater than the default (64), e.g. to 128 or 256. 
Since a non-group-aware application/process can only run on 1 Processor Group, increasing this limit is needed to allow non-group-aware multithreaded applications to scale seamlessly for parallel computation across more than 64 cores simultaneously on systems
with more than 64 cores (for example, a quad-socket Intel Xeon system with 24 cores per socket, including true and hyper cores).  If that limit can be set via registry key or command line, or if there is another way to force all the cores on
all 4 sockets (4 NUMA nodes in this case) to join in 1 Processor Group, then non-group-aware applications will not need to be rewritten using group-aware thread-affinity APIs.  

Answer 1

Hi John,
Are you trying to specify a static NUMA-node-to-group assignment, rather than letting Windows dynamically assign NUMA nodes to groups at boot time? If yes, You can refer the
following related KB:
How to manually configure K-Group assignment on multiprocessor machines
http://support.microsoft.com/kb/2506384
Systems with fewer than 64 logical processors always have a single group, Group 0. On systems with 64 or fewer processors, existing applications will operate correctly without
modification. Applications that do not call any functions that use processor affinity masks or processor numbers will operate correctly on all systems, regardless of the number of processors. To operate correctly on systems with more than 64 logical processors,
the following kinds of applications might require modification:
•Applications that manage, maintain, or display per-processor information for the entire system must be modified to support more than 64 logical processors. An example of
such an application is Windows Task Manager, which displays the workload of each processor in the system.
•Applications for which performance is critical and that can scale efficiently beyond 64 logical processors must be modified to run on such systems. For example, database
applications might benefit from modifications.
•If an application uses a DLL that has per-processor data structures, and the DLL has not been modified to support more than 64 logical processors, all threads in the application
that call functions exported by the DLL must be assigned to the same group.
By design, a non-NUMA computer is considered to have one NUMA node. Because NUMA nodes cannot span groups, the system creates a node for each group after you restart the computer.
The related KB:
Processor Groups
http://msdn.microsoft.com/en-us/library/windows/desktop/dd405503(v=vs.85).aspx
More information:
Uneven Windows Processor Groups
http://blogs.msdn.com/b/saponsqlserver/archive/2011/10/08/uneven-windows-processor-groups.aspx
Boot Parameters to Test Drivers for Multiple Processor Group Support
http://msdn.microsoft.com/en-us/library/windows/hardware/ff542298(v=vs.85).aspx
I’m glad to be of help to you!
We
are trying to better understand customer views on social support experience, so your participation in this
interview project would be greatly appreciated if you have time.
Thanks for helping make community forums a great place.