Volume rendering by ray casting is computationally expensive. For inte
ractive volume visualization, rendering must be done in real time (30
frames/s). Since the typical size of a 3D dataset is 256(3), parallel
processing is imperative. In this paper, we present an O(log n) EREW a
lgorithm for volume rendering. We use O(n(3)) processors that can be o
ptimized to O(log(3) n) time with O(n(3)/log(3)n) processors. We have
implemented our algorithm on a MasPar MP-1. The implementation results
show that a frame of size 256(3) is generated in Ils by 4096 processo
rs. This time can be further reduced by the use of large number of pro
cessors.