USENIX | The Advanced Computing Systems Association

Accelerating Design Space Exploration for {LLM} Training Systems with Multi-experiment Parallel Simulation

Gui F, Gao K, Chen L, Li D, Liu V, Zhang R, Yang H, Xiong D. 2025. Accelerating Design Space Exploration for {LLM} Training Systems with Multi-experiment Parallel Simulation. 22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25). :473--488.

{CellReplay}: Towards accurate record-and-replay for cellular networks

Sentosa W, Chandrasekaran B, P. Godfrey B, Hassanieh H. 2025. {CellReplay}: Towards accurate record-and-replay for cellular networks. 22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25). :1169--1186.

Beehive: A Scalable Disaggregated Memory Runtime Exploiting Asynchrony of Multithreaded Programs

Li Q, Huang H, Liu Y, Xia Y, Zhang J, Zhou M, Feng X, Cui H, Chen Q, Shan Y et al.. 2025. Beehive: A Scalable Disaggregated Memory Runtime Exploiting Asynchrony of Multithreaded Programs. 22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25). :167--187.

{AutoCCL}: Automated Collective Communication Tuning for Accelerating Distributed and Parallel {DNN} Training

Xu G, Le Z, Chen Y, Lin Z, Jin Z, Miao Y, Li C. 2025. {AutoCCL}: Automated Collective Communication Tuning for Accelerating Distributed and Parallel {DNN} Training. 22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25). :667--683.

{DISC}: Backpressure Mitigation In Multi-tier Applications With Distributed Shared Connection

Ekane B, Mvondo D, Lachaize R, Bromberg Y-D, Tchana A, Hagimont D. 2025. {DISC}: Backpressure Mitigation In Multi-tier Applications With Distributed Shared Connection. 22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25). :55--70.

Making Serverless {Pay-For-Use} a Reality with Leopard

Cao T, Arpaci-Dusseau AC, Arpaci-Dusseau RH, Caraza-Harter T. 2025. Making Serverless {Pay-For-Use} a Reality with Leopard. 22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25). :189--204.

Mitigating Scalability Walls of {RDMA-based} Container Networks

Liu W, Qian K, Li Z, Qian F, Xu T, Liu Y, Guan Y, Zhu S, Xu H, Xi L et al.. 2025. Mitigating Scalability Walls of {RDMA-based} Container Networks. 22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25). :1049--1065.

Ladder: A Convergence-based Structured {DAG} Blockchain for High Throughput and Low Latency

Hu D, Wang J, Liu X, Xu H, Wu X, Shahzad M, Liu G, Li K. 2025. Ladder: A Convergence-based Structured {DAG} Blockchain for High Throughput and Low Latency. 22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25). :779--794.

High-level Programming for Application Networks

Zhu X, Wang Y, Liu B, Wu Y, Bojanic N, Chen J, Bernstein GLouis, Krishnamurthy A, Kumar S, Mahajan R et al.. 2025. High-level Programming for Application Networks. 22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25). :915--935.

Holmes: Localizing Irregularities in {LLM} Training with Mega-scale {GPU} Clusters

Yao Z, Hu P, Miao C, Jia X, Liang Z, Xu Y, He C, Lu H, Chen M, Li X et al.. 2025. Holmes: Localizing Irregularities in {LLM} Training with Mega-scale {GPU} Clusters. 22nd USENIX Symposium on Networked Systems Design and Implementation (NSDI 25). :523--540.

Pages