RM-SSD: In-Storage Computing for Large-Scale Recommendation Inference

To meet the strict service level agreement requirements of recommendation systems, the entire set of embeddings in recommendation systems needs to be loaded into the memory. However, as the model and dataset for production-scale recommendation systems scale up, the size of the embeddings is approaching the limit of memory capacity. Limited physical memory constrains the algorithms that can be trained and deployed, posing a severe challenge for deploying advanced recommendation systems. Recent studies offload the embedding lookups into SSDs, which targets the embedding-dominated recommendation models. This paper takes it one step further and proposes to offload the entire recommendation system into SSD with in-storage computing capability. The proposed SSD-side FPGA solution leverages a low -end FPGA to speed up both the embedding-dominated and MLP-dominated models with high resource efficiency. We evaluate the performance of the proposed solution with a prototype SSD. Results show that we can achieve 20-100x throughput improvement compared with the baseline SSD and 1.5-15x improvement compared with the state-of-art.

Automated summary

This paper proposes a solution to offload the entire recommendation system into an SSD with in-storage computing capability, and improves the resource efficiency of the models using a low-end FPGA. Experimental results show significant throughput improvement compared to the baseline solution.