| 标题 |
 Hybrid-Granularity Network Load balancing for Distributed AI Model Training 用于分布式人工智能模型训练的混合粒度网络负载均衡
相关领域
粒度
计算机科学
分布式计算
负载平衡（电力）
培训（气象学）
计算机体系结构
计算机网络
人工智能
操作系统
气象学
网格
物理
几何学
数学
|
| 网址 | |
| DOI | |
| 其它 |
摘要：With the increasing number of parameters in artificial intelligence (AI) models, distributed AI model training using numerous servers within data centers has become commonplace. However, traditional load balancing strategies in data center networks are not suitable for distributed training. Unbalanced network load can reduce net work throughput and degrade application performance. Therefore, we propose a hybrid-granularity network load balancing strategy to address the issue, consisting of global path planing in advance, periodic flow scheduling and real-time packet rerouting. The simu lation experiment results demonstrate that our method can reduce throughput imbalance and exposed communication time. |
| 求助人 | |
| 下载 | |
|
温馨提示：该文献已被科研通 学术中心 收录，前往查看
科研通『学术中心』是文献索引库，收集文献的基本信息（如标题、摘要、期刊、作者、被引量等），不提供下载功能。如需下载文献全文，请通过文献求助获取。
|
SYLH
科研小民工
杳鸢
劲秉
迟大猫
Zn
小林太郎
故意的傲玉
sweetbearm
lalala
36456657
RC_Wang
浦肯野
kingwill
iNk
火星上的菲鹰
pluto
从心
坚强的广山
maox1aoxin
皖公网安备34019202002308
