Flattened Clos: Designing High-performance Deadlock-free Expander Data Center Networks Using Graph Contraction
Shizhen Zhao, Qizhou Zhang, Peirui Cao, Xiao Zhang, Xinbing Wang, Chenghu Zhou
Abstract
Clos networks have witnessed the successful deployment of RoCE in production data centers. However, as DCN bandwidth keeps increasing, building Clos networks is becoming cost-prohibitive and thus the more cost-efficient expander graph has received much attention in recent literature. Unfortunately, the existing expander graphs' topology and routing designs may contain Cyclic Buffer Dependency (CBD) and incur deadlocks in PFC-enabled RoCE networks.
We propose Flattened Clos (FC), a topology/routing codesigned approach, to eliminate the PFC-induced deadlocks in expander networks. FC's topology and routing are designed in three steps: 1) logically divide each ToR switch into k virtual layers and establish connections only between adjacent virtual layers; 2) generate virtual up-down paths for routing; 3) flatten the virtual multi-layered network and the virtual up-down paths using graph contraction. We rigorously prove that FC's design is deadlock-free and validate this property using a real testbed and packet-level simulation. Compared to expander graphs with the edge-disjoint-spanning-tree (EDST) based routing (a state-of-art CBD-free routing algorithm for expander graphs), FC reduces the average hop count by at least 50% and improves network throughput by 2 -10× or more. Compared to Clos networks with up-down routing, FC increases network throughput by 1.1 -2× under all-to-all and uniform random traffic patterns.
Ask about this paper
Your agent reads all of it.
Lune indexed this paper to the last equation, along with the top-tier papers that cite it. Ask a question and the answer quotes them.
Your agent calls
Luneget_paper_fulltext
Free to start. No credit card required.
Terminal
Install the CLIlune papers fulltext 8eb08883-92ae-4ce8-861d-73fdc5e5ee68Cited by top-tier papers1
Ask how each one uses itBuilds on7
- Jupiter evolving: transforming google's datacenter network via optical circuit switches and software-defined networkingLeon Poutievski, Omid Mashayekhi, Joon Ong, Arjun Singh et al.SIGCOMM 2022 · 230 citations
- When Cloud Storage Meets RDMAYixiao Gao, Qiang Li, Lingbo Tang, Yongqing Xi et al.NSDI 2021 · 228 citations
- Sirius: A Flat Datacenter Network with Nanosecond Optical SwitchingHitesh Ballani, Paolo Costa, Raphael Behrendt, Daniel Cletheroe et al.SIGCOMM 2020 · 204 citations
- A throughput-centric view of the performance of datacenter topologiesPooria Namyar, Sucha Supittayapornpong, Mingyang Zhang, Minlan Yu et al.SIGCOMM 2021 · 29 citations
- FatPaths: routing in supercomputers and data centers when shortest paths fall shortMaciej Besta, Marcel Schneider, Marek Konieczny, Karolina Cynk et al.SC 2020 · 25 citations
Related papers
- FLB: Fine-grained Load Balancing for Lossless Datacenter NetworksJinbin Hu, Wenxue Li, Xiangzhou Liu, Junfeng Wang et al.USENIX ATC 2025 · 10 citations
- Cost-Effective Live Expansion of Three-Stage Switching Networks without Blocking or Connection RearrangementTakeru Inoue, Toru Mano, Takeaki UnoINFOCOM 2023 · 6 citations
- Expanding across time to deliver bandwidth efficiency and low latencyWilliam M. Mellette, Rajdeep Das, Yibo Guo, Rob McGuinness et al.NSDI 2020 · 194 citations
- SEEC: stochastic escape express channelMayank Parasar, Natalie D. Enright Jerger, Paul V. Gratz, Joshua San Miguel et al.SC 2021 · 15 citations
- Stay in your Lane: A NoC with Low-overhead Multi-packet BypassingHossein Farrokhbakht, Paul V. Gratz, Tushar Krishna, Joshua San Miguel et al.HPCA 2022 · 21 citations
