Abstract

Recent compressed suffix trees targeted to highly repetitive text collections reach excellent compression performance, but operation times in the order of milliseconds. We design a new suffix tree representation for this scenario that still achieves very low space usage, only slightly larger than the best previous one, but supports the operations within microseconds. This puts the data structure in the same performance level of compressed suffix trees designed for standard text collections, which on repetitive collections use many times more space than our new structure.