SeqBalance: Congestion-Aware Load Balancing with no Reordering for RoCE
Fuente:
arXiv
Salvato in:
| Autori principali: | , , , , , |
|---|---|
| Natura: | Preprint |
| Pubblicazione: |
2024
|
| Soggetti: | |
| Accesso online: | |
| Tags: |
Aggiungi Tag
Nessun Tag, puoi essere il primo ad aggiungerne!!
|
| _version_ | 1866929419787960320 |
|---|---|
| author | Luo, Huimin Zhang, Jiao Yu, Mingxuan Pan, Yongchen Pan, Tian Huang, Tao |
| author_facet | Luo, Huimin Zhang, Jiao Yu, Mingxuan Pan, Yongchen Pan, Tian Huang, Tao |
| contents | Remote Direct Memory Access (RDMA) is widely used in data center networks because of its high performance. However, due to the characteristics of RDMA's retransmission strategy and the traffic mode of AI training, current load balancing schemes for data center networks are unsuitable for RDMA. In this paper, we propose SeqBalance, a load balancing framework designed for RDMA. SeqBalance implements fine-grained load balancing for RDMA through a reasonable design and does not cause reordering problems. SeqBalance's designs are all based on existing commercial RNICs and commercial programmable switches, so they are compatible with existing data center networks. We have implemented SeqBalance in Mellanox CX-6 RNICs and Tofino switches. The results of hardware testbed experiments and large-scale simulations show that compared with existing load balancing schemes, SeqBalance improves 18.7% and 33.2% on average FCT and 99th percentile FCT. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2407_09808 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | SeqBalance: Congestion-Aware Load Balancing with no Reordering for RoCE Luo, Huimin Zhang, Jiao Yu, Mingxuan Pan, Yongchen Pan, Tian Huang, Tao Networking and Internet Architecture Remote Direct Memory Access (RDMA) is widely used in data center networks because of its high performance. However, due to the characteristics of RDMA's retransmission strategy and the traffic mode of AI training, current load balancing schemes for data center networks are unsuitable for RDMA. In this paper, we propose SeqBalance, a load balancing framework designed for RDMA. SeqBalance implements fine-grained load balancing for RDMA through a reasonable design and does not cause reordering problems. SeqBalance's designs are all based on existing commercial RNICs and commercial programmable switches, so they are compatible with existing data center networks. We have implemented SeqBalance in Mellanox CX-6 RNICs and Tofino switches. The results of hardware testbed experiments and large-scale simulations show that compared with existing load balancing schemes, SeqBalance improves 18.7% and 33.2% on average FCT and 99th percentile FCT. |
| title | SeqBalance: Congestion-Aware Load Balancing with no Reordering for RoCE |
| topic | Networking and Internet Architecture |
| url | https://arxiv.org/abs/2407.09808 |