Related papers: FlowDyn: Towards a Dynamic Flowlet Gap Detection u…
Distributed computing systems often consist of hundreds of nodes, executing tasks with different resource requirements. Efficient resource provisioning and task scheduling in such systems are non-trivial and require close monitoring and…
Any-to-any generation seeks to translate between arbitrary subsets of modalities, enabling flexible cross-modal synthesis. Despite recent success, existing flow-based approaches are challenged by their inefficiency, as they require…
Federated learning (FL) is a subfield of machine learning that avoids sharing local data with a central server, which can enhance privacy and scalability. The inability to consolidate data leads to a unique problem called dataset imbalance,…
Deep Neural Networks (DNNs) have revolutionized numerous applications, but the demand for ever more performance remains unabated. Scaling DNN computations to larger clusters is generally done by distributing tasks in batch mode using…
With the increasing use of renewable generation in power systems, responsive resources will be necessary to support primary frequency control in future low-inertia/under-damped power systems. Flexible loads can provide fast-frequency…
Recent literature has proved that stable dynamic routing algorithms have solid theoretical foundation that makes them suitable to be implemented in a real protocol, and used in practice in many different operational network contexts. Such…
We consider multiuser scheduling in wireless networks with channel variations and flow-level dynamics. Recently, it has been shown that the MaxWeight algorithm, which is throughput-optimal in networks with a fixed number users, fails to…
AI deployment increasingly resembles a pipeline of data transformation, fine-tuning, and agent interactions rather than a monolithic LLM job; recent examples include RLHF/RLAIF training and agentic workflows. To cope with this shift, we…
Structured LLM workflows, where specialized LLM sub-agents execute according to a predefined graph, have become a powerful abstraction for solving complex tasks. Optimizing such workflows, i.e., selecting configurations for each sub-agent…
Deep neural networks with more parameters and FLOPs have higher capacity and generalize better to diverse domains. But to be deployed on edge devices, the model's complexity has to be constrained due to limited compute resource. In this…
While there is a large body of research on efficient processing of deep neural networks (DNNs), ultra-low-latency realization of these models for applications with stringent, sub-microsecond latency requirements continues to be an…
Deep Reinforcement Learning (DRL) has recently been proposed as a methodology to discover complex Active Flow Control (AFC) strategies [Rabault, J., Kuchta, M., Jensen, A., Reglade, U., & Cerardi, N. (2019): "Artificial neural networks…
We introduce FatPaths: a simple, generic, and robust routing architecture that enables state-of-the-art low-diameter topologies such as Slim Fly to achieve unprecedented performance. FatPaths targets Ethernet stacks in both HPC…
We consider an optimal flow distribution problem in which the goal is to find a radial configuration that minimizes resistance-induced quadratic distribution costs while ensuring delivery of inputs from multiple sources to all sinks to meet…
In this work, we show that the impact of model capacity varies across timesteps: it is crucial for the early and late stages but largely negligible during the intermediate stage. Accordingly, we propose FlowBlending, a stage-aware…
The use of under-utilized Internet resources is widely recognized as a viable form of high performance computing. Sustained processing power of roughly 40T FLOPS using 4 million volunteered Internet hosts has been reported for…
Datacenter networks routinely support the data transfers of distributed computing frameworks in the form of coflows, i.e., sets of concurrent flows related to a common task. The vast majority of the literature has focused on the problem of…
Transformer-based language models spread FLOPs uniformly across input sequences. In this work we demonstrate that transformers can instead learn to dynamically allocate FLOPs (or compute) to specific positions in a sequence, optimising the…
We consider a basic model of a dynamical distribution network, modeled as a directed graph with storage variables corresponding to every vertex and flow inputs corresponding to every edge, subject to unknown but constant inflows and…
Inspired by biological swimming and flying with distributed sensing, we propose a data-driven approach for load estimation that relies on complex networks. We exploit sparse, real-time pressure inputs, combined with pre-trained transition…