In embodiments, apparatuses, methods and storage media (transitory and non-transitory) are described that are associated with receiving a call from an application at a shared library, accessing a first resource based at least in part on the first call, and storing a prefetch entry in a prefetch engine based at least in part on an address of a second resource in preparation to service a second call to the shared library that requires traversal of a plurality of stages at the shared library. A prefetch request may be performed based at least in part on the second call, and the second resource may be accessed based at least in part on a result of the prefetch request. In embodiments, the shared library may be a Message Passing Interface (MPI) library. Other embodiments may be described and/or claimed.