BEGIN:VCALENDAR
VERSION:2.0
PRODID:Linklings LLC
BEGIN:VTIMEZONE
TZID:America/Denver
X-LIC-LOCATION:America/Denver
BEGIN:DAYLIGHT
TZOFFSETFROM:-0700
TZOFFSETTO:-0600
TZNAME:MDT
DTSTART:19700308T020000
RRULE:FREQ=YEARLY;BYMONTH=3;BYDAY=2SU
END:DAYLIGHT
BEGIN:STANDARD
TZOFFSETFROM:-0600
TZOFFSETTO:-0700
TZNAME:MST
DTSTART:19701101T020000
RRULE:FREQ=YEARLY;BYMONTH=11;BYDAY=1SU
END:STANDARD
END:VTIMEZONE
BEGIN:VEVENT
DTSTAMP:20260422T000712Z
LOCATION:405-406-407
DTSTART;TZID=America/Denver:20231115T113000
DTEND;TZID=America/Denver:20231115T120000
UID:submissions.supercomputing.org_SC23_sess164_pap203@linklings.com
SUMMARY:ADT-FSE: A New Encoder for SZ
DESCRIPTION:Tao Lu (DapuStor Corporation); Yu Zhong, Zibin Sun, Xiang Chen
 , You Zhou, and Fei Wu (Huazhong University of Science & Technology); and 
 Ying Yang, Yunxin Huang, and Yafei Yang (DapuStor Corporation)\n\nSZ is a 
 lossy floating-point data compressor that excels in compression ratio and 
 throughput for high-performance computing (HPC), time series databases, an
 d deep learning applications. However, SZ performs poorly for small chunks
  and has slow decompression. We pinpoint the Huffman tree in the quantizat
 ion factor encoder as the bottleneck of SZ. In this paper, we propose ADT-
 FSE, a new quantization factor encoder for SZ. Based on the Gaussian distr
 ibution of quantization factors, we design an adaptive data transcoding (A
 DT) scheme to map quantization factors to codes for better compressibility
 , and then use finite state entropy (FSE) to compress the codes. Experimen
 ts show that ADT-FSE improves the quantization factor compression ratio, c
 ompression and decompression throughput by up to 5x, 2x and 8x, respective
 ly, over the original SZ Huffman encoder. On average, SZ_ADT is over 2x fa
 ster than ZFP in decompression.\n\nTag: Accelerators, Data Analysis, Visua
 lization, and Storage, Data Compression\n\nRegistration Category: Tech Pro
 gram Reg Pass\n\nReproducibility Badges: Artifact Available, Artifact Func
 tional, Results Reproduced\n\nSession Chair: Kazutomo Yoshii (Argonne Nati
 onal Laboratory (ANL))\n\n
END:VEVENT
END:VCALENDAR