Haneul Park

Haneul Park

Auto-generated profile, based on publicly available information

Haneul Park

University of Illinois Urbana-Champaign, Urbana, IL, USA

Research Focus & Metrics

Academic Career Length

4Years (2023-Present)

(Calculated once Scilit publication data is synced: current year − year of first publication + 1)

Representative Publications

View All Publications
17 June 2023
Verified
Conference Paper
Open access
WJWenjing Jin
Wenjing Jin
WJWonsuk Jang
Wonsuk Jang
HPHaneul Park
Haneul Park
JLJongsung Lee
Jongsung Lee
SKSoosung Kim
Soosung Kim
JLJae W. Lee
Jae W. Lee

20 June 2025
Verified
Conference Paper
Open access
HPHaneul Park
Haneul Park
JLJiaqi Lou
Jiaqi Lou
SLSangjin Benjamin Lee
Sangjin Benjamin Lee
YYYifan Yuan
Yifan Yuan
KPKyoungSoo Park
KyoungSoo Park
NKNam Sung Kim
Nam Sung Kim

23 March 2026 in IEEE Computer Architecture Letters
Verified
Journal Article
Open access
HPHaneul Park
Haneul Park
GAGrant Ayers
Grant Ayers
NKNam Sung Kim
Nam Sung Kim
PLPhilip Levis
Philip Levis
BMBrian Morris
Brian Morris
DRAM dominates the cost of hyperscale datacenter servers today and AI-driven memory supply shortages are pushing costs even higher. CXL memory expanders are a new technology to reduce server costs by reusing legacy DDR DRAM modules from older servers. Memory expanders further save cost through inline compression that increases the effective capacity. This paper investigates two critical design points of a memory expander: the compression algorithm and the compression block size. Both design points introduce tradeoffs between latency and capacity that involve complex interactions with the memory expander architecture, access patterns to the device, and the entropy of stored data. Current commercial devices use simple, fast compression and a 4kB block as the default. Using a compressibility benchmark and the latest traces of cold memory accesses from a hyperscaler, we find that using a different compression algorithm and block size can simultaneously increase expander capacity and reduce access latency. The key insight is counter-intuitive: using a slower, more complex compression algorithm improves latency because the reduction in DRAM traffic more than offsets the additional decompression time.