# TENET: A Framework for Modeling Tensor Dataflow Based on Relation-centric Notation

arXiv: https://arxiv.org/abs/2105.01892 · HTML: https://mlsanity.com/paper/2105.01892/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 5 May 2021.

## Built on by (most cited first, 9 of 9)

- [Full Stack Optimization of Transformer Inference: a Survey](https://mlsanity.com/paper/2302.14017/) (2023)
- [WaferLLM: Large Language Model Inference at Wafer Scale](https://mlsanity.com/paper/2502.04563/) (2025)
- [SCAR: Scheduling Multi-Model AI Workloads on Heterogeneous Multi-Chiplet Module Accelerators](https://mlsanity.com/paper/2405.00790/) (2024)
- [Scaling Deep Learning Computation over the Inter-Core Connected Intelligence Processor with T10](https://mlsanity.com/paper/2408.04808/) (2024)
- [SPADE: Sparse Pillar-based 3D Object Detection Accelerator for Autonomous Driving](https://mlsanity.com/paper/2305.07522/) (2023)
- [DOSA: Differentiable Model-Based One-Loop Search for DNN Accelerators](https://mlsanity.com/paper/2509.10702/) (2025)
- [LEGO: Spatial Accelerator Generation and Optimization for Tensor Applications](https://mlsanity.com/paper/2509.12053/) (2025)
- [KAPLA: Pragmatic Representation and Fast Solving of Scalable NN Accelerator Dataflow](https://mlsanity.com/paper/2306.15676/) (2023)
- [Hardware-Software Co-design for 3D-DRAM-based LLM Serving Accelerator](https://mlsanity.com/paper/2603.04797/) (2026)

## Abstract (arXiv)

Accelerating tensor applications on spatial architectures provides high performance and energy-efficiency, but requires accurate performance models for evaluating various dataflow alternatives. Such modeling relies on the notation of tensor dataflow and the formulation of performance metrics. Recent proposed compute-centric and data-centric notations describe the dataflow using imperative directives. However, these two notations are less expressive and thus lead to limited optimization opportunities and inaccurate performance models. In this paper, we propose a framework TENET that models hardware dataflow of tensor applications. We start by introducing a relation-centric notation, which formally describes the hardware dataflow for tensor computation. The relation-centric notation specifies the hardware dataflow, PE interconnection, and data assignment in a uniform manner using relations. The relation-centric notation is more expressive than the compute-centric and data-centric notations by using more sophisticated affine transformations. Another advantage of relation-centric notation is that it inherently supports accurate metrics estimation, including data reuse, bandwidth, latency, and energy. TENET computes each performance metric by counting the relations using integer set structures and operators. Overall, TENET achieves 37.4% and 51.4% latency reduction for CONV and GEMM kernels compared with the state-of-the-art data-centric notation by identifying more sophisticated hardware dataflows.