Ennan Zhai
where else to find them
linkedin:ennan-zhai-26818b17
what they talk about
all 12 in the filters
where they keep going back
- NSDI2 editions2020–2023
all 10 in the filters
on a stage with
all 156 in the filters
who talks about the same things
nobody close enough yet
| title | conference | date |
|---|---|---|
| StriaTrace: Efficient Tracing and Diagnosis for Online LLM Inference | 20th USENIX Symposium on Operating Systems Design and Implementation (OSDI '26) | 2026-07-13 |
StriaTrace implements host and device collectors that record only 13 synchronization spans and kernel boundaries plus memory transfers, with a pass-by sampler… | ||
| TrainMover: An Interruption-Resilient Runtime for ML Training | 20th USENIX Symposium on Operating Systems Design and Implementation (OSDI '26) | 2026-07-13 |
TrainMover is an ML training runtime that prepares standby GPUs ahead of interruptions by pre-recording output tensors and running communication-free sandboxed… | ||
| EROICA: Online Performance Troubleshooting for Large-scale Model Training | NSDI '26 | 2026-05-04 |
EROICA is an online profiler that triggers only on iteration slowdowns, profiles all workers simultaneously, extracts beta, mu, sigma fingerprints for… | ||
| Come Hell or Still Water: Alleviating Tail Latency in Cloud Block Store | NSDI '26 | 2026-05-04 |
Per-proxy dual-bucket token buckets add a main bucket (refilled every 10 ms) and a check bucket that only steady segments can draw from; hot segments drain the… | ||
| ServeGen: Workload Characterization and Generation of Large Language... | NSDI '26 | 2026-05-04 |
Characterization uses production traces spanning 3.5 billion requests over 4 months across 10 models and finds that traffic is dominated by a few customers —… | ||
| HeteCCL: Synthesizing Near-Optimal Collective Communication Schedules for Heterogeneous | NSDI '26 | 2026-05-04 |
HeteCCL generates near‑optimal collective schedules by quantizing data into step‑level chunks, modeling topology and link bandwidth precisely, and casting… | ||
| Roaming Free in the VR World with MP2 | USENIX ATC '25 | 2025-07-07 |
A user-space centralized overlay called MP2 coordinates per-headset two Wi‑Fi interfaces via a centralized controller that collects RSSI and frame-size stats… | ||
| SimAI: Unifying Architecture Design and Performance Tuning for Large-Scale Large... | NSDI '25 | 2025-04-28 |
SimAI claims a unified, high‑precision LLM training simulator that matches real runs 98.1% alignment by hijacking training frameworks (Megatron, DeepSpeed)… | ||
| Learning Production-Optimized Congestion Control Selection for Alibaba Cloud CDN | NSDI '25 | 2025-04-28 |
AliCCS is a production system that learns to pick among existing congestion controllers using a GAN-based invariant representation to remove /24 hidden-state… | ||
| Evolution of Aegis: Fault Diagnosis for AI Model Training Service in Production | NSDI '25 | 2025-04-28 |
Aegis is a production fault diagnosis system for AI model training that evolves from enhanced legacy tools to a customized collective communication library… | ||
| Mitigating Scalability Walls of RDMA-based Container Networks | NSDI '25 | 2025-04-28 |
ScalaCN infers RNIC internals via combinatorial causal testing and links specific config dimensions to failures, finding that increasing flow-group query cost… | ||
| Burstable Cloud Block Storage with Data Processing Units | 18th USENIX Symposium on Operating Systems Design and Implementation (OSDI '24) | 2024-07-10 |
An end-to-end system called BurstCBS runs on DPUs to balance per-core load with added FPGA queues, uses a two-tier memory pool to prefill buffers without… | ||
| Sirius: Composing Network Function Chains into P4-Capable Edge Gateways | 21st USENIX Symposium on Networked Systems Design and Implementation (NSDI '24) | 2024-04-16 |
System composes dozens of distinct NF chains into a single P4 program, synthesizing traffic identification tables, pipelining loops via recirculations, and… | ||
| LuoShen: A Hyper-Converged Programmable Gateway for Multi-Tenant Multi-Service Edge... | 21st USENIX Symposium on Networked Systems Design and Implementation (NSDI '24) | 2024-04-16 |
LuoShen consolidates a full VPC network into a 2U server-switch with a P4-centric pipeline, using pipeline folding and meticulous on-chip resource budgeting to… | ||
| Reasoning about Network Traffic Load Property at Production Scale | 21st USENIX Symposium on Networked Systems Design and Implementation (NSDI '24) | 2024-04-16 |
NSDI '24 - Reasoning about Network Traffic Load Property at Production Scale
Ruihan Li, Peking University and Alibaba Cloud; Fangdan Ye, Yifei Yuan, Ruizhen… | ||
| Norma: Towards Practical Network Load Testing | NSDI 2023 | 2023-04-17 |
Norma implements a practical network load tester on programmable switch ASICs (Tofino) with a stateful protocol abstraction mapped into the pipeline, a… | ||
| Cetus: Releasing P4 Programmers from the Chore of Trial and Error Compiling | 19th USENIX Symposium on Networked Systems Design and Implementation (NSDI '22) | 2022-04-04 |
NSDI '22 - Cetus: Releasing P4 Programmers from the Chore of Trial and Error Compiling
Yifan Li, Tsinghua University and Alibaba Group; Jiaqi Gao, Ennan Zhai… | ||
| Static Detection of Silent Misconfigurations with Deep Interaction Analysis | OOPSLA 2021 | 2021-10-17 |
Builds a tool called ConfigX that statically links configuration entries to code via LLVM IR, applies customized data- and control-flow tainting to derive… | ||
| Check before You Change: Preventing Correlated Failures in Service Updates | NSDI 2020 | 2020-02-25 |
CloudCanary audits service-update snapshots in real time by reusing previous audit caches and translating dependency graphs into SAT problems, using SnapAudit… | ||
| Lock-Free Collaboration Support for Cloud Storage Services with Operation Inference | FAST '20 | 2020-02-24 |
Reverse engineering and measurements reveal long synchronization tails, missing locks, and many false conflicts in popular cloud drives. The system infers… | ||
| nothing matches | ||