Xiang Li
where else to find them
linkedin:xiang-li-40551a16a
what they talk about
all 16 in the filters
where they keep going back
- KubeCon + CloudNativeCon China2 editions2018–2019
all 15 in the filters
on a stage with
all 111 in the filters
who talks about the same things
- Lei Zhang33 talks
- Kevin Wang24 talks
- Tim Hockin21 talks
- Kris Nova23 talks
- Clayton Coleman12 talks
- Nikhita Raghunath31 talks
| title | conference | date |
|---|---|---|
| SwiftEP: Accelerating MoE Inference with Buffer Fusion and TMA Offloading | NSDI '26 | 2026-05-04 |
SwiftEP is an all-to-all library that fuses compute and RDMA buffers and offloads NVLink transfers to the Hopper TMA to remove redundant copies and free SMs… | ||
| Rowhammer-Based Trojan Injection: One Bit Flip Is Sufficient... | USENIX Security '25 | 2025-08-13 |
Claims a practical one-bit flip Rowhammer backdoor that works on full-precision models by flipping one exponent bit in the classification layer, finding… | ||
| Unlocking ECMP Programmability for Precise Traffic Control | NSDI '25 | 2025-04-28 |
P-ECMP implements a programming model, compiler, and runtime that reuses seldom-used ECMP groups to deterministically steer flows; compiler inputs are topology… | ||
| Holmes: Localizing Irregularities in LLM Training with Mega-scale GPU Clusters | NSDI '25 | 2025-04-28 |
Holmes localizes slow or spiky iterations in multi‑thousand GPU LLM training by logging only communication operators, classifying abnormal operators with a… | ||
| Statically Contextualizing Large Language Models with Typed Holes | OOPSLA 2024 | 2024-10-20 |
Integration of a typed live editor with LLM completions produces fewer hallucinations by sending a Hazel language server–constructed static slice: inferred… | ||
| More is Different: Prototyping and Analyzing a New Form of Edge Server with... | USENIX ATC 2024 | 2024-07-10 |
A commercial SoC Cluster with 60 Snapdragon 865 SoCs in a 2U rack and deployed for mobile cloud gaming is measured against servers using a benchmark suite that… | ||
| FwdLLM: Efficient Federated Finetuning of Large Language Models with Perturbed... | USENIX ATC 2024 | 2024-07-10 |
Introduces a BP-free federated finetuning pipeline that estimates gradients via perturbed inferences (forward gradients) and updates only small adapter… | ||
| MegaScale: Scaling Large Language Model Training to More Than 10,000 GPUs | 21st USENIX Symposium on Networked Systems Design and Implementation (NSDI '24) | 2024-04-16 |
MegaScale is a production system that trains LLMs on more than 10,000 GPUs and achieves concrete gains: 55.2% Model FLOPs Utilization when training a 175B… | ||
| Efficient Bottom-Up Synthesis for Programs with Local Variables | POPL 2024 | 2024-01-14 |
Introduces lifted interpretation, an algorithm that evaluates all grammar programs simultaneously to enumerate binding contexts and collapse equivalent… | ||
| TsuKing: Coordinating DNS Resolvers and Queries into Potent DoS Amplifiers | Black Hat Europe 2023 | 2023-12-04 |
A new DNS amplification method called TsuKing coordinates vulnerable resolvers via crafted referral chains and an attacker-controlled authoritative king to… | ||
| Auto-Paizo Games: Towards Understanding the Design of Games that Aim to Unify a Player’s Physical... | CHI PLAY 2023 | 2023-10-10 |
Paper builds and tests a setup that loans control of one hand to Electrical Muscle Stimulation (EMS) while the other hand, holding a phone, drives it via… | ||
| Fused Spectatorship: Designing Bodily Experiences Where Spectators Become Players | CHI PLAY 2023 | 2023-10-10 |
A computational Fused Spectatorship system using Electrical Muscle Stimulation (EMS) actuates both hands to play three custom games; one game, Godai, is a… | ||
| MaginotDNS: Attacking the Boundary of DNS Caching Protection | Black Hat USA 2023 | 2023-08-05 |
MaginotDNS is a cache‑poisoning technique that targets CDNS (resolvers that forward and recurse) by breaking bailiwick checking and injecting forged NS records… | ||
| Phoenix Domain Attack: Vulnerable Links in Domain Name Delegation and Revocation | Black Hat Asia 2023 | 2023-05-09 |
Claim: Phoenix Domain lets revoked domains stay resolvable at scale by abusing DNS cache operations; two exploit types (T1 injects NS records as cached entries… | ||
| Scalog: Seamless Reconfiguration and Total Order in a Scalable Shared Log | NSDI 2020 | 2020-02-25 |
Scalog implements a shared log as per-shard local ordering plus an aggregator-based ordering layer that builds a global cut vector and uses a deterministic… | ||
| Panel: Improving and Managing Kubernetes at Scale | KubeCon + CloudNativeCon North America 2019 | 2019-11-18 |
Multiple companies report concrete limits and fixes: control plane and etcd benchmarking shows storage and API server as bottlenecks; an 8k nodes -> 40k pods… | ||
| Panel Discussion: Leverage Cloud Native to Transform Your Enterprise – The China region | KubeCon + CloudNativeCon China 2019 | 2019-06-24 |
Large Chinese enterprises describe migrating legacy apps to Kubernetes since 2014–2016, running thousands of nodes and facing upgrade, scalability and… | ||
| Panel Discussion: Learn, Grow and Succeed in Kubernetes Contributions | KubeCon + CloudNativeCon + Open Source Summit China 2019 | 2019-06-24 |
Chinese Kubernetes contributions double down on practical work: contributors start with docs and small bug fixes, graduate to PR reviews, become… | ||
| Panel Discussion: Learn, Grow and Succeed in Kubernetes Contributions | KubeCon + CloudNativeCon China 2019 | 2019-06-24 |
Panelists map a clear contributor path: start with docs, tests or small bugfixes, submit PRs, earn member → reviewer → approver roles; core tools discussed… | ||
| Keynote: Cloud Native at Alibaba Scale | KubeCon + CloudNativeCon China 2019 | 2019-06-24 |
Alibaba runs Kubernetes clusters of tens of thousands of nodes and manages more than ten thousand heterogeneous applications, with stress-tested… | ||
| Panel Discussion: Leverage Cloud Native to Transform Your Enterprise – The China region | KubeCon + CloudNativeCon + Open Source Summit China 2019 | 2019-06-24 |
Panelists describe concrete migrations: OpenStack/VM fleets move to Kubernetes starting 2014–2016, clusters grow to thousands of nodes, and operators build a… | ||
| Keynote: Cloud Native at Alibaba Scale | KubeCon + CloudNativeCon + Open Source Summit China 2019 | 2019-06-24 |
Keynote: Cloud Native at Alibaba Scale - Xiang Li, Senior Staff Engineer, Alibaba Group and Technical Oversight Committee Member, CNCF
As one of the biggest… | ||
| Evolving Sigma: Adapting Kubernetes at Alibaba | KubeCon + CloudNativeCon China 2018 | 2018-11-13 |
Sigma is a homegrown cluster manager that integrates Kubernetes to run tens of thousands of servers and legacy non-container workloads; the team adapts… | ||
| nothing matches | ||