CMU-SAFARI / pim-mlLinks
PIM-ML is a benchmark for training machine learning algorithms on the UPMEM architecture, which is the first publicly-available real-world processing-in-memory (PIM) architecture. Described in the ISPASS 2023 paper by Gomez-Luna et al. (https://arxiv.org/pdf/2207.07886.pdf).
☆24Updated 6 months ago
Alternatives and similar repositories for pim-ml
Users that are interested in pim-ml are comparing it to the libraries listed below
Sorting:
- STONNE Simulator integrated into SST Simulator☆20Updated last year
- MultiPIM: A Detailed and Configurable Multi-Stack Processing-In-Memory Simulator☆55Updated 4 years ago
- A PIM instrumentation, compilation, execution, simulation, and evaluation repository for BLIMP-style architectures.☆18Updated 3 years ago
- The simulator for SPADA, an SpGEMM accelerator with adaptive dataflow☆38Updated 2 years ago
- Heterogenous ML accelerator☆18Updated 3 months ago
- A scheduler for spatial DNN accelerators that generate high-performance schedules in one shot using mixed integer programming (MIP)☆82Updated last year
- ☆28Updated 3 years ago
- ☆65Updated 4 years ago
- dMazeRunner: Dataflow acceleration optimization infrastructure for coarse-grained programmable accelerators☆46Updated 3 years ago
- ☆25Updated last year
- A Cycle-level simulator for M2NDP☆29Updated 3 months ago
- A simulation framework for modeling efficiency of Graph Neural Network Dataflows☆22Updated 5 months ago
- Domain-Specific Architecture Generator 2☆21Updated 2 years ago
- ☆25Updated last year
- A reference implementation of the Mind Mappings Framework.☆30Updated 3 years ago
- Dynamically Reconfigurable Architecture Template and Cycle-level Microarchitecture Simulator for Dataflow AcCelerators☆29Updated 2 years ago
- ☆92Updated last year
- GNNear: Accelerating Full-Batch Training of Graph NeuralNetworks with Near-Memory Processing☆13Updated 2 years ago
- agile hardware-software co-design☆50Updated 3 years ago
- ☆28Updated 2 years ago
- PIMeval simulator and PIMbench suite☆32Updated this week
- SimplePIM is the first high-level programming framework for real-world processing-in-memory (PIM) architectures. Described in the PACT 20…☆27Updated last year
- ☆11Updated 3 years ago
- ☆30Updated 9 months ago
- A Full-System Framework for Simulating NDP devices from Caches to DRAM☆17Updated last year
- PyGim is the first runtime framework to efficiently execute Graph Neural Networks (GNNs) on real Processing-in-Memory systems. It provide…☆27Updated 3 months ago
- Fibertree emulator☆12Updated 9 months ago
- High Bandwidth Memory (HBM) timing model based on DRAMSim2☆42Updated 8 years ago
- Artifact for paper "PIM is All You Need: A CXL-Enabled GPU-Free System for LLM Inference", ASPLOS 2025☆82Updated 3 months ago
- NeuraChip Accelerator Simulator☆13Updated last year