# Efficient Orchestrated AI Workflows Execution on Scale-out Spatial Architecture

arXiv: https://arxiv.org/abs/2405.17221 · HTML: https://mlsanity.com/paper/2405.17221/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 5 of its 72 references that are in the map.
- First posted on arXiv 21 May 2024.

## Builds on (most cited first, 5 of 5)

- [Taurus: A Data Plane Architecture for Per-Packet ML](https://mlsanity.com/paper/2002.08987/) (2020)
- [Wafer-scale Computing: Advancements, Challenges, and Future Perspectives](https://mlsanity.com/paper/2310.09568/) (2023)
- [Homunculus: Auto-Generating Efficient Data-Plane ML Pipelines for Datacenter Networks](https://mlsanity.com/paper/2206.05592/) (2022)
- [ARENA: Asynchronous Reconfigurable Accelerator Ring to Enable Data-Centric Parallel Computing](https://mlsanity.com/paper/2011.04931/) (2020)
- [Towards Efficient Control Flow Handling in Spatial Architecture via Architecting the Control Flow Plane](https://mlsanity.com/paper/2307.02847/) (2023)

## Abstract (arXiv)

Given the increasing complexity of AI applications, traditional spatial architectures frequently fall short. Our analysis identifies a pattern of interconnected, multi-faceted tasks encompassing both AI and general computational processes. In response, we have conceptualized "Orchestrated AI Workflows," an approach that integrates various tasks with logic-driven decisions into dynamic, sophisticated workflows. Specifically, we find that the intrinsic Dual Dynamicity of Orchestrated AI Workflows, namely dynamic execution times and frequencies of Task Blocks, can be effectively represented using the Orchestrated Workflow Graph. Furthermore, the intrinsic Dual Dynamicity poses challenges to existing spatial architecture, namely Indiscriminate Resource Allocation, Reactive Load Rebalancing, and Contagious PEA Idleness. To overcome these challenges, we present Octopus, a scale-out spatial architecture and a suite of advanced scheduling strategies optimized for executing Orchestrated AI Workflows, such as the Discriminate Dual-Scheduling Mechanism, Adaptive TBU Scheduling Strategy, and Proactive Cluster Scheduling Strategy. Our evaluations demonstrate that Octopus significantly outperforms traditional architectures in handling the dynamic demands of Orchestrated AI Workflows, and possesses robust scalability in large scale hardware such as wafer-scale chip.