SeqCDC speeds up content-defined chunking by detecting monotonic byte sequences, skipping unfavorable regions, and vectorizing with SSE/AVX, achieving up to roughly 10x throughput over unaccelerated and 1.2x to 1.35x over vector-accelerated chunkers at 8 to 16 KB chunks.
A comprehensive study of the past, present, and future of data deduplication,
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
citation-role summary
background 1
citation-polarity summary
fields
cs.DC 1years
2025 1verdicts
CONDITIONAL 1roles
background 1polarities
background 1representative citing papers
citing papers explorer
-
Vectorized Sequence-Based Chunking for Data Deduplication
SeqCDC speeds up content-defined chunking by detecting monotonic byte sequences, skipping unfavorable regions, and vectorizing with SSE/AVX, achieving up to roughly 10x throughput over unaccelerated and 1.2x to 1.35x over vector-accelerated chunkers at 8 to 16 KB chunks.