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

Was this helpful?

  1. Infrastructure
  2. Vast Data Platform

Vast Database

The VAST DataBase is an all-flash transactional and analytical system designed for the AI era.

It simplifies data engineering by eliminating tiers of data management and combining the transactional properties of a relational database with the schema of a data warehouse.

The DataBase breaks the tradeoffs between row-based OLTP databases and columnar-based analytical queries, offering the scale and affordability of a data lake and the capabilities of a data lakehouse.

The key concepts and insights are as follows:

Semantic Layer

The VAST database serves as a critical component of the VAST data platform by providing a semantic or contextual layer.

It enables the application of semantic understanding to the data in both transactional and analytical contexts, allowing for real-time data processing and analysis.

What is the semantic layer?

The semantic layer is a critical component in tabular databases that acts as an intermediary between end-users and the underlying database structure.

It provides a simplified and user-friendly view of the data, enabling users to interact with it in a more meaningful and intuitive manner.

The semantic layer translates complex database structures into familiar concepts such as tables, columns, and relationships, abstracting away the underlying complexity.

Definition and Function: The semantic layer serves several key functions

Data Interpretation: It acts as a translator, converting technical database terminology into business-friendly terms. This allows users to understand and analyse data without needing deep technical expertise.

Relationship Definition: The semantic layer defines relationships between tables, making it easier for users to navigate and explore related data.

Security and Governance: It controls access to data based on user roles and permissions, ensuring data privacy and security.

Interplay with Tabular Databases

Tabular databases organise data into rows and columns, similar to a spreadsheet. While efficient for storage and retrieval, working directly with tabular databases can be challenging for end-users due to complex schemas and numerous tables.

The semantic layer enhances tabular databases by:

  1. Simplifying the data view, making it more accessible to non-technical users.

  2. Enabling the creation of calculated fields, aggregates, and hierarchical structures for more sophisticated analysis.

  3. Providing a centralised repository for data definitions and business rules, ensuring consistency and accuracy.

Benefits: Implementing a semantic layer brings several key benefits

  1. Improved Data Accessibility: Users can access and interact with data independently, without relying on IT or database professionals.

  2. Enhanced Data Security: The semantic layer enforces access control, auditing, and monitoring to safeguard sensitive data.

  3. Increased Efficiency: By centralizing data definitions and rules, the semantic layer promotes consistency, reduces errors, and streamlines data management.

Challenges: Implementing a semantic layer also presents some challenges

  1. Technical Issues: Ensuring optimal performance, scalability, and seamless integration with existing systems can be complex.

  2. Data Integrity: Maintaining accuracy and consistency between the semantic layer and the underlying database requires robust synchronisation and validation mechanisms.

  3. Change Management: Adopting a semantic layer may require significant planning, training, and support to overcome resistance and ensure smooth transition.

Future Trends: As technology advances, the semantic layer is expected to evolve

  1. Natural Language Processing and Machine Learning: These technologies will enhance query understanding and enable more intuitive user interactions.

  2. Intelligent Insights: By analysing user behavior, the semantic layer will proactively provide relevant recommendations and insights.

  3. Advanced Analytics: The semantic layer will facilitate the application of machine learning and predictive modeling, unlocking new opportunities for data-driven decision-making.

In conclusion, the semantic layer is a powerful tool that revolutionises data management and analysis in tabular databases.

By providing a user-friendly interface, enhancing security, and enabling advanced analytics, it empowers organisations to harness the full potential of their data and drive innovation in a data-driven future.

Design Principles

The VAST database is built on three key design principles: embracing standards (e.g., SQL, Apache Arrow), simplifying data management by eliminating complex tiers, and enabling deployment flexibility across public and private clouds.

Transactional and Analytical Capabilities

VAST challenges the traditional separation between transactional databases (OLTP) and analytical databases (OLAP).

Typically, OLTP systems are optimised for fast, small transactions, while OLAP systems are designed for large-scale data analysis.

VAST's aim is to create a unified system that can efficiently handle both types of workloads. This convergence is a significant shift from the norm, where separate systems handle transactional and analytical needs.

The result is that the VAST database combines transactional and analytical capabilities into a single system. It acts as a transactional system for real-time data ingestion and cataloguing while also providing analytical capabilities for quick understanding and querying of the data within the system.

Simplifying Data Engineering

The VAST database aims to simplify data engineering by eliminating constraints associated with managing semi-structured data objects in data science contexts.

Overcoming Limitations of Traditional Systems

Traditional database management systems often consist of separate systems for online transactional processing (OLTP) and online analytical processing (OLAP).

The VAST database addresses the limitations of these systems by providing a scalable solution that combines both transactional and analytical capabilities.

Disaggregated and Shared Everything (DASE) Architecture

The VAST database is built on top of the DASE architecture, which abstracts the cluster architecture and creates a data centre-scale computer.

All logic runs in stateless Linux containers, with data presented in parallel across a high-speed, low-latency data centre fabric. This architecture eliminates the need for machine coordination in the read/write path, which is a common bottleneck in traditional systems.

This architecture allows for high transactional services and scalability in terms of capacity and performance.

Write Buffer and Data Transformation: The VAST database uses a write buffer in storage class memory to absorb incoming data and provide time for data manipulation before storing it in low-cost flash storage. During this process, data is transformed from standard database record form into a columnar data format optimised for analytics.

Columnar Object and Metadata Store: The VAST database introduces a new style of columnar object that is significantly smaller than standard parquet row groups. These columnar objects are organised using a metadata store that enables efficient querying and data access.

SQL and Query Engine Integration: The VAST database supports native SQL querying and integrates with popular query engines like Apache Spark, Trino, and Dremio through push-down plugins.

Scalability and Performance: The VAST database is designed to scale linearly in terms of performance by adding CPUs, GPUs, and SSDs to the system. It can achieve high transactional throughput and sustained streaming performance for queries.

Similarity-Based Data Reduction: The VAST database employs a novel data reduction technique called similarity, which acts as a modern form of global compression. It identifies similar blocks of data across the entire data store and compresses them together, resulting in significant data reduction.

Cost Efficiency: By leveraging low-cost flash storage and efficient data reduction techniques, the VAST database offers a cost-effective solution compared to other flash-based data platforms.

Comprehensive Data Science Support: The VAST database provides a unified system that supports both structured data (databases and data warehouses) and unstructured data (file and object formats) commonly used in deep learning pipelines.

In summary, the VAST database introduces a transformative approach to database management by combining transactional and analytical capabilities, leveraging the DASE architecture, and employing innovative data reduction techniques.

It simplifies data engineering, enables real-time processing and analysis, and provides a scalable and cost-effective solution for modern data science and deep learning workloads.

PreviousVast DatastoreNextVast Data Engine

Last updated 10 months ago

Was this helpful?

Page cover image