The NYU Ultracomputer project continues to pioneer the study of architecture and software for large-scale, shared-memory parallel computers. During this past year, we have achieved several very significant milestones, most notably we fabricated and used that first-ever combining switches and we increased our industrial involvement. Other important accomplishments include porting our Symunix operation system to Ultra 3 prototypes; further developing a very high quality, portable C compiler needed for our prototypes, that has attracted considerable commercial attention; producing a fast solver for Laplace's equation on multiply connected domains; and furthering the analysis of buffered interconnection networks and parallel random number generators. In addition to further developments in the areas mentioned above, we plan two new activities for next year. First, we will obtain extensive measurements of the effect of combining on scientific and other application software using both the Ultra 3 hardware prototypes and a new simulation environment that we are presently constructing. Our successful VLSI development of combining switches has already demonstrated that the additional cost for combining is about 100% using modest packaging 50% with 300 pins, and zero given next generation densities and 400 pins. Hopefully, these chips will refute the often-quoted claim that combining increases themore » cost of the network by a factor of between 6 and 30. Second, we will port our operating system to the new NCR series of Intel 486-based multiprocessors. NCR has agreed to donate a machine during the first quarter of 92 for this effort, which will strengthen the ties between our project and NCR.« less