Related papers: Avoiding Cross-Datacenter Collective Congestion vi…

Data center networks leverage multiple parallel paths connecting end host pairs to offer high bisection bandwidth for cluster computing applications. However, state of the art distributed multi-pathing protocols such as Equal Cost Multipath…

We consider a scenario involving computations over a massive dataset stored distributedly across multiple workers, which is at the core of distributed learning algorithms. We propose Lagrange Coded Computing (LCC), a new framework to…

Accurate traffic flow prediction is vital for optimizing urban mobility, yet it remains difficult in many cities due to complex spatio-temporal dependencies and limited high-quality data. While deep graph-based models demonstrate strong…

Accurate traffic prediction is vital for effective traffic management during hurricane evacuation. This paper proposes a predictive modeling system that integrates Multilayer Perceptron (MLP) and Long-Short Term Memory (LSTM) models to…

LAPS identifies and disaggregates requests with different prompt lengths in LLM serving to reduce TTFT latency. While recent systems have decoupled the prefill and decode stages to improve throughput, they still rely on unified scheduling…

Heterogeneous wireless networks have evolved to reach application requirements for low latency and high throughput on Internet access. Recent studies have improved network performance employing the Multipath TCP, which aggregates flows from…

The interconnection network is a crucial subsystem in High-Performance Computing clusters and Data-centers, guaranteeing high bandwidth and low latency to the applications' communication operations. Unfortunately, congestion situations may…

As the scale of training large language models (LLMs) increases, one emergent failure is silent data corruption (SDC), where hardware produces incorrect computations without explicit failure signals. In this work, we are the first to…

We present MonkeyTree, the first system to mitigate network congestion in multi-tenant GPU clusters through job-migration based defragmentation rather than network-layer techniques. As cloud operators co-locate ML training jobs on shared,…

This paper provides an in-depth characterization of GPU-accelerated systems, to understand the interplay between overlapping computation and communication which is commonly employed in distributed training settings. Due to the large size of…

Network congestion and packet loss pose an ever-increasing challenge to video streaming. Despite the research efforts toward making video encoding schemes resilient to lossy network conditions, forwarding devices have not considered…

Load balancing at transport layer is an important function in data centers, content delivery networks, and mobile networks, where per-connection consistency (PCC) has to be met for optimal performance. Cloud-native L4 load balancers are…

With the rapid adoption of Large Language Models (LLMs), LLM-adapters have become increasingly common, providing lightweight specialization of large-scale models. Serving hundreds or thousands of these adapters on a single GPU allows…

Modern machine learning frameworks can train neural networks using multiple nodes in parallel, each computing parameter updates with stochastic gradient descent (SGD) and sharing them asynchronously through a central parameter server. Due…

Coflow is a network abstraction used to represent communication patterns in data centers. The coflow scheduling problem in large data centers is one of the most important $NP$-hard problems. Many previous studies on coflow scheduling mainly…

Distributed gradient descent (DGD) is an efficient way of implementing gradient descent (GD), especially for large data sets, by dividing the computation tasks into smaller subtasks and assigning to different computing servers (CSs) to be…

Nowadays, service providers often deploy multiple types of LLM services within shared clusters. While the service colocation improves resource utilization, it introduces significant interference risks for latency-sensitive (LS)…

Fine-tuning large language models (LLMs) often exceeds GPU memory limits, prompting systems to offload model states to CPU memory. However, existing offloaded training frameworks like ZeRO-Offload treat all parameters equally and update the…

Data centers (DCs) nowadays house tens of thousands of servers and switches, interconnected by high-speed communication links. With the rapid growth of cloud DCs, in both size and number, tremendous efforts have been undertaken to…

When gradient descent (GD) is scaled to many parallel workers for large scale machine learning problems, its per-iteration computation time is limited by the straggling workers. Straggling workers can be tolerated by assigning redundant…