neutral A buzzword applied to Sun's Java programming language that is meant to imply its ability to execute under a number of computer architectures without requiring recompilation or code changes.
Microcontrollers based on the Harvard Architecture have separate data and instruction busses. This allows execution instructions to occur in parallel. As an instruction is being "pre-fetched", the instruction is executing on the data bus. Once the current instruction is complete, the next instruction is ready. This pre-fetch theoretically allows for much faster execution than a Von-Neuman architecture, but a certain amount of added silicon complexity exists. The PPC 405 core is built on Harvard architecture. Microcontrollers based on the Von-Neuman architecture have a single "data" bus that is used to fetch both instructions and data. Program instructions and data are stored in a common main memory. When such a controller addresses main memory, it first fetches an instruction, then fetches the data to support the instruction. The two separate fetches slow down the controller's operation.
(As applied to computers), the structured organization of elements in a computational mechanism -- for example, the difference between sequential processing and parallel distributed processing machines is said to be a difference in architecture. See also Computation, (non-) von Neumann Architecture.