Running Open Large Language Models in Production with Ollama and Serverless GPUs
Explore the deployment of open large language models in production environments using Ollama and serverless GPUs. Learn why companies are increasingly interested in running open models like Gemma and Llama, which offer full control over deployment, model…
One Billion Row Challenge: Optimizing Java for Massive Data Processing
Explore efficient techniques for processing massive CSV files in Java, including parallelization, optimized parsing, and custom data structures. Learn from the One Billion Row Challenge.
Sustainable Development of Stencil-based HPC Applications - JuliaCon 2024
Explore sustainable HPC application development using Julia packages for stencil-based computations, achieving scalable performance and portability across diverse hardware architectures.
NeuroTree - A Differentiable Tree Operator for Tabular Data
Explore NeuroTree, a differentiable tree operator for tabular data, addressing traditional tree limitations through simultaneous node learning and built-in ensemble techniques.
Introducing DMRjulia v1.0 - Tensor Network Computations in Julia
Explore DMRjulia v1.0, a tensor network library for quantum systems modeling. Learn design principles, hardware optimization, and cutting-edge research applications in this comprehensive introduction.
Training Llama 2 in Julia - Scaling Large Language Models
Explore scaling techniques for training large neural networks like Llama2 using Julia, including multi-GPU parallelization and fine-tuning methods for efficient model adaptation.
Data Versioning in Generative AI: A Pathway to Cost-Effective ML
Explore data versioning strategies for Generative AI, focusing on managing large-scale unstructured data and optimizing resource usage in model scoring and API interactions.
Training LLMs: Lessons from the Trenches
Discover key insights and challenges in training large language models across hundreds of GPUs, from data handling to resource management.
Creating the World's Premier Biological Foundation Model
Explore Recursion's groundbreaking biological foundation model, revolutionizing drug discovery through advanced infrastructure, data management, and large-scale training techniques.
Supercharging Recommender Systems - Unleashing the Power of Distributed Model Training
Explore distributed model training techniques to enhance recommender systems, boosting performance and scalability for production environments.
From Idea to Production: AI Infrastructure for Scaling LLM Applications
Explore strategies for scaling LLM applications from beta to production, addressing challenges and building adaptable AI infrastructure for evolving models and workflows.
Data Versioning in Generative AI - A Pathway to Cost-effective Machine Learning
Explore data versioning strategies for generative AI, focusing on cost-effective techniques to minimize processing time and API calls while enhancing collaboration and efficient versioning of annotations and embeddings.
Running ML Inference Services in Shared Hosting Environments
Optimize ML inference in shared hosting: Learn to boost latency, throughput, and resource utilization through queue management and OpenMP tuning.
Scale and Accelerate Distributed Model Training in Kubernetes Clusters
Orchestrate and accelerate distributed deep learning workloads across multiple GPUs and nodes using Kubernetes, Kubeflow PytorchJob, RDMA, RoCE, and SRIOV for near-linear performance scaling.
MLOps is Just HPC in Disguise - A Real-World, No-Nonsense Guide to Upgrading Your Workflow
Upgrade your ML workflow with efficient tools and techniques for automating tasks, managing resources, and streamlining production deployment in this practical guide to MLOps.
Critical Use of MLOps in Finance Using Cloud-Managed ML Services That Scale
Leverage MLOps to improve quality and consistency of machine learning solutions in finance. Accelerate model experimentation, deployment, and tracking using cloud-managed services for efficient scaling.
WarpDrive: Orders of Magnitude Faster Multi-Agent Deep Reinforcement Learning on a GPU
Accelerate multi-agent deep reinforcement learning with WarpDrive, an open-source framework enabling end-to-end GPU execution. Explore features like automatic GPU tuning and distributed training for improved efficiency and scalability.
Best Practices in HPC and HTC Environments
Discover best practices for optimizing performance in High Performance and High Throughput Computing environments.
File Systems and Efficient Data Handling - HPC Café
Explore file systems and efficient data handling in HPC environments, focusing on NHR@FAU's offerings, properties, usage scenarios, and best practices for optimal performance.
Faster Model Serving with Ray and Anyscale - Ray Summit 2024
Explore Ray Serve's capabilities for distributed model serving and deployment, focusing on Anyscale's solutions for large-scale AI models and the challenges of building AI applications in the era of generative AI.
Unlock the Power of Parallel Computing With SWAR (SIMD Within A Register)
Unlock parallel computing power with SWAR technique. Explore SIMD-like operations using conventional CPU registers, boosting performance for hash tables, string functions, and more. Dive into benchmarks and real-world applications.