# UpDLRM: Accelerating Personalized Recommendation using Real-World PIM Architecture

arXiv: https://arxiv.org/abs/2406.13941 · HTML: https://mlsanity.com/paper/2406.13941/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 6 of its 20 references that are in the map.
- First posted on arXiv 20 Jun 2024.

## Builds on (most cited first, 6 of 6)

- [Deep Learning Recommendation Model for Personalization and Recommendation Systems](https://mlsanity.com/paper/1906.00091/) (2019)
- [The Architectural Implications of Facebook's DNN-based Personalized Recommendation](https://mlsanity.com/paper/1906.03109/) (2019)
- [Fine-Grained Spoiler Detection from Large-Scale Review Corpora](https://mlsanity.com/paper/1905.13416/) (2019)
- [RecNMP: Accelerating Personalized Recommendation with Near-Memory Processing](https://mlsanity.com/paper/1912.12953/) (2019)
- [Deep Learning Training in Facebook Data Centers: Design of Scale-up and Scale-out Systems](https://mlsanity.com/paper/2003.09518/) (2020)
- [Accelerating Recommendation System Training by Leveraging Popular Choices](https://mlsanity.com/paper/2103.00686/) (2021)

## Built on by (most cited first, 4 of 4)

- [Toward Robust and Efficient ML-Based GPU Caching for Modern Inference](https://mlsanity.com/paper/2509.20979/) (2025)
- [UpANNS: Enhancing Billion-Scale ANNS Efficiency with Real-World PIM Architecture](https://mlsanity.com/paper/2410.23805/) (2024)
- [Faster Distributed Inference-Only Recommender Systems via Bounded Lag Synchronous Collectives](https://mlsanity.com/paper/2512.19342/) (2025)
- [TokenStack: A Heterogeneous HBM-PIM Architecture and Runtime for Efficient LLM Inference](https://mlsanity.com/paper/2605.05639/) (2026)

## Abstract (arXiv)

Deep Learning Recommendation Models (DLRMs) have gained popularity in recommendation systems due to their effectiveness in handling large-scale recommendation tasks. The embedding layers of DLRMs have become the performance bottleneck due to their intensive needs on memory capacity and memory bandwidth. In this paper, we propose UpDLRM, which utilizes real-world processingin-memory (PIM) hardware, UPMEM DPU, to boost the memory bandwidth and reduce recommendation latency. The parallel nature of the DPU memory can provide high aggregated bandwidth for the large number of irregular memory accesses in embedding lookups, thus offering great potential to reduce the inference latency. To fully utilize the DPU memory bandwidth, we further studied the embedding table partitioning problem to achieve good workload-balance and efficient data caching. Evaluations using real-world datasets show that, UpDLRM achieves much lower inference time for DLRM compared to both CPU-only and CPU-GPU hybrid counterparts.