Related papers: Host-Side Telemetry for Performance Diagnosis in C…
Mixture-of-Experts (MoE) models offer computational efficiency during inference by activating only a subset of specialized experts for a given input. This enables efficient model scaling on multi-GPU systems that use expert parallelism…
Soft real-time applications require timely delivery of messages conforming to the soft real-time constraints. Satisfying such requirements is a complex task both due to the volatile nature of distributed environments, as well as due to…
Graph Neural Networks (GNNs) have gained significant momentum recently due to their capability to learn on unstructured graph data. Dynamic GNNs (DGNNs) are the current state-of-the-art for point cloud applications; such applications (viz.…
The cache plays a key role in determining the performance of applications, no matter for sequential or concurrent programs on homogeneous and heterogeneous architecture. Fixing cache misses requires to understand the origin and the type of…
Kubernetes multi-cluster deployments demand scalable and privacy-preserving anomaly detection. Existing eBPF-based monitors provide low-overhead system and network visibility but are limited to single clusters, while centralized approaches…
Cloud systems have rapidly expanded worldwide in the last decade, shifting computational tasks to cloud servers where clients submit their requests. Among cloud workloads, latency-critical applications -- characterized by high-percentile…
Cloud GPU servers have become the de facto way for deep learning practitioners to train complex models on large-scale datasets. However, it is challenging to determine the appropriate cluster configuration---e.g., server type and…
Computing on graphics processors is maybe one of the most important developments in computational science to happen in decades. Not since the arrival of the Beowulf cluster, which combined open source software with commodity hardware to…
Sustaining a large fraction of single GPU performance in parallel computations is considered to be the major problem of GPU-based clusters. In this article, this topic is addressed in the context of a lattice Boltzmann flow solver that is…
Large Language Model (LLM) inference is widely used in interactive assistants and agentic systems. In latency-sensitive deployments, inference time can become dominated by host-side overheads. Existing approaches typically expose this cost…
Graphics Processing Units (GPUs) have become an integral part of High-Performance Computing to achieve an Exascale performance. The main goal of application developers of GPU is to tune their code extensively to obtain optimal performance,…
Increase in the number of antennas in the front-end increases the volume of data to be processed at the back-end. This establishes a need for acceleration in back-end processing. To solve the issue of high volume data processing at…
All computing infrastructure suffers from performance variability, be it bare-metal or virtualized. This phenomenon originates from many sources: some transient, such as noisy neighbors, and others more permanent but sudden, such as changes…
Graphics processing units (GPUs) are gaining widespread use in computational chemistry and other scientific simulation contexts because of their huge performance advantages relative to conventional CPUs. However, the reliability of GPUs in…
In LHC Run 3, ALICE will increase the data taking rate significantly to 50 kHz continuous read out of minimum bias Pb-Pb collisions. The reconstruction strategy of the online offline computing upgrade foresees a first synchronous online…
A lot of deep learning applications are desired to be run on mobile devices. Both accuracy and inference time are meaningful for a lot of them. While the number of FLOPs is usually used as a proxy for neural network latency, it may be not…
In LHC Run 3, ALICE will increase the data taking rate significantly to 50 kHz continuous read-out of minimum bias Pb-Pb collisions. The reconstruction strategy of the online-offline computing upgrade foresees a first synchronous online…
The optimization of submodular functions constitutes a viable way to perform clustering. Strong approximation guarantees and feasible optimization w.r.t. streaming data make this clustering approach favorable. Technically, submodular…
Cloud performance diagnosis and prediction is a challenging problem due to the stochastic nature of the cloud systems. Cloud performance is affected by a large set of factors including (but not limited to) virtual machine types, regions,…
Hardware peripherals such as GPUs and FPGAs are commonly available in server-grade computing to accelerate specific compute tasks, from database queries to machine learning. CSPs have integrated these accelerators into their infrastructure…