Related papers: Replicating Data Pipelines with GrimoireLab
Large-scale language models have become increasingly challenging and expensive to train. Among various methods addressing this issue, Pipeline Parallelism has been widely employed to accommodate massive model weights within limited GPU…
Training large deep learning models at scale is very challenging. This paper proposes Chimera, a novel pipeline parallelism scheme which combines bidirectional pipelines for efficiently training large-scale models. Chimera is a synchronous…
Recent literature highlights the potential of graph-based approaches within large language model (LLM) retrieval-augmented generation (RAG) pipelines for answering queries of varying complexity, particularly those that fall outside the…
Resolution of complex post-production issues in large-scale open-source software (OSS) projects requires significant cognitive effort, as developers need to go through long, unstructured and fragmented issue discussion threads before that.…
River is a machine learning library for dynamic data streams and continual learning. It provides multiple state-of-the-art learning methods, data generators/transformers, performance metrics and evaluators for different stream learning…
As the particle physics community needs higher and higher precisions in order to test our current model of the subatomic world, larger and larger datasets are necessary. With upgrades scheduled for the detectors of colliding-beam…
Biosciences have been revolutionized by next generation sequencing (NGS) technologies in last years, leading to new perspectives in medical, industrial and environmental applications. And although our motivation comes from biosciences, the…
Most of the space projects or large observatories do have official tools like simulators, end-to-end pipelines developed during years by a large team of contributors. They are like {\em cathedrals}. In this paper, we show that very…
We design and implement a ready-to-use library in PyTorch for performing micro-batch pipeline parallelism with checkpointing proposed by GPipe (Huang et al., 2019). In particular, we develop a set of design components to enable…
Data is a valuable asset, and sharing it as a product across organizations is key to building comprehensive and useful insights in fields such as science and industry. Before sharing, data often requires transformation to comply with…
Microscopy is a primary source of information on materials structure and functionality at nanometer and atomic scales. The data generated is often well-structured, enriched with metadata and sample histories, though not always consistent in…
Pipeline parallelism is one of the key components for large-scale distributed training, yet its efficiency suffers from pipeline bubbles which were deemed inevitable. In this work, we introduce a scheduling strategy that, to our knowledge,…
GitHub has become a popular social application platform, where a large number of users post their open source projects. In particular, an increasing number of researchers release repositories of source code related to their research papers…
Management and analysis of big data are systematically associated with a data distributed architecture in the Hadoop and now Spark frameworks. This article offers an introduction for statisticians to these technologies by comparing the…
This paper introduces a lightweight and auditable method for asynchronous information exchange between distributed entities using Git as the coordination medium. The proposed approach replaces traditional APIs and message brokers with a…
We develop a tool called PipeGen for efficient data transfer between database management systems (DBMSs). PipeGen targets data analytics workloads on shared-nothing engines. It supports scenarios where users seek to perform different parts…
Machine learning inference pipelines commonly encountered in data science and industries often require real-time responsiveness due to their user-facing nature. However, meeting this requirement becomes particularly challenging when certain…
Finite mixture models are frequently used to uncover latent structures in high-dimensional datasets (e.g.\ identifying clusters of patients in electronic health records). The inference of such structures can be performed in a Bayesian…
We present GHTraffic, a dataset of significant size comprising HTTP transactions extracted from GitHub data and augmented with synthetic transaction data. The dataset facilitates reproducible research on many aspects of service-oriented…
Discovering atom-level phenomena requires molecular dynamics (MD) simulations with ab initio accuracy. Machine learning interatomic potentials (MLIPs) enable stable, high-accuracy MD simulations, and their models exhibit scaling-law trends…