Logical Synchrony Networks: A formal model for deterministic distribution
Fuente:
arXiv
Enregistré dans:
| Auteurs principaux: | , , , , , , |
|---|---|
| Format: | Preprint |
| Publié: |
2024
|
| Sujets: | |
| Accès en ligne: | |
| Tags: |
Ajouter un tag
Pas de tags, Soyez le premier à ajouter un tag!
|
| _version_ | 1866909216952811520 |
|---|---|
| author | Kenwright, Logan Roop, Partha Allen, Nathan Lall, Sanjay Cascaval, Calin Spalink, Tammo Izzard, Martin |
| author_facet | Kenwright, Logan Roop, Partha Allen, Nathan Lall, Sanjay Cascaval, Calin Spalink, Tammo Izzard, Martin |
| contents | Kahn Process Networks (KPNs) are a deterministic Model of Computation (MoC) for distributed systems. KPNs supports non-blocking writes and blocking reads, with the consequent assumption of unbounded buffers between processes. Variants such as Finite FIFO Platforms (FFP) have been developed, which enforce boundedness. One issue with existing models is that they mix process synchronisation with process execution. In this paper we address how these two facets may be decoupled.
This paper explores a recent alternative called bittide, which decouples the execution of a process from the control needed for process synchronisation, and thus preserves determinism and boundedness while ensuring pipelined execution for better throughput. Our intuition is that such an approach could leverage not only determinism and buffer boundedness but may potentially offer better overall throughput.
To understand the behavior of these systems we define a formal model -- a deterministic MoC called Logical Synchrony Networks (LSNs). LSNs describes a network of processes modelled as a graph, with edges representing invariant logical delays between a producer process and the corresponding consumer process. We show that this abstraction is satisfied by KPNs. Subsequently, we show that both FFPs and bittide faithfully implement this abstraction. Thus, we show for the first time that FFPs and bittide offer two alternative ways of implementing deterministic distributed systems with the latter being more performant. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_07433 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Logical Synchrony Networks: A formal model for deterministic distribution Kenwright, Logan Roop, Partha Allen, Nathan Lall, Sanjay Cascaval, Calin Spalink, Tammo Izzard, Martin Distributed, Parallel, and Cluster Computing Formal Languages and Automata Theory Kahn Process Networks (KPNs) are a deterministic Model of Computation (MoC) for distributed systems. KPNs supports non-blocking writes and blocking reads, with the consequent assumption of unbounded buffers between processes. Variants such as Finite FIFO Platforms (FFP) have been developed, which enforce boundedness. One issue with existing models is that they mix process synchronisation with process execution. In this paper we address how these two facets may be decoupled. This paper explores a recent alternative called bittide, which decouples the execution of a process from the control needed for process synchronisation, and thus preserves determinism and boundedness while ensuring pipelined execution for better throughput. Our intuition is that such an approach could leverage not only determinism and buffer boundedness but may potentially offer better overall throughput. To understand the behavior of these systems we define a formal model -- a deterministic MoC called Logical Synchrony Networks (LSNs). LSNs describes a network of processes modelled as a graph, with edges representing invariant logical delays between a producer process and the corresponding consumer process. We show that this abstraction is satisfied by KPNs. Subsequently, we show that both FFPs and bittide faithfully implement this abstraction. Thus, we show for the first time that FFPs and bittide offer two alternative ways of implementing deterministic distributed systems with the latter being more performant. |
| title | Logical Synchrony Networks: A formal model for deterministic distribution |
| topic | Distributed, Parallel, and Cluster Computing Formal Languages and Automata Theory |
| url | https://arxiv.org/abs/2402.07433 |