OPA driver optimizations are based on the MPI model where it is
expected to have multiple endpoints between two given nodes. To
enable this optimization for Lustre, we need to make it possible,
via an LND-specific tuneable, to create multiple endpoints and to
balance the traffic over them.
Both
OPA driver optimizations are based on the MPI model where it is
expected to have multiple endpoints between two given nodes. To
enable this optimization for Lustre, we need to make it possible,
via an LND-specific tuneable, to create multiple endpoints and to
balance the traffic over them.
Both
On Thu, Mar 15, 2018 at 02:53:46PM -0700, Doug Oucharek wrote:
> OPA driver optimizations are based on the MPI model where it is
> expected to have multiple endpoints between two given nodes. To
> enable this optimization for Lustre, we need to make it possible,
> via an LND-specific tuneable, to
On Thu, Mar 15, 2018 at 02:53:46PM -0700, Doug Oucharek wrote:
> OPA driver optimizations are based on the MPI model where it is
> expected to have multiple endpoints between two given nodes. To
> enable this optimization for Lustre, we need to make it possible,
> via an LND-specific tuneable, to
OPA driver optimizations are based on the MPI model where it is
expected to have multiple endpoints between two given nodes. To
enable this optimization for Lustre, we need to make it possible,
via an LND-specific tuneable, to create multiple endpoints and to
balance the traffic over them.
Both
OPA driver optimizations are based on the MPI model where it is
expected to have multiple endpoints between two given nodes. To
enable this optimization for Lustre, we need to make it possible,
via an LND-specific tuneable, to create multiple endpoints and to
balance the traffic over them.
Both
6 matches
Mail list logo