HPCA 2026
Sat 31 January - Wed 4 February 2026 Sydney, Australia
co-located with HPCA/CGO/PPoPP/CC 2026
Tue 3 Feb 2026 11:50 - 12:10 at Collaroy - Caching and Prefetching Chair(s): David Schall

In this paper, we present the Streamline temporal prefetcher, which introduces a stream-based metadata representation that produces three significant benefits over Triangel, the previous state-of-the-art in temporal prefetching. First, it removes redundancy present in Triangel’s metadata. Second, it prioritizes the storage of those metadata entries that have higher prefetch utility. Third, it eliminates the need for the untenable LLC traffic that is induced when Triangel dynamically adjusts the size of its metadata store. The end result is that for memory-intensive SPEC 2006, SPEC 2017, and GAP benchmarks, Streamline outperforms Triangel by 6.7 percentage points on an 8-core system with a stride prefetcher. This performance benefit comes from Streamline’s improved storage efficiency: It holds 33% more correlations than Triangel, which translates to 12.5 percentage points better prefetch coverage. Significantly, Streamline matches the performance of Triangel even when Triangel is given twice the metadata storage.

Tue 3 Feb

Displayed time zone: Hobart change

11:30 - 12:50
Caching and PrefetchingMain Conference at Collaroy
Chair(s): David Schall Technical University of Munich
11:30
20m
Talk
Athena: Synergizing Data Prefetching and Off-Chip Prediction via Online Reinforcement Learning
Main Conference
Zhenrong Lang ETH Zürich, Rahul Bera ETH Zurich, Caroline Hengartner ETH Zürich, Konstantinos Kanellopoulos ETH Zurich, Rakesh Kumar NTNU, Mohammad Sadrosadati ETH Zürich, Onur Mutlu ETH Zurich
11:50
20m
Talk
Streamlined On-Chip Temporal Prefetching
Main Conference
Quang Duong The University of Texas at Austin, Calvin Lin The University of Texas at Austin
12:10
20m
Talk
Intermittence-Aware Cache Compression
Main Conference
Gan Fang Purdue University, Jianping Zeng Arizona State University, Yuchen Zhou Purdue University, Changhee Jung Purdue University, USA
12:30
20m
Talk
TENET-v2: Applying Relation-centric Notation to Model and Optimize Data Swizzle in the Cache of Modern NPU
Main Conference
Hanyu Zhang Zhejiang University, Fangxu Guo Zhejiang University, Liqiang Lu Zhejiang University, Long Wang Huawei Technologies, Yunfei Du Huawei Technologies, Zhe Wang Huawei Technologies, Jinghan Zhang Huawei Technologies, Jie Zhang Peking University, Chenli Xue Zhejiang University, Chengpeng Wu Zhejiang University, Ziyi Zhang Zhejiang University, Yun Liang Peking University, Size Zheng Tsinghua University, Jianwei Yin Zhejiang University