分布式、并行与集群计算

Distributed applications increasingly demand low end-to-end latency, especially in edge and cloud environments where co-located workloads contend for limited resources. Traditional load-balancing strategies are typically reactive and rely…

We introduce xLLM, an intelligent and efficient Large Language Model (LLM) inference framework designed for high-performance, large-scale enterprise-grade serving, with deep optimizations for diverse AI accelerators. To address these…

The evolution of blockchain technology, from its origins as a decentralized ledger for cryptocurrencies to its broader applications in areas like decentralized finance (DeFi), has significantly transformed financial ecosystems while…

Unmanned aerial vehicle (UAV) swarms are increasingly used in critical applications such as aerial mapping, environmental monitoring, and autonomous delivery. However, the reliability of these systems is highly dependent on uninterrupted…

The rapid adoption of large language models and multimodal foundation models has made multimodal data preparation pipelines critical AI infrastructure. These pipelines interleave CPU-heavy preprocessing with accelerator-backed (GPU/NPU/TPU)…

We consider an asynchronous network of $n$ parties connected to each other via secure channels, up to $t$ of which are byzantine. We study common coin tossing, a task where the parties try to agree on an unpredictable random value, with…

Large language model (LLM) services have become an integral part of search, assistance, and decision-making applications. However, unlike traditional web or microservices, the hardware and software stack enabling LLM inference deployment is…

With the increasing computational capability of mobile devices, deploying agentic retrieval-augmented generation (RAG) locally on heterogeneous System-on-Chips (SoCs) has become a promising way to enhance LLM-based applications. However,…

Shared L1-memory clusters of streamlined instruction processors (processing elements - PEs) are commonly used as building blocks in modern, massively parallel computing architectures (e.g. GP-GPUs). Scaling out these architectures by…

Edge computing environments host increasingly complex microservice-based IoT applications that are prone to performance anomalies propagating across dependent services. Identifying the faulty component (root cause localization) and the…

This is the first of five papers comprising The Semantic Arrow of Time. The argument begins with a claim: computing's arrow of time is semantic, not thermodynamic. The direction in which meaning is preserved or destroyed across transactions…

Message passing is widely assumed to be a fundamental primitive of distributed systems. This paper argues that conventional message systems embed a category mistake: they misinterpret logical dependency relations as temporal propagation…

Speculative Decoding (SD) has emerged as a premier technique for accelerating Large Language Model (LLM) inference by decoupling token generation into rapid drafting and parallel verification. While recent advancements in self-speculation…

Unix tools such as ls, cp, mv, and rename expose a filesystem abstraction that appears to present a single, authoritative state evolving through atomic transitions. This abstraction is false. We present a systematic Forward-In-Time-Only…

We introduce an algorithm that performs a one-directional mesh overset of a parallel forest of octrees with another distributed mesh of unrelated partition. The forest mesh consists of several adaptively refined octrees. Individual smooth…

Federated learning (FL) enables collaborative model training over distributed private data. However, sustaining open participation requires incentive mechanisms that compensate contributors for their resources and risks. Enabled by Web3…

The increasing adoption of UAVs equipped with advanced sensors and GPU-accelerated edge computing has enabled real-time AI-driven applications in domains such as precision agriculture, wildfire monitoring, and environmental conservation.…

Stencil computation constitutes a cornerstone of scientific computing, serving as a critical kernel in domains ranging from fluid dynamics to weather simulation. While stencil computations are conventionally regarded as memory-bound and…

Multi-tenant AI inference platforms must balance resource utilization against service-level guarantees under variable demand. Conventional approaches fail to achieve this balance: dedicated endpoints strand capacity on idle models, while…

GPU-accelerated server platforms that share most of their hardware architecture often require separate firmware images due to minor hardware differences--different component identifiers, thermal profiles, or interconnect topologies. I built…