LogoLogo
Continuum WebsiteContinuum ApplicationsContinuum KnowledgeAxolotl Platform
Continuum Knowledge
Continuum Knowledge
  • Continuum
  • Data
    • Datasets
      • Pre Training Data
      • Types of Fine Tuning
      • Self Instruct Paper
      • Self-Alignment with Instruction Backtranslation
      • Systematic Evaluation of Instruction-Tuned Large Language Models on Open Datasets
      • Instruction Tuning
      • Instruction Fine Tuning - Alpagasus
      • Less is More For Alignment
      • Enhanced Supervised Fine Tuning
      • Visualising Data using t-SNE
      • UMAP: Uniform Manifold Approximation and Projection for Dimension Reduction
      • Training and Evaluation Datasets
      • What is perplexity?
  • MODELS
    • Foundation Models
      • The leaderboard
      • Foundation Models
      • LLama 2 - Analysis
      • Analysis of Llama 3
      • Llama 3.1 series
      • Google Gemini 1.5
      • Platypus: Quick, Cheap, and Powerful Refinement of LLMs
      • Mixtral of Experts
      • Mixture-of-Agents (MoA)
      • Phi 1.5
        • Refining the Art of AI Training: A Deep Dive into Phi 1.5's Innovative Approach
      • Phi 2.0
      • Phi-3 Technical Report
  • Training
    • The Fine Tuning Process
      • Why fine tune?
        • Does Fine-Tuning LLMs on New Knowledge Encourage Hallucinations?
        • Explanations in Fine Tuning
      • Tokenization
        • Tokenization Is More Than Compression
        • Tokenization - SentencePiece
        • Tokenization explore
        • Tokenizer Choice For LLM Training: Negligible or Crucial?
        • Getting the most out of your tokenizer for pre-training and domain adaptation
        • TokenMonster
      • Parameter Efficient Fine Tuning
        • P-Tuning
          • The Power of Scale for Parameter-Efficient Prompt Tuning
        • Prefix-Tuning: Optimizing Continuous Prompts for Generation
        • Harnessing the Power of PEFT: A Smarter Approach to Fine-tuning Pre-trained Models
        • What is Low-Rank Adaptation (LoRA) - explained by the inventor
        • Low Rank Adaptation (Lora)
        • Practical Tips for Fine-tuning LMs Using LoRA (Low-Rank Adaptation)
        • QLORA: Efficient Finetuning of Quantized LLMs
        • Bits and Bytes
        • The Magic behind Qlora
        • Practical Guide to LoRA: Tips and Tricks for Effective Model Adaptation
        • The quantization constant
        • QLORA: Efficient Finetuning of Quantized Language Models
        • QLORA and Fine-Tuning of Quantized Language Models (LMs)
        • ReLoRA: High-Rank Training Through Low-Rank Updates
        • SLoRA: Federated Parameter Efficient Fine-Tuning of Language Models
        • GaLora: Memory-Efficient LLM Training by Gradient Low-Rank Projection
      • Hyperparameters
        • Batch Size
        • Padding Tokens
        • Mixed precision training
        • FP8 Formats for Deep Learning
        • Floating Point Numbers
        • Batch Size and Model loss
        • Batch Normalisation
        • Rethinking Learning Rate Tuning in the Era of Language Models
        • Sample Packing
        • Gradient accumulation
        • A process for choosing the learning rate
        • Learning Rate Scheduler
        • Checkpoints
        • A Survey on Efficient Training of Transformers
        • Sequence Length Warmup
        • Understanding Training vs. Evaluation Data Splits
        • Cross-entropy loss
        • Weight Decay
        • Optimiser
        • Caching
      • Training Processes
        • Extending the context window
        • PyTorch Fully Sharded Data Parallel (FSDP)
        • Train Short, Test Long: Attention with Linear Biases Enables Input Length Extrapolation
        • YaRN: Efficient Context Window Extension of Large Language Models
        • Sliding Window Attention
        • LongRoPE
        • Reinforcement Learning
        • An introduction to reinforcement learning
        • Reinforcement Learning from Human Feedback (RLHF)
        • Direct Preference Optimization: Your Language Model is Secretly a Reward Model
  • INFERENCE
    • Why is inference important?
      • Grouped Query Attention
      • Key Value Cache
      • Flash Attention
      • Flash Attention 2
      • StreamingLLM
      • Paged Attention and vLLM
      • TensorRT-LLM
      • Torchscript
      • NVIDIA L40S GPU
      • Triton Inference Server - Introduction
      • Triton Inference Server
      • FiDO: Fusion-in-Decoder optimised for stronger performance and faster inference
      • Is PUE a useful measure of data centre performance?
      • SLORA
  • KNOWLEDGE
    • Vector Databases
      • A Comprehensive Survey on Vector Databases
      • Vector database management systems: Fundamental concepts, use-cases, and current challenges
      • Using the Output Embedding to Improve Language Models
      • Decoding Sentence-BERT
      • ColBERT: Efficient and Effective Passage Search via Contextualized Late Interaction over BERT
      • SimCSE: Simple Contrastive Learning of Sentence Embeddings
      • Questions Are All You Need to Train a Dense Passage Retriever
      • Improving Text Embeddings with Large Language Models
      • Massive Text Embedding Benchmark
      • RocketQAv2: A Joint Training Method for Dense Passage Retrieval and Passage Re-ranking
      • LLM2Vec: Large Language Models Are Secretly Powerful Text Encoders
      • Embedding and Fine-Tuning in Neural Language Models
      • Embedding Model Construction
      • Demystifying Embedding Spaces using Large Language Models
      • Fine-Tuning Llama for Multi-Stage Text Retrieval
      • Large Language Model Based Text Augmentation Enhanced Personality Detection Model
      • One Embedder, Any Task: Instruction-Finetuned Text Embeddings
      • Vector Databases are not the only solution
      • Knowledge Graphs
        • Harnessing Knowledge Graphs to Elevate AI: A Technical Exploration
        • Unifying Large Language Models and Knowledge Graphs: A Roadmap
      • Approximate Nearest Neighbor (ANN)
      • High Dimensional Data
      • Principal Component Analysis (PCA)
      • Vector Similarity Search - HNSW
      • FAISS (Facebook AI Similarity Search)
      • Unsupervised Dense Retrievers
    • Retrieval Augmented Generation
      • Retrieval-Augmented Generation for Large Language Models: A Survey
      • Fine-Tuning or Retrieval?
      • Revolutionising Information Retrieval: The Power of RAG in Language Models
      • A Survey on Retrieval-Augmented Text Generation
      • REALM: Retrieval-Augmented Language Model Pre-Training
      • Retrieve Anything To Augment Large Language Models
      • Generate Rather Than Retrieve: Large Language Models Are Strong Context Generators
      • Active Retrieval Augmented Generation
      • DSPy: LM Assertions: Enhancing Language Model Pipelines with Computational Constraints
      • DSPy: Compiling Declarative Language Model Calls
      • DSPy: In-Context Learning for Extreme Multi-Label Classification
      • Optimizing Instructions and Demonstrations for Multi-Stage Language Model Programs
      • HYDE: Revolutionising Search with Hypothetical Document Embeddings
      • Enhancing Recommender Systems with Large Language Model Reasoning Graphs
      • Retrieval Augmented Generation (RAG) versus fine tuning
      • RAFT: Adapting Language Model to Domain Specific RAG
      • Summarisation Methods and RAG
      • Lessons Learned on LLM RAG Solutions
      • Stanford: Retrieval Augmented Language Models
      • Overview of RAG Approaches with Vector Databases
      • Mastering Chunking in Retrieval-Augmented Generation (RAG) Systems
    • Semantic Routing
    • Resource Description Framework (RDF)
  • AGENTS
    • What is agency?
      • Rephrase and Respond: Let Large Language Models Ask Better Questions for Themselves
      • Types of Agents
      • The risk of AI agency
      • Understanding Personality in Large Language Models: A New Frontier in AI Psychology
      • AI Agents - Reasoning, Planning, and Tool Calling
      • Personality and Brand
      • Agent Interaction via APIs
      • Bridging Minds and Machines: The Legacy of Newell, Shaw, and Simon
      • A Survey on Language Model based Autonomous Agents
      • Large Language Models as Agents
      • AI Reasoning: A Deep Dive into Chain-of-Thought Prompting
      • Enhancing AI Reasoning with Self-Taught Reasoner (STaR)
      • Exploring the Frontier of AI: The "Tree of Thoughts" Framework
      • Toolformer: Revolutionising Language Models with API Integration - An Analysis
      • TaskMatrix.AI: Bridging Foundational AI Models with Specialised Systems for Enhanced Task Completion
      • Unleashing the Power of LLMs in API Integration: The Rise of Gorilla
      • Andrew Ng's presentation on AI agents
      • Making AI accessible with Andrej Karpathy and Stephanie Zhan
  • Regulation and Ethics
    • Regulation and Ethics
      • Privacy
      • Detecting AI Generated content
      • Navigating the IP Maze in AI: The Convergence of Blockchain, Web 3.0, and LLMs
      • Adverse Reactions to generative AI
      • Navigating the Ethical Minefield: The Challenge of Security in Large Language Models
      • Navigating the Uncharted Waters: The Risks of Autonomous AI in Military Decision-Making
  • DISRUPTION
    • Data Architecture
      • What is a data pipeline?
      • What is Reverse ETL?
      • Unstructured Data and Generatve AI
      • Resource Description Framework (RDF)
      • Integrating generative AI with the Semantic Web
    • Search
      • BM25 - Search Engine Ranking Function
      • BERT as a reranking engine
      • BERT and Google
      • Generative Engine Optimisation (GEO)
      • Billion-scale similarity search with GPUs
      • FOLLOWIR: Evaluating and Teaching Information Retrieval Models to Follow Instructions
      • Neural Collaborative Filtering
      • Federated Neural Collaborative Filtering
      • Latent Space versus Embedding Space
      • Improving Text Embeddings with Large Language Models
    • Recommendation Engines
      • On Interpretation and Measurement of Soft Attributes for Recommendation
      • A Survey on Large Language Models for Recommendation
      • Model driven recommendation systems
      • Recommender AI Agent: Integrating Large Language Models for Interactive Recommendations
      • Foundation Models for Recommender Systems
      • Exploring the Impact of Large Language Models on Recommender Systems: An Extensive Review
      • AI driven recommendations - harming autonomy?
    • Logging
      • A Taxonomy of Anomalies in Log Data
      • Deeplog
      • LogBERT: Log Anomaly Detection via BERT
      • Experience Report: Deep Learning-based System Log Analysis for Anomaly Detection
      • Log-based Anomaly Detection with Deep Learning: How Far Are We?
      • Deep Learning for Anomaly Detection in Log Data: A Survey
      • LogGPT
      • Adaptive Semantic Gate Networks (ASGNet) for log-based anomaly diagnosis
  • Infrastructure
    • The modern data centre
      • Enhancing Data Centre Efficiency: Strategies to Improve PUE
      • TCO of NVIDIA GPUs and falling barriers to entry
      • Maximising GPU Utilisation with Kubernetes and NVIDIA GPU Operator
      • Data Centres
      • Liquid Cooling
    • Servers and Chips
      • The NVIDIA H100 GPU
      • NVIDIA H100 NVL
      • Lambda Hyperplane 8-H100
      • NVIDIA DGX Servers
      • NVIDIA DGX-2
      • NVIDIA DGX H-100 System
      • NVLink Switch
      • Tensor Cores
      • NVIDIA Grace Hopper Superchip
      • NVIDIA Grace CPU Superchip
      • NVIDIA GB200 NVL72
      • Hopper versus Blackwell
      • HGX: High-Performance GPU Platforms
      • ARM Chips
      • ARM versus x86
      • RISC versus CISC
      • Introduction to RISC-V
    • Networking and Connectivity
      • Infiniband versus Ethernet
      • NVIDIA Quantum InfiniBand
      • PCIe (Peripheral Component Interconnect Express)
      • NVIDIA ConnectX InfiniBand adapters
      • NVMe (Non-Volatile Memory Express)
      • NVMe over Fabrics (NVMe-oF)
      • NVIDIA Spectrum-X
      • NVIDIA GPUDirect
      • Evaluating Modern GPU Interconnect
      • Scalable Hierarchical Aggregation and Reduction Protocol (SHARP)
      • Next-generation networking in AI environments
      • NVIDIA Collective Communications Library (NCCL)
    • Data and Memory
      • NVIDIA BlueField Data Processing Units (DPUs)
      • Remote Direct Memory Access (RDMA)
      • High Bandwidth Memory (HBM3)
      • Flash Memory
      • Model Requirements
      • Calculating GPU memory for serving LLMs
      • Transformer training costs
      • GPU Performance Optimisation
    • Libraries and Complements
      • NVIDIA Base Command
      • NVIDIA AI Enterprise
      • CUDA - NVIDIA GTC 2024 presentation
      • RAPIDs
      • RAFT
    • Vast Data Platform
      • Vast Datastore
      • Vast Database
      • Vast Data Engine
      • DASE (Disaggregated and Shared Everything)
      • Dremio and VAST Data
    • Storage
      • WEKA: A High-Performance Storage Solution for AI Workloads
      • Introduction to NVIDIA GPUDirect Storage (GDS)
        • GDS cuFile API
      • NVIDIA Magnum IO GPUDirect Storage (GDS)
      • Vectors in Memory
Powered by GitBook
LogoLogo

Continuum - Accelerated Artificial Intelligence

  • Continuum Website
  • Axolotl Platform

Copyright Continuum Labs - 2023

On this page
  • InfiniBand Architecture
  • Definition - Switched Fabric Topology
  • Some worthwhile reading on InfiniBand
  • Interconnect for GPUs
  • Summary
  • Three applications for NVIDIA Quantum InfiniBand

Was this helpful?

  1. Infrastructure
  2. Networking and Connectivity

NVIDIA Quantum InfiniBand

Networking Solution

PreviousInfiniband versus EthernetNextPCIe (Peripheral Component Interconnect Express)

Last updated 11 months ago

Was this helpful?

NVIDIA Quantum InfiniBand is a high-performance networking solution designed for AI and high-performance computing (HPC) workloads in data centres.

It enables fast, low-latency communication between servers, storage systems, and NVIDIA GPUs.

InfiniBand Architecture

  • NVIDIA Quantum InfiniBand is based on the InfiniBand networking standard, which provides high bandwidth and low latency.

  • It uses a switched fabric topology, allowing multiple devices to communicate simultaneously without contention.

  • The latest generation, NVIDIA Quantum-2, offers speeds up to 400 Gb/s per port.

This speed of 400 Gb/s per port is incredibly fast.

At 400 Gb/s, you could transfer a 100 GB dataset in just 2 seconds. In one minute, you could transfer 3 TB of data, which is equivalent to the storage capacity of a high-end consumer desktop computer.

To put this speed into context, let's compare it with some common networking standards:

Gigabit Ethernet (GbE)

Gigabit Ethernet offers a maximum speed of 1 Gb/s per port. - NVIDIA Quantum InfiniBand's 400 Gb/s speed is 400 times faster than GbE.

10 Gigabit Ethernet (10GbE)

10 Gigabit Ethernet provides speeds up to 10 Gb/s per port. NVIDIA Quantum InfiniBand is 40 times faster than 10GbE.

PCI Express (PCIe) Gen 4

PCIe Gen 4 provides a bandwidth of up to 16 GT/s (GigaTransfers per second) per lane, with a x16 link offering a maximum theoretical bandwidth of 32 GB/s.

NVIDIA Quantum InfiniBand's 400 Gb/s speed is equivalent to around 50 GB/s, exceeding the bandwidth of a PCIe Gen 4 x16 link.

Definition - Switched Fabric Topology

Fabric topology in networking refers to the layout or structure of interconnected nodes, including switches, servers, and storage devices, within a network.

It is designed to support high levels of data transmission and communication efficiency. The term "fabric" comes from the idea of interweaving threads, symbolising the complex and interconnected nature of the network paths.

NVIDIA Quantum InfiniBand uses a switched fabric topology, which means it can easily be scaled up by adding more switches or nodes. It also means the network can ensure continued operation even if a component fails, which is critical for mission-critical applications in data centres.

Some worthwhile reading on InfiniBand

Interconnect for GPUs

NVIDIA Quantum InfiniBand is designed to work with NVIDIA GPUs

With GPUDirect RDMA (Remote Direct Memory Access), GPUs can bypass the CPU and directly access data from remote servers or storage systems over the InfiniBand network.

In-Network Computing

NVIDIA Quantum InfiniBand supports In-Network Computing, which offloads certain computations to the network fabric itself.

These engines accelerate collective operations, reduce network traffic, and improve overall application performance.

Performance Isolation

NVIDIA Quantum InfiniBand provides proactive monitoring and congestion management to ensure performance isolation.

It minimises performance jitter and guarantees predictable performance for applications, as if they were running on dedicated systems.

This is particularly important in multi-tenant environments where multiple users or applications share the same infrastructure.

Performance jitter refers to the variability in computational performance or network latency over time. Factors contributing to performance jitter may include fluctuating network traffic, shared system resources, or varying workloads.

Cloud-Native Supercomputing

It provides bare-metal performance, user management, data protection, and on-demand provisioning of HPC and AI services in a cloud environment.

This allows organisations to leverage the flexibility and scalability of the cloud while maintaining the performance characteristics of dedicated supercomputing systems.

Adapters and Switches

  • These adapters include advanced In-Network Computing capabilities and programmable engines for data preprocessing and offloading application control paths to the network.

  • NVIDIA Quantum-2 switches offer high-density, high-bandwidth switching with up to 64 400 Gb/s ports or 128 200 Gb/s ports in a compact 1U form factor.

Cables and Transceivers

  • NVIDIA Quantum InfiniBand supports a variety of connectivity options, including , , , and .

  • These options provide flexibility in building network topologies and enable backward compatibility with existing 200 Gb/s or 100 Gb/s infrastructures.

Summary

NVIDIA Quantum InfiniBand is a networking solution that offers extreme performance and efficiency for modern data centres.

As data centres continue to evolve and adopt GPU-accelerated computing and cloud-native architectures, NVIDIA Quantum InfiniBand will play a role in ensuring optimal system performance, scalability, and flexibility.

By investing in this technology, organisations can future-proof their data centre infrastructure and unlock new possibilities for innovation and discovery.

Three applications for NVIDIA Quantum InfiniBand

Real-time, high-resolution video processing in media and entertainment

NVIDIA Quantum InfiniBand could enable distributed, GPU-accelerated processing of high-resolution video content (e.g., 8K or higher) in real-time.

This would allow media and entertainment companies to collaborate on complex video editing, visual effects, and animation projects across multiple locations, with minimal latency and maximum performance.

Federated learning for healthcare and medical research

Quantum InfiniBand could facilitate secure, high-speed data sharing and model training across multiple healthcare institutions or research centres.

This would enable federated learning, where AI models are trained on decentralised data without compromising patient privacy. The low latency and high bandwidth of Quantum InfiniBand would ensure rapid model updates and faster discovery of new medical insights.

Real-time, GPU-accelerated intrusion detection and cybersecurity

NVIDIA Quantum InfiniBand could power distributed, GPU-accelerated intrusion detection systems (IDS) for large-scale networks.

By leveraging GPUs and high-speed, low-latency networking, these systems could analyse massive amounts of network traffic in real-time, detecting and responding to potential security threats with unprecedented speed and accuracy.

This would help organisations to better protect their critical assets and data from increasingly sophisticated cyber attacks.

It supports , a technology that allows GPUs to directly access the memory of other GPUs or network adapters, reducing latency and improving performance.

It includes preconfigured and programmable compute engines, such as , Message Passing Interface (MPI) Tag Matching, and MPI All-to-All.

NVIDIA Quantum InfiniBand, combined with , enables cloud-native supercomputing.

, available in various form factors, provide single or dual network ports at 400 Gb/s.

NVIDIA GPUDirect
NVIDIA Scalable Hierarchical Aggregation and Reduction Protocol (SHARPv3)
NVIDIA BlueField Data Processing Units (DPUs)
NVIDIA ConnectX-7 InfiniBand adapters
320KB
Introduction to Infiniband.pdf
pdf
From NVIDIA
Page cover image