Groups | Search | Server Info | Keyboard shortcuts | Login | Register [http] [https] [nntp] [nntps]

Groups > linux.kernel > #1682811

Re: [RFC v2 0/5] surface heterogeneous memory performance information

Show all headers | View raw

On 07/06/2017 04:08 PM, Jerome Glisse wrote:
>> So, for applications that need to differentiate between memory ranges based
>> on their performance, what option would work best for you?  Is the local
>> (initiator,target) performance provided by patch 5 enough, or do you
>> require performance information for all possible (initiator,target)
>> pairings?
> 
> Am i right in assuming that HBM or any faster memory will be relatively small
> (1GB - 8GB maybe 16GB ?) and of fix amount (ie size will depend on the exact
> CPU model you have) ?

For HBM, that's certainly consistent with the Xeon Phi MCDRAM.

But, please remember that this patch set is for fast memory *and* slow
memory (vs. plain DRAM).

> If so i am wondering if we should not restrict NUMA placement policy for such
> node to vma only. Forbid any policy that would prefer those node globally at
> thread/process level. This would avoid wide thread policy to exhaust this
> smaller pool of memory.

You would like to take the NUMA APIs and bifurcate them?  Make some of
them able to work on this memory, and others not?  So, set_mempolicy()
would work if you passed it one of these "special" nodes with
MPOL_F_ADDR, but would fail otherwise?


> Drawback of doing so would be that existing applications would not benefit
> from it. So workload where is acceptable to exhaust such memory wouldn't
> benefit until their application are updated.

I think the guys running 40-year-old fortran binaries might not be so
keen on this restriction.  I bet there are a pretty substantial number
of folks out there that would love to get new hardware and just do:

	numactl --membind=fast-node ./old-binary

If I were working for a hardware company, I'd sure like to just be able
to sell somebody some fancy new hardware and have their existing
software "just work" with a minimal wrapper.

Back to linux.kernel | Previous | Next — Previous in thread | Next in thread | Find similar | Unroll thread

Thread

[RFC v2 0/5] surface heterogeneous memory performance information Ross Zwisler <ross.zwisler@linux.intel.com> - 2017-07-07 00:00 +0200
  [RFC v2 3/5] hmem: add heterogeneous memory sysfs support Ross Zwisler <ross.zwisler@linux.intel.com> - 2017-07-07 00:00 +0200
    Re: [RFC v2 3/5] hmem: add heterogeneous memory sysfs support John Hubbard <jhubbard@nvidia.com> - 2017-07-07 08:00 +0200
      Re: [RFC v2 3/5] hmem: add heterogeneous memory sysfs support Ross Zwisler <ross.zwisler@linux.intel.com> - 2017-07-07 18:40 +0200
  [RFC v2 5/5] hmem: add performance attributes Ross Zwisler <ross.zwisler@linux.intel.com> - 2017-07-07 00:00 +0200
  [RFC v2 2/5] acpi: HMAT support in acpi_parse_entries_array() Ross Zwisler <ross.zwisler@linux.intel.com> - 2017-07-07 00:00 +0200
    Re: [RFC v2 2/5] acpi: HMAT support in acpi_parse_entries_array() "Rafael J. Wysocki" <rafael@kernel.org> - 2017-07-07 00:20 +0200
      Re: [RFC v2 2/5] acpi: HMAT support in acpi_parse_entries_array() Ross Zwisler <ross.zwisler@linux.intel.com> - 2017-07-07 00:30 +0200
        Re: [RFC v2 2/5] acpi: HMAT support in acpi_parse_entries_array() "Rafael J. Wysocki" <rafael@kernel.org> - 2017-07-07 00:40 +0200
  [RFC v2 1/5] acpi: add missing include in acpi_numa.h Ross Zwisler <ross.zwisler@linux.intel.com> - 2017-07-07 00:00 +0200
    Re: [RFC v2 1/5] acpi: add missing include in acpi_numa.h "Rafael J. Wysocki" <rafael@kernel.org> - 2017-07-07 00:10 +0200
  Re: [RFC v2 0/5] surface heterogeneous memory performance information Jerome Glisse <jglisse@redhat.com> - 2017-07-07 01:10 +0200
    Re: [RFC v2 0/5] surface heterogeneous memory performance information Dave Hansen <dave.hansen@intel.com> - 2017-07-07 01:40 +0200
  Re: [RFC v2 0/5] surface heterogeneous memory performance information John Hubbard <jhubbard@nvidia.com> - 2017-07-07 07:40 +0200
    Re: [RFC v2 0/5] surface heterogeneous memory performance information Ross Zwisler <ross.zwisler@linux.intel.com> - 2017-07-07 18:40 +0200
  Re: [RFC v2 0/5] surface heterogeneous memory performance  information Balbir Singh <bsingharora@gmail.com> - 2017-07-07 08:30 +0200
    Re: [RFC v2 0/5] surface heterogeneous memory performance information Dave Hansen <dave.hansen@intel.com> - 2017-07-07 18:30 +0200
    Re: [RFC v2 0/5] surface heterogeneous memory performance information Ross Zwisler <ross.zwisler@linux.intel.com> - 2017-07-07 18:30 +0200

csiph-web