Cover Image

Cost-Efficient LLM Training with Lifetime-Aware Tensor Offloading via GPUDirect Storage

Fuente: arXiv

Saved in:

Bibliographic Details
Main Authors:	Yuan, Ziqi, Zhang, Haoyang, Zhou, Yirui Eric, Mohan, Apoorve, Chung, I-Hsin, Seelam, Seetharami, Huang, Jian
Format:	Preprint
Published:	2025
Subjects:	Distributed, Parallel, and Cluster Computing Artificial Intelligence Machine Learning Performance
Online Access:	Acceder al recurso
Tags:	Add Tag No Tags, Be the first to tag this record!

Similar Items

RAPID-LLM: Resilience-Aware Performance analysis of Infrastructure for Distributed LLM Training and Inference
by: Karfakis, George, et al.
Published: (2025)

A Pilot Study on Tunable Precision Emulation via Automatic BLAS Offloading
by: Liu, Hang, et al.
Published: (2025)

Fault-Tolerant Hybrid-Parallel Training at Scale with Reliable and Efficient In-memory Checkpointing
by: Wang, Yuxin, et al.
Published: (2023)

Efficient allocation of image recognition and LLM tasks on multi-GPU system
by: Lawenda, Marcin, et al.
Published: (2025)

Can Tensor Cores Benefit Memory-Bound Kernels? (No!)
by: Zhang, Lingqi, et al.
Published: (2025)

HybridGen: Efficient LLM Generative Inference via CPU-GPU Hybrid Computing
by: Lin, Mao, et al.
Published: (2026)

THEAS: Efficient Power Management in Multi-Core CPUs via Cache-Aware Resource Scheduling
by: Muhammad, Said, et al.
Published: (2025)

DUAL-BLADE: Dual-Path NVMe-Direct KV-Cache Offloading for Edge LLM Inference
by: Jeong, Bodon, et al.
Published: (2026)

Learning-Augmented Performance Model for Tensor Product Factorization in High-Order FEM
by: Ren, Xuanzhengbo, et al.
Published: (2026)

The Energy Cost of Execution-Idle in GPU Clusters
by: Lei, Yiran, et al.
Published: (2026)

Fast and Scalable Mixed Precision Euclidean Distance Calculations Using GPU Tensor Cores
by: Curless, Brian, et al.
Published: (2025)

ProTrain: Efficient LLM Training via Memory-Aware Techniques
by: Yang, Hanmei, et al.
Published: (2024)

Evaluating HPC-Style CPU Performance and Cost in Virtualized Cloud Infrastructures
by: Tharwani, Jay, et al.
Published: (2025)

Cost-Performance Evaluation of General Compute Instances: AWS, Azure, GCP, and OCI
by: Tharwani, Jay, et al.
Published: (2024)

Energy-Aware Computing in the Year 2026
by: Tchakoute, Roblex Nana, et al.
Published: (2026)

KVPR: Efficient LLM Inference with I/O-Aware KV Cache Partial Recomputation
by: Jiang, Chaoyi, et al.
Published: (2024)

Communication-Aware Diffusion Load Balancing for Persistently Interacting Objects
by: Taylor, Maya, et al.
Published: (2026)

KEET: Explaining Performance of GPU Kernels Using LLM Agents
by: Davis, Joshua H., et al.
Published: (2026)

LMDeploy Accelerates Mixed-Precision LLM Inference with TurboMind
by: Zhang, Li, et al.
Published: (2025)

Scaling Large-scale GNN Training to Thousands of Processors on CPU-based Supercomputers
by: Zhuang, Chen, et al.
Published: (2024)

Fine-Grained Energy Prediction For Parallellized LLM Inference With PIE-P
by: Dutt, Anurag, et al.
Published: (2025)

FAILS: A Framework for Automated Collection and Analysis of LLM Service Incidents
by: Battaglini-Fischer, Sándor, et al.
Published: (2025)

CASA: A Framework for SLO and Carbon-Aware Autoscaling and Scheduling in Serverless Cloud Computing
by: Qi, S., et al.
Published: (2024)

ReLATE: Learning Efficient Sparse Encoding for High-Performance Tensor Decomposition
by: Helal, Ahmed E., et al.
Published: (2025)

Modeling the Impact of Fiber Latency on Compute-Communication Overlap in Geo-Distributed Multi-Datacenter AI Training
by: Papavasileiou, Ioannis, et al.
Published: (2026)

BurstGPT: A Real-world Workload Dataset to Optimize LLM Serving Systems
by: Wang, Yuxin, et al.
Published: (2024)

DataStates-LLM: Scalable Checkpointing for Transformer Models Using Composable State Providers
by: Maurya, Avinash, et al.
Published: (2026)

ADELIA: Automatic Differentiation for Efficient Laplace Inference Approximations
by: Boudaoud, Afif, et al.
Published: (2026)

Towards a Scalable and Efficient PGAS-based Distributed OpenMP
by: Shan, Baodi, et al.
Published: (2024)

Fantasy: Efficient Large-scale Vector Search on GPU Clusters with GPUDirect Async
by: Liu, Yi, et al.
Published: (2025)

Efficient GPU-Centered Singular Value Decomposition Using the Divide-and-Conquer Method
by: Liu, Shifang, et al.
Published: (2025)

TaxBreak: Unmasking the Hidden Costs of LLM Inference Through Overhead Decomposition
by: Vellaisamy, Prabhu, et al.
Published: (2026)

Opt4GPTQ: Co-Optimizing Memory and Computation for 4-bit GPTQ Quantized LLM Inference on Heterogeneous Platforms
by: Zhang, Yaozheng, et al.
Published: (2025)

Efficient Serverless Cold Start: Reducing Library Loading Overhead by Profile-guided Optimization
by: Tariq, Syed Salauddin Mohammad, et al.
Published: (2025)

Efficient Fault Localization in a Cloud Stack Using End-to-End Application Service Topology
by: Mathews, Dhanya R, et al.
Published: (2025)

Towards a Peer-to-Peer Data Distribution Layer for Efficient and Collaborative Resource Optimization of Distributed Dataflow Applications
by: Scheinert, Dominik, et al.
Published: (2023)

Prism: Unleashing GPU Sharing for Cost-Efficient Multi-LLM Serving
by: Yu, Shan, et al.
Published: (2025)

AutoSP: Unlocking Long-Context LLM Training Via Compiler-Based Sequence Parallelism
by: Gupta, Ahan, et al.
Published: (2026)

MassiveGNN: Efficient Training via Prefetching for Massively Connected Distributed Graphs
by: Sarkar, Aishwarya, et al.
Published: (2024)

CoNST: Code Generator for Sparse Tensor Networks
by: Raje, Saurabh, et al.
Published: (2024)