AI and Memory Wall
Reviewed by Pithpith:P3FJJATJopen to challenge →
read the original abstract
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.
This paper has not been read by Pith yet.
Forward citations
Cited by 5 Pith papers
-
AlphaQ: Calibration-Free Bit Allocation for Mixture-of-Experts Quantization
AlphaQ performs calibration-free mixed-precision quantization of MoE models by allocating higher bits to experts whose weight spectra exhibit stronger heavy-tailed structure according to HT-SR theory, outperforming ca...
-
Structured Recurrent Mixers for Massively Parallelized Sequence Generation
Structured Recurrent Mixers enable algebraic switching between parallel training and recurrent inference representations, delivering higher efficiency, information capacity, and throughput than other linear-complexity models.
-
Structured Recurrent Mixers for Massively Parallelized Sequence Generation
Structured Recurrent Mixers enable algebraic switching between parallel training and recurrent inference representations, yielding higher throughput, concurrency, and training efficiency than comparable linear-complex...
-
Structured Recurrent Mixers for Massively Parallelized Sequence Generation
Structured Recurrent Mixers provide a dual parallel-recurrent representation for sequence models, claiming superior training efficiency, information capacity, and inference throughput over linear complexity alternatives.
-
Rethinking Network Topologies for Cost-Effective Mixture-of-Experts LLM Serving
Switchless topologies such as 3D full-mesh are 20.6-56.2% more cost-effective than scale-up networks for MoE LLM serving, with current link bandwidths over-provisioned by up to 27%.
discussion (0)
Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.