PhD Seminar • Networks and Distributed Systems — Work-Stealing, Locality-Aware Actor SchedulingExport this event to calendar

Friday, May 18, 2018 1:30 PM EDT

Saman Barghi, PhD candidate
David R. Cheriton School of Computer Science

The actor programming model is gaining popularity due to the prevalence of multi-core systems along with the rising need for highly scalable and distributed applications. Frameworks such as Akka, Orleans, Pony, and C++ Actor Framework (CAF) have been developed to address these application requirements. Each framework provides a runtime system to schedule and run millions of actors, potentially on multi-socket platforms with non-uniform memory access (NUMA). However, the literature provides only limited research that studies or improves the performance of actor-based applications on NUMA systems. 

We study the performance penalties that are imposed on actors running on a NUMA system and characterize applications based on the actor type, behavior, and communication pattern. This information is used to identify workloads that can benefit from improved locality on a NUMA system. In addition, two locality- and NUMA-aware work-stealing schedulers are proposed and their respective execution overhead in CAF is studied on both AMD and Intel machines. The performance of the proposed work-stealing schedulers is evaluated against the default scheduler in CAF.

Location 
DC - William G. Davis Computer Research Centre
1304
200 University Avenue West

Waterloo, ON N2L 3G1
Canada

S M T W T F S
29
30
31
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
1
2
  1. 2024 (100)
    1. April (23)
    2. March (27)
    3. February (25)
    4. January (25)
  2. 2023 (296)
    1. December (20)
    2. November (28)
    3. October (15)
    4. September (25)
    5. August (30)
    6. July (30)
    7. June (22)
    8. May (23)
    9. April (32)
    10. March (31)
    11. February (18)
    12. January (22)
  3. 2022 (245)
  4. 2021 (210)
  5. 2020 (217)
  6. 2019 (255)
  7. 2018 (217)
  8. 2017 (36)
  9. 2016 (21)
  10. 2015 (36)
  11. 2014 (33)
  12. 2013 (23)
  13. 2012 (4)
  14. 2011 (1)
  15. 2010 (1)
  16. 2009 (1)
  17. 2008 (1)