A Parallel Multiresolution Volume Rendering Algorithm for Large Data Visualization?

We present a new parallel multiresolution volume rendering algorithm for visual-izing large data sets. Using the wavelet transform, the raw data is rst converted to a multiresolution wavelet tree. To eliminate the data dependency between proces-sors at run-time, and achieve load-balanced rendering, we design a novel algorithm to partition the tree and distribute the data along a hierarchical space-lling curve with error-guided bucketization. Further optimization is achieved by storing recon-structed data at pre-selected tree nodes for each processor based on the available storage resources to reduce the overall wavelet reconstruction cost. At run time, the wavelet tree is rst traversed according to the user-specied error tolerance. Data blocks of dierent resolutions that satisfy the error tolerance are then decompressed and rendered to compose the nal image in parallel. Experimental results showed that our algorithm can reduce the run-time communication cost to a minimum and ensure a well-balanced workload among processors when visualizing gigabytes of data with arbitrary error tolerances. Key words: large scale data, multiresolution, wavelet transform, space-lling curve, load balancing, volume rendering, parallel renderin