分布式、并行与集群计算

Recently, there has been increased interest in globally distributed training, which has the promise to both reduce training costs and democratize participation in building large-scale foundation models. However, existing models trained in a…

Scaling Mixture-of-Experts (MoE) training introduces systems challenges absent in dense models. Because each token activates only a subset of experts, this sparsity allows total parameters to grow much faster than per-token computation,…

Modern enterprise platforms increasingly depend on distributed microservices, analytical data platforms, and external APIs to construct composite responses for applications. Orchestrating data retrieval across these heterogeneous systems is…

Function-as-a-Service (FaaS) platforms provide scalable and cost-efficient execution but suffer from increased latency and resource overheads in complex applications comprising multiple functions, particularly due to double billing when…

Large language models (LLMs) are increasingly deployed in AI infrastructure, driving the need for high throughput, resource efficient serving systems. Disaggregated LLM serving, which separates prompt prefill from auto-regressive decode,…

We present SPDL (Scalable and Performant Data Loading), an open-source, framework-agnostic library designed for efficiently loading array data to GPU. Data loading is often a bottleneck in AI applications, and is challenging to optimize…

Complex systems often show macroscopic coherent behavior due to the interactions of microscopic agents like molecules, cells, or individuals in a population with their environment. However, simulating such systems poses several…

On the one side, the formalism of Global Transformations comes with the claim of capturing any transformation of space that is local, synchronous and deterministic. The claim has been proven for different classes of models such as mesh…

In this paper we propose a method for analyzing services deployed in serverless platforms. These services typically consists of orchestrated functions that can exhibit complex and non-conservative information flows due to the interaction of…

Data replication is a critical aspect of data center design, as it ensures high availability, scalability, and fault tolerance. However, replicas need to be coordinated to maintain convergence and database integrity constraints under…

In high-speed rail (HSR) systems, federated learning (FL) enables cross-departmental flow prediction without sharing raw data. However, existing schemes suffer from two key limitations: (1) insufficient incentives, leading to free-riding…

Containerized microservices are widely adopted for latency-sensitive and compute-intensive applications, with Kubernetes (K8s) as the dominant orchestration platform. However, automating the deployment and management of multi-service…

Autoscaling in cloud-native platforms like Kubernetes is reactive and metric-driven, leading to a strategic void problem. This comes from the decoupling of higher-level business policies from lower-level resource provisioning. The strategic…

For fifty years, networking has fragmented whenever new workloads exposed hidden assumptions about time, ordering, failure, and trust. This paper argues that the current interconnect landscape -- NVLink, UALink, Ultra Ethernet,…

Operating a global, real-time platform at Uber's scale requires infrastructure that is both resilient and cost-efficient. Historically, reliability was ensured through a costly 2x capacity model--each service provisioned to handle global…

iCloud Drive presents a filesystem interface but implements cloud synchronization semantics that diverge from POSIX in fundamental ways. This divergence is not an implementation bug; it is a Category Mistake -- the same one that pervades…

Ultra-reliable low-latency vehicular communications (URLLC) require sufficient physical-layer (PHY) compute headroom at the network edge, where roadside units (RSUs) and compact next-generation base stations (gNBs) must meet strict timing…

To meet strict Service-Level Objectives (SLOs),contemporary Large Language Models (LLMs) decouple the prefill and decoding stages and place them on separate GPUs to mitigate the distinct bottlenecks inherent to each phase. However, the…

In-memory key-value datastores have become indispensable building blocks of modern cloud-native infrastructures, yet their evolution faces scalability, compatibility, and sustainability constraints. The current literature lacks an…

Leader election is a fundamental problem in distributed computing, particularly within programmable matter systems, where coordination among simple computational entities is crucial for solving complex tasks. In these systems, particles…