PhD Seminar • Programming Languages | Algorithms and Complexity • Reclaiming Memory in Concurrent Data Structures: Reactive Synchronization Approach

Wednesday, September 18, 2024 12:00 pm - 1:00 pm EDT (GMT -04:00)

Please note: This PhD seminar will take place in DC 1304.

Ajay Singh, PhD candidate
David R. Cheriton School of Computer Science

Supervisors: Professors Trevor Brown, Peter Buhr

What will you learn?

  • Why is reclaiming memory in concurrent data structures — the backbone of today’s scalable multicore software — a challenge?
  • What are the cutting-edge solutions, aka safe memory reclamation (SMR) algorithms, addressing this issue?
  • Existing SMR algorithms impose an asymmetric synchronization overhead on concurrent data structures. This overhead arises because, although reclamation events are infrequent, all threads frequently perform reclamation-related tasks during their common-case execution. This leads to non-trivial overhead, hindering performance in practice.
  • A new SMR algorithm that resolves the issue of asymmetric synchronization overhead.