分布式、并行与集群计算

We revisit the relationship between two fundamental models of distributed computation: the asynchronous message-passing model with up to $f$ crash failures ($\operatorname{AMP}_f$) and the Heard-Of model with up to $f$ message omissions…

This paper introduces MonadBFT, a novel Byzantine Fault Tolerant (BFT) consensus protocol that advances both performance and robustness. MonadBFT is implemented as the consensus protocol in the Monad blockchain. As a HotStuff-family…

Approximate Membership Query (AMQ) structures are essential for high-throughput systems in databases, networking, and bioinformatics. While Bloom filters offer speed, they lack support for deletions. Existing GPU-based dynamic alternatives,…

The rapid proliferation of the Internet of Things (IoT) and smart applications has led to a surge in data generated by distributed sensing devices. Edge computing is a mainstream approach to managing this data by pushing computation closer…

Multi-agent LLM orchestration incurs synchronization costs scaling as O(n x S x |D|) in agents, steps, and artifact size under naive broadcast -- a regime I term broadcast-induced triply-multiplicative overhead. I argue this pathology is a…

Protocols for causal message delivery are widely used in distributed systems. Traditionally, causal delivery can be enforced either on the message sender's side or on the receiver's side. The traditional sender-side approach avoids the…

The ongoing convergence of HPC and cloud computing presents a fundamental challenge: HPC applications, designed for static and homogeneous supercomputers, are ill-suited for the dynamic, heterogeneous, and volatile nature of the cloud.…

For deep learning inference on edge devices, hardware configurations achieving the same throughput can differ by 2$\times$ in power consumption, yet operators often struggle to find the efficient ones without exhaustive profiling. Existing…

Parallel Byzantine Fault Tolerant (BFT) protocols based on committee-based sharding improve scalability but weaken safety since smaller node groups are responsible for consensus. Recent approaches integrate trusted execution environments…

The datacenter industry is converging on SmartNIC-based resource management. Wave (Humphries et al., ASPLOS '25) demonstrates the practical feasibility of offloading kernel thread scheduling, memory management, and RPC stacks to the ARM…

The Markov approximation is arguably the most ubiquitous tool in physics, underpinning quantum master equations, stochastic processes, and -- via Shannon's channel model and Lamport's logical clocks -- the foundational assumptions of…

Despite advances in assistive technologies, Blind and Low-Vision (BLV) individuals continue to face challenges in understanding their surroundings. Delivering concise, useful, and timely scene descriptions for ambient perception remains a…

For nearly two decades, population protocols have been extensively studied, yielding efficient solutions for central problems in distributed computing, including leader election, and majority computation, a predicate type in Presburger…

The scientific computing ecosystem in Python is largely confined to single-node parallelism, creating a gap between high-level prototyping in NumPy and high-performance execution on modern supercomputers. The increasing prevalence of…

LLM serving is increasingly dominated by decode attention, which is a memory-bound operation due to massive KV cache loading from global memory. Meanwhile, real-world workloads exhibit substantial, hierarchical shared prefixes across…

Collective communication is becoming increasingly important in data center and supercomputer workloads with an increase in distributed AI related jobs. However, existing libraries that provide collective support such as NCCL, RCCL, and…

Scheduling deep learning (DL) models to train on powerful clusters with accelerators like GPUs and TPUs, presently falls short, either lacking fine-grained heterogeneity awareness or leaving resources substantially under-utilized. To fill…

Distributed deep neural network training necessitates efficient GPU collective communications, which are inherently susceptible to deadlocks. GPU collective deadlocks arise easily in distributed deep learning applications when multiple…

Traditional lock-free parallel algorithms for combinatorial optimization problems, such as shortest paths, stable matching, and job scheduling require programmers to write problem-specific routines and synchronization code. We propose a…

Agentic reinforcement learning (RL) has emerged as a transformative workload in cloud clusters, enabling large language models (LLMs) to solve complex problems through interactions with real world. However, unlike traditional RL, agentic RL…