Heaps and stacks in distributed shared memory

Abstract

Software-based distributed shared memory (DSM) systems do usually not provide any means to use shared memory regions as stacks or via an efficient heap memory allocator. Instead DSM users are forced to work with very rudimentary and coarse grain memory (de-)allocation primitives. As a consequence most DSM applications have to "reinvent the wheel", that is to implement simple stack or heap semantics within the shared regions. Obviously, this has several disadvantages. It is error-prone, timeconsuming and inefficient. This paper presents an all in software DSM that does not suffer from these drawbacks. Stack and heap organization is adapted to the changed requirements in DSM environments and both, stacks and heaps, are transparently placed in DSM space by the operating system.