Blocking Analysis for Spin Locks in Real-Time Parallel Tasks

Dinh, Son; Li, Jing; Agrawal, Kunal; Gill, Chris; Lu, Chenyang

doi:10.1109/tpds.2017.2777454

Cited by 25 publications

(28 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Recently [21], spin locks were studied for parallel real-time tasks in where each parallel task is scheduled exclusively on several pre-assigned processors (i.e., by the federated scheduling approach). This reduces the need for locks or scheduling to synchronize concurrent access (through shared resources may introduce locking as considered by Dinh et al [5]). The authors developed new schedulability analysis techniques for parallel tasks with spin locks, and analysed blocking times but the analysis is pessimistic.…”

Section: Introductionmentioning

confidence: 99%

Analysis of Polka Contention Manager for use in Multicore Hard Real-Time Systems

Quillet

Queudet

Lime

2020

Proceedings of the 28th International Conference on Real-Time Networks and Systems

View full text Add to dashboard Cite

Transactional memory (TM) draws the attention of both academic and development groups; indeed this concept offers an alternative to lock-based approaches, easing programmers' work. Despite the large amount of investigations around this topic, the question of the correctness of most TM implementations remains open. More specifically, the lack of upper bounds on the execution time of transactions prevents the use of TM in real-time systems. To address this issue, we introduce new realistic assumptions relative to real-time systems, which allow to ensure wait-freedom guarantees progress (i.e. all transactions progress) when Polka contention manager is considered. In that context, through a thorough formalization of the system, we prove upper bounds both on the number of abortions and on the execution time of transactions. CCS CONCEPTS• Computer systems organization → Real-time operating systems; Multicore architectures.

show abstract

Section: Introductionmentioning

confidence: 99%

Analysis of Polka Contention Manager for use in Multicore Hard Real-Time Systems

Quillet

Queudet

Lime

2020

Proceedings of the 28th International Conference on Real-Time Networks and Systems

View full text Add to dashboard Cite

show abstract

Section: Introductionmentioning

confidence: 99%

“…Recently, spin locks were studied for parallel real-time tasks in [9] where each parallel task is scheduled exclusively on several pre-assigned processors (i.e., by the federated *Corresponding author: Nan Guan scheduling approach [6]). However, the analysis in [9] is pessimistic. The contribution of our work is to develop new techniques for the schedulability analysis of real-time parallel tasks with spin locks and significantly improve the analysis precision against the state-of-the-art.…”

Section: Introductionmentioning

confidence: 99%

“…Both [9] and our work only require knowledge of the total worst-case execution time (WCET) C i and longest path length L i of each task, but not the exact graph structure (the benefits of only using the abstract C i and L i information in the analysis will be discussed in Section 2.4). In [9]'s analysis, all blocking time caused by spin locks is considered to contribute to the workload that delays the finishing time of a parallel task, which is added to C i and L i in their worstcase scenarios separately. This is quite pessimistic since many blocking time can not delay the finishing time of a parallel task due to the parallelism and intra-dependencies.…”

Section: Introductionmentioning

confidence: 99%

“…We conduct experiments to evaluate the precision improvement using our new techniques compared with [9], with both randomly generated tasks and workload generated according to realistic OpenMP programs. Experimental results show that our techniques consistently outperform [9] under different settings.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

On the Analysis of Parallel Real-Time Tasks with Spin Locks

Jiang

Guan

et al. 2020

Preprint

View full text Add to dashboard Cite

Locking protocol is an essential component in resource management of real-time systems, which coordinates mutually exclusive accesses to shared resources from different tasks. Although the design and analysis of locking protocols have been intensively studied for sequential real-time tasks, there has been little work on this topic for parallel real-time tasks. In this paper, we study the analysis of parallel real-time tasks using spin locks to protect accesses to shared resources in three commonly used request serving orders (unordered, FIFO-order and priority-order). A remarkable feature making our analysis method more accurate is to systematically analyze the blocking time which may delay a task's finishing time, where the impact to the total workload and the longest path length is jointly considered, rather than analyzing them separately and counting all blocking time as the workload that delays a task's finishing time, as commonly assumed in the state-of-the-art.

show abstract