# AI and Memory Wall

arXiv: https://arxiv.org/abs/2403.14123 · HTML: https://mlsanity.com/paper/2403.14123/

## Key facts

- Cited by 92 papers in the map (data through 14 Oct 2026).
- Builds on 26 of its 46 references that are in the map.
- First posted on arXiv 21 Mar 2024.

## Builds on (most cited first, 26 of 26)

- [Attention Is All You Need](https://mlsanity.com/paper/1706.03762/) (2017)
- [BERT: Pre-training of Deep Bidirectional Transformers for Language Understanding](https://mlsanity.com/paper/1810.04805/) (2018)
- [Language Models are Few-Shot Learners](https://mlsanity.com/paper/2005.14165/) (2020)
- [Llama 2: Open Foundation and Fine-Tuned Chat Models](https://mlsanity.com/paper/2307.09288/) (2023)
- [PaLM: Scaling Language Modeling with Pathways](https://mlsanity.com/paper/2204.02311/) (2022)
- [Mistral 7B](https://mlsanity.com/paper/2310.06825/) (2023)
- [Scaling Instruction-Finetuned Language Models](https://mlsanity.com/paper/2210.11416/) (2022)
- [Training Compute-Optimal Large Language Models](https://mlsanity.com/paper/2203.15556/) (2022)
- [Emergent Abilities of Large Language Models](https://mlsanity.com/paper/2206.07682/) (2022)
- [ZeRO: Memory Optimizations Toward Training Trillion Parameter Models](https://mlsanity.com/paper/1910.02054/) (2019)
- [Mixed Precision Training](https://mlsanity.com/paper/1710.03740/) (2017)
- [Optimization Methods for Large-Scale Machine Learning](https://mlsanity.com/paper/1606.04838/) (2016)
- [SparseGPT: Massive Language Models Can Be Accurately Pruned in One-Shot](https://mlsanity.com/paper/2301.00774/) (2023)
- [A Survey of Quantization Methods for Efficient Neural Network Inference](https://mlsanity.com/paper/2103.13630/) (2021)
- [It's Not Just Size That Matters: Small Language Models Are Also Few-Shot Learners](https://mlsanity.com/paper/2009.07118/) (2020)
- [Sparsity in Deep Learning: Pruning and growth for efficient inference and training in neural networks](https://mlsanity.com/paper/2102.00554/) (2021)
- [ZeroQuant: Efficient and Affordable Post-Training Quantization for Large-Scale Transformers](https://mlsanity.com/paper/2206.01861/) (2022)
- [SpQR: A Sparse-Quantized Representation for Near-Lossless LLM Weight Compression](https://mlsanity.com/paper/2306.03078/) (2023)
- [Deep Learning Recommendation Model for Personalization and Recommendation Systems](https://mlsanity.com/paper/1906.00091/) (2019)
- [Reducing Activation Recomputation in Large Transformer Models](https://mlsanity.com/paper/2205.05198/) (2022)
- [ADAHESSIAN: An Adaptive Second Order Optimizer for Machine Learning](https://mlsanity.com/paper/2006.00719/) (2020)
- [FP8 Formats for Deep Learning](https://mlsanity.com/paper/2209.05433/) (2022)
- [A Fast Post-Training Pruning Framework for Transformers](https://mlsanity.com/paper/2204.09656/) (2022)
- [Checkmate: Breaking the Memory Wall with Optimal Tensor Rematerialization](https://mlsanity.com/paper/1910.02653/) (2019)
- [Full Stack Optimization of Transformer Inference: a Survey](https://mlsanity.com/paper/2302.14017/) (2023)
- [Accelerating Recommender Systems via Hardware "scale-in"](https://mlsanity.com/paper/2009.05230/) (2020)

## Built on by (most cited first, 30 of 92)

- [Block Transformer: Global-to-Local Language Modeling for Fast Inference](https://mlsanity.com/paper/2406.02657/) (2024)
- [GuidedQuant: Large Language Model Quantization via Exploiting End Loss Guidance](https://mlsanity.com/paper/2505.07004/) (2025)
- [ShiftAddLLM: Accelerating Pretrained LLMs via Post-Training Multiplication-Less Reparameterization](https://mlsanity.com/paper/2406.05981/) (2024)
- [A Survey: Collaborative Hardware and Software Design in the Era of Large Language Models](https://mlsanity.com/paper/2410.07265/) (2024)
- [SambaNova SN40L: Scaling the AI Memory Wall with Dataflow and Composition of Experts](https://mlsanity.com/paper/2405.07518/) (2024)
- [BitMoD: Bit-serial Mixture-of-Datatype LLM Acceleration](https://mlsanity.com/paper/2411.11745/) (2024)
- [Harmony in Divergence: Towards Fast, Accurate, and Memory-efficient Zeroth-order LLM Fine-tuning](https://mlsanity.com/paper/2502.03304/) (2025)
- [ETS: Efficient Tree Search for Inference-Time Scaling](https://mlsanity.com/paper/2502.13575/) (2025)
- [Flex-PE: Flexible and SIMD Multi-Precision Processing Element for AI Workloads](https://mlsanity.com/paper/2412.11702/) (2024)
- [Efficient Heterogeneous Large Language Model Decoding with Model-Attention Disaggregation](https://mlsanity.com/paper/2405.01814/) (2024)
- [Recent and Upcoming Developments in Randomized Numerical Linear Algebra for Machine Learning](https://mlsanity.com/paper/2406.11151/) (2024)
- [Efficient LLM Inference using Dynamic Input Pruning and Cache-Aware Masking](https://mlsanity.com/paper/2412.01380/) (2024)
- [BlockDialect: Block-wise Fine-grained Mixed Format Quantization for Energy-Efficient LLM Inference](https://mlsanity.com/paper/2501.01144/) (2025)
- [HiFloat4 Format for Language Model Inference](https://mlsanity.com/paper/2602.11287/) (2026)
- [STAR-KV: Low-Rank KV Cache Compression via Soft Thresholding for Adaptive Rank Control](https://mlsanity.com/paper/2606.08382/) (2026)
- [TinyLLM: A Framework for Training and Deploying Language Models at the Edge Computers](https://mlsanity.com/paper/2412.15304/) (2024)
- [Fire-Flyer AI-HPC: A Cost-Effective Software-Hardware Co-Design for Deep Learning](https://mlsanity.com/paper/2408.14158/) (2024)
- [FlashAttention on a Napkin: A Diagrammatic Approach to Deep Learning IO-Awareness](https://mlsanity.com/paper/2412.03317/) (2024)
- [Leveraging the true depth of LLMs](https://mlsanity.com/paper/2502.02790/) (2025)
- [AlphaQ: Calibration-Free Bit Allocation for Mixture-of-Experts Quantization](https://mlsanity.com/paper/2606.04980/) (2026)
- [A Diagonal Structured State Space Model on Loihi 2 for Efficient Streaming Sequence Processing](https://mlsanity.com/paper/2409.15022/) (2024)
- [SKIM: Any-bit Quantization Pushing The Limits of Post-Training Quantization](https://mlsanity.com/paper/2412.04180/) (2024)
- [TerEffic: Highly Efficient Ternary LLM Inference on FPGA](https://mlsanity.com/paper/2502.16473/) (2025)
- [Fine-tuning Quantized Neural Networks with Zeroth-order Optimization](https://mlsanity.com/paper/2505.13430/) (2025)
- [Tawa: Automatic Warp Specialization for Modern GPUs with Asynchronous References](https://mlsanity.com/paper/2510.14719/) (2025)
- [CoreQ: Learning-Free Mismatch Correction and Successive Rounding for Quantization](https://mlsanity.com/paper/2602.05902/) (2026)
- [EfficientRollout: System-Aware Self-Speculative Decoding for RL Rollouts](https://mlsanity.com/paper/2606.18967/) (2026)
- [A Theory of Inference Compute Scaling: Reasoning through Directed Stochastic Skill Search](https://mlsanity.com/paper/2507.00004/) (2025)
- [Determinant Estimation under Memory Constraints and Neural Scaling Laws](https://mlsanity.com/paper/2503.04424/) (2025)
- [POLARON: Precision-aware On-device Learning and Adaptive Runtime-cONfigurable AI acceleration](https://mlsanity.com/paper/2506.08785/) (2025)

## Abstract (arXiv)

The availability of unprecedented unsupervised training data, along with neural scaling laws, has resulted in an unprecedented surge in model size and compute requirements for serving/training LLMs. However, the main performance bottleneck is increasingly shifting to memory bandwidth. Over the past 20 years, peak server hardware FLOPS has been scaling at 3.0x/2yrs, outpacing the growth of DRAM and interconnect bandwidth, which have only scaled at 1.6 and 1.4 times every 2 years, respectively. This disparity has made memory, rather than compute, the primary bottleneck in AI applications, particularly in serving. Here, we analyze encoder and decoder Transformer models and show how memory bandwidth can become the dominant bottleneck for decoder models. We argue for a redesign in model architecture, training, and deployment strategies to overcome this memory limitation.