Singnification improvements in real-time efficientcy have been obtained for plasma fluid turbulence calculatins by microtasking the nonlinear fluid code KITE in which they are implemented on the CRY Y-MP C90 at the National Energy Research SUpercomputer Center (NESC) The number of processors accessed concurrenly scales linearly with problem size. Close to six concurrent processors have so far been obtained with a three-deinensional memory size of 80 Mword. With a calculatinn size corresponding to eht maximum allowed mwmory of 200 Mword int he next system configuration, we expect to be able to access close to ten procssors of the C90 cocurrently with acommensurate implement in real-time efficiency. These improvements in performance are comparable to those expected from a mssively parellal implementation of the same calculations on the intel paragon