+ <para>Network request scheduler. A subcomponent of the PTLRPC layer, which specifies the
+ order in which RPCs are handled at servers. This allows optimizing large numbers of
+ incoming requests for disk access patterns, fairness between clients, and other
+ administrator-selected policies.</para>
+ </glossdef>
+ </glossentry>
+ <glossentry xml:id="NUMAdef">
+ <glossterm>NUMA
+ </glossterm>
+ <glossdef>
+ <para>Non-uniform memory access. Describes a multi-processing architecture where the time
+ taken to access given memory differs depending on memory location relative to a given
+ processor. Typically machines with multiple sockets are NUMA architectures.</para>