TSC-IBR: A Variant of Interval-Based Memory Reclamation Using CPU’s Time-Stamp Counter

Hardware timestamps are increasingly used in concurrent data structures. On the x86_64 platform, a clock with cycle-level resolution maintained by the CPU’s time-stamp counter register is frequently employed to implement parallel algorithms, and its value can be accessed swiftly without a system call. In line with this emerging trend, our research focuses on leveraging the CPU’s time-stamp counter in a memory management algorithm to secure safe memory reclamation in concurrent data structures. Existing quiescent-state-based memory reclamation algorithms rely on a global epoch counter maintained via a software timestamp. Although most quiescent-state-based methods demonstrate good performance when employed in lock-free concurrent data structures, they grapple with issues concerning robustness. Specifically, in Interval-Based Reclamation, a thread reclaiming a shared memory block must check whether the lifetime of the block intersects with the reserved epoch intervals of all threads. Interval-Based Reclamation can provide more fine-grained memory management than Epoch-Based Reclamation and hazard eras. In this study, we propose TSC-IBR, a variant of Interval-Based Reclamation. TSC-IBR leverages the CPU’s timestamp counter instead of the software timestamp. Our experiments confirmed the practicability of employing the CPU’s time-stamp counter in a memory reclamation algorithm. Compared to other quiescent-state-based methods using the software timestamp, TSC-IBR exhibits excellent robustness across all our lock-free concurrent data structure benchmarks and attains high throughput in most cases.

Authors

Institutions

Publication Details

Journal
IEEE Access
Published
2025-01-01
DOI
https://doi.org/10.1109/access.2025.3569072
Primary Topic
Parallel Computing and Optimization Techniques
Type
article
Field-Weighted Citation Impact
0.00
Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
article

TSC-IBR: A Variant of Interval-Based Memory Reclamation Using CPU’s Time-Stamp Counter

C. J. Zhang, Xinghui Zhu, Wendong Li
1 links
IEEE Access
Parallel Computing and Optimization Techniques
article

TSC-IBR: A Variant of Interval-Based Memory Reclamation Using CPU’s Time-Stamp Counter

C. J. Zhang, Xinghui Zhu, Wendong Li
article en

Abstract

Hardware timestamps are increasingly used in concurrent data structures. On the x86_64 platform, a clock with cycle-level resolution maintained by the CPU’s time-stamp counter register is frequently employed to implement parallel algorithms, and its value can be accessed swiftly without a system call. In line with this emerging trend, our research focuses on leveraging the CPU’s time-stamp counter in a memory management algorithm to secure safe memory reclamation in concurrent data structures. Existing quiescent-state-based memory reclamation algorithms rely on a global epoch counter maintained via a software timestamp. Although most quiescent-state-based methods demonstrate good performance when employed in lock-free concurrent data structures, they grapple with issues concerning robustness. Specifically, in Interval-Based Reclamation, a thread reclaiming a shared memory block must check whether the lifetime of the block intersects with the reserved epoch intervals of all threads. Interval-Based Reclamation can provide more fine-grained memory management than Epoch-Based Reclamation and hazard eras. In this study, we propose TSC-IBR, a variant of Interval-Based Reclamation. TSC-IBR leverages the CPU’s timestamp counter instead of the software timestamp. Our experiments confirmed the practicability of employing the CPU’s time-stamp counter in a memory reclamation algorithm. Compared to other quiescent-state-based methods using the software timestamp, TSC-IBR exhibits excellent robustness across all our lock-free concurrent data structure benchmarks and attains high throughput in most cases.

IEEE AccessVol. 13
Hunan Agricultural University (CN)
Climate action
Openalex Percentile: Top 92%
Parallel Computing and Optimization Techniques
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.