Blog

LLM推理论文精读2 -- PowerNPU[ASPLOS'25]

PowerNPU:Fast On-device LLM Inference with NPUs

Oct 15, 2024

LLM推理论文精读1 -- FlexGen[ICML'23]

Oct 12, 2024

Transformer Model Learning

Transformer模型入门

Oct 10, 2024

Distributed Training

ElasticFlow: Serverless DDL Training is better?

ASPLOS'23弹性调度论文ElasticFlow精读

Jul 21, 2023

Math and Algorithm

Combinatorial Online Optimization For Online DML Job Scheduling

在线调度问题的组合优化求解方法

Jul 7, 2023

Math and Algorithm

A Preemptive Scheduler for DML Jobs In Edge-Cloud·22'JSAC

云边协同场景中抢占式调度策略

May 30, 2023

Distributed Training

A Deadline-Aware Scheduler for DLT Jobs·21'SoCC

Chronus论文精读

Apr 13, 2023

Math and Algorithm

Primal-Dual Approximate Algorithm Based On LP

基于LP的近似算法

Apr 13, 2023

Distributed Training

Deep Learning Workloads Scheduling in GPU Clusters

GPU集群中的DLT任务调度工作研讨

Apr 8, 2023

Distributed Training

How To Support Elastic Trainging

弹性伸缩深度学习分布式训练

Mar 26, 2023