mlsanityData to 7 Oct 2026

D-NOVA: In-Storage Retrieval Accelerator via Dual-Bound 3D NAND-Optimized Similarity Search with Vector Adaptation

arXiv 2607.17538 ↗First posted 20 Jul 2026cs.AR cs.CL cs.DBcited by 0

On the map: Retrieval-augmented generation → RAG pipeline design and evaluation

Abstract (arXiv)

Retrieval-Augmented Generation (RAG) enhances the factual grounding of large language model (LLM) inference by retrieving relevant information from external knowledge bases. However, its dense vector retrieval introduces significant latency and energy overhead, becoming the primary performance bottleneck. Although recent in-storage accelerators aim to reduce data movement, they still rely on host or embedded processors outside the memory, where nearly 70% of the total retrieval time is spent. As a result, they cannot fully overcome the bandwidth limitations, leading to yet another memory bottleneck. To tackle these limitations, we present D-NOVA, a hardware-software co-designed in-storage retrieval accelerator. D-NOVA executes an inverted file (IVF)-based hierarchical retrieval pipeline by deeply embedding the search functionality directly into the NAND memory array. This is achieved by incorporating a new distance metric, Dual-Bound Tight Similarity Sensing (DTS), which is specifically tailored for searching within the NAND string. In addition, we introduce a lightweight contrastive adapter that maps embedding vectors into a DTS-friendly domain, recovering near-software recall while improving performance and energy efficiency. D-NOVA is up to 41.7x faster and 71x more energy-efficient than a CPU baseline, and achieves 12.13x higher throughput while being up to 1.26x more energy-efficient than state-of-the-art in-storage RAG accelerators, demonstrating the potential of fully in-storage vector search for scalable RAG acceleration.

Key facts. No citing papers in the map yet (data through 7 Oct 2026). Placed in RAG pipeline design and evaluation, one of 15 directions in Retrieval-augmented generation. Builds on 22 of its 85 references that are in the map. First posted on arXiv 20 Jul 2026.

Builds on

22 of 85 references.

Closest ancestors · 1

Same direction, newest first

  1. 2025RAGO: Systematic Performance Optimization for Retrieval-Augmented Generation Serving18 Mar

Other references · 21

  1. 2018Representation Learning with Contrastive Predictive Codingcited by 7,662
  2. 2023Efficient Memory Management for Large Language Model Serving with PagedAttentioncited by 5,767
  3. 2018HotpotQA: A Dataset for Diverse, Explainable Multi-hop Question Answeringcited by 3,686
  4. 2017Billion-scale similarity search with GPUscited by 2,477
  5. 2020MiniLM: Deep Self-Attention Distillation for Task-Agnostic Compression of Pre-Trained Transformerscited by 1,414
  6. 2018FEVER: a large-scale dataset for Fact Extraction and VERificationcited by 1,362
  7. 2021BEIR: A Heterogenous Benchmark for Zero-shot Evaluation of Information Retrieval Modelscited by 1,225
  8. 2020MPNet: Masked and Permuted Pre-training for Language Understandingcited by 973
  9. 2023InfoNCE Loss Provably Learns Cluster-Preserving Representationscited by 37
  10. 2021SPANN: Highly-efficient Billion-scale Approximate Nearest Neighbor Searchcited by 35
  11. 2023CAGRA: Highly Parallel Graph Construction and Approximate Nearest Neighbor Search for GPUscited by 22
  12. 2023Pathfinding Future PIM Architectures by Demystifying a Commercial PIM Technologycited by 13
9 more
  1. 2022Sparse Attention Acceleration with Synergistic In-Memory Pruning and On-Chip Recomputationcited by 12
  2. 2024AIR-Bench: Automated Heterogeneous Information Retrieval Benchmarkcited by 12
  3. 2017Error Characterization, Mitigation, and Recovery in Flash Memory Based Solid-State Drivescited by 9
  4. 2024FSL-HDnn: A 5.7 TOPS/W End-to-end Few-shot Learning Classifier Accelerator with Feature Extraction and Hyperdimensional Computingcited by 4
  5. 2025REIS: A High-Performance and Energy-Efficient Retrieval System with In-Storage Processingcited by 4
  6. 2022Accelerating Large-Scale Graph-based Nearest Neighbor Search on a Computational Storage Platformcited by 3
  7. 2023Proxima: Near-storage Acceleration for Graph-based Approximate Nearest Neighbor Search in 3D NANDcited by 3
  8. 2025Flexible In-NAND Cryptographic Processing for Secure Flash Storagecited by 1
  9. 2025FeNOMS: Enhancing Open Modification Spectral Library Search with In-Storage Processing on Ferroelectric NAND (FeNAND) Flashcited by 1

Built on by

Not cited yet.

Similar, not cited

Close in wording, no citation either way

  1. 2024Accelerating Retrieval-Augmented Generation
  2. 2026HAVEN: High-Bandwidth Flash Augmented Vector Engine for Large-Scale Approximate Nearest-Neighbor Search Acceleration
  3. 2025Cosmos: A CXL-Based Full In-Memory System for Approximate Nearest Neighbor Search
  4. 2025A2ATS: Retrieval-Based KV Cache Reduction via Windowed Rotary Position Embedding and Query-Aware Vector Quantization
  5. 2023Chameleon: a Heterogeneous and Disaggregated Accelerator System for Retrieval-Augmented Language Models
  6. 2025CTkvr: KV Cache Retrieval for Long-Context LLMs via Centroid then Token Indexing
  7. 2024Retro-li: Small-Scale Retrieval Augmented Generation Supporting Noisy Similarity Searches and Domain Shift Generalization
  8. 2025FreeKV: Boosting KV Cache Retrieval for Efficient LLM Inference