LogoLogo
API ReferenceGitHubSlackService StatusLogin
v3.8.16
v3.8.16
  • Deep Lake Docs
  • Vector Store Quickstart
  • Deep Learning Quickstart
  • Storage & Credentials
    • Storage Options
    • User Authentication
    • Storing Deep Lake Data in Your Own Cloud
      • Microsoft Azure
        • Provisioning Federated Credentials
        • Enabling CORS
      • Amazon Web Services
        • Provisioning Role-Based Access
        • Enabling CORS
  • List of ML Datasets
  • 🏢High-Performance Features
    • Introduction
    • Performant Dataloader
    • Tensor Query Language (TQL)
      • TQL Syntax
      • Sampling Datasets
    • Deep Memory
      • How it Works
    • Index for ANN Search
      • Caching and Optimization
    • Managed Tensor Database
      • REST API
      • Migrating Datasets to the Tensor Database
  • 📚EXAMPLE CODE
    • Getting Started
      • Vector Store
        • Step 1: Hello World
        • Step 2: Creating Deep Lake Vector Stores
        • Step 3: Performing Search in Vector Stores
        • Step 4: Customizing Vector Stores
      • Deep Learning
        • Step 1: Hello World
        • Step 2: Creating Deep Lake Datasets
        • Step 3: Understanding Compression
        • Step 4: Accessing and Updating Data
        • Step 5: Visualizing Datasets
        • Step 6: Using Activeloop Storage
        • Step 7: Connecting Deep Lake Datasets to ML Frameworks
        • Step 8: Parallel Computing
        • Step 9: Dataset Version Control
        • Step 10: Dataset Filtering
    • Tutorials (w Colab)
      • Vector Store Tutorials
        • Vector Search Options
          • Deep Lake Vector Store API
          • REST API
          • LangChain API
        • Image Similarity Search
        • Deep Lake Vector Store in LangChain
        • Deep Lake Vector Store in LlamaIndex
        • Improving Search Accuracy using Deep Memory
      • Deep Learning Tutorials
        • Creating Datasets
          • Creating Complex Datasets
          • Creating Object Detection Datasets
          • Creating Time-Series Datasets
          • Creating Datasets with Sequences
          • Creating Video Datasets
        • Training Models
          • Splitting Datasets for Training
          • Training an Image Classification Model in PyTorch
          • Training Models Using MMDetection
          • Training Models Using PyTorch Lightning
          • Training on AWS SageMaker
          • Training an Object Detection and Segmentation Model in PyTorch
        • Updating Datasets
        • Data Processing Using Parallel Computing
      • Concurrent Writes
        • Concurrency Using Zookeeper Locks
    • Playbooks
      • Querying, Training and Editing Datasets with Data Lineage
      • Evaluating Model Performance
      • Training Reproducibility Using Deep Lake and Weights & Biases
      • Working with Videos
    • Low-Level API Summary
  • 🔬Technical Details
    • Best Practices
      • Creating Datasets at Scale
      • Training Models at Scale
      • Storage Synchronization and "with" Context
      • Restoring Corrupted Datasets
      • Concurrent Writes
    • Data Layout
    • Version Control and Querying
    • Dataset Visualization
    • Tensor Relationships
    • Visualizer Integration
    • Shuffling in dataloaders
    • How to Contribute
Powered by GitBook
On this page
  • How to Extract Maximum Performance from Your Vector Search
  • Tuning the Index Parameters
  • Caching of Embeddings and Index
  • Caching of Other Tensors

Was this helpful?

  1. High-Performance Features
  2. Index for ANN Search

Caching and Optimization

Understanding Caching to Increase Query Performance in Deep Lake

PreviousIndex for ANN SearchNextManaged Tensor Database

Was this helpful?

How to Extract Maximum Performance from Your Vector Search

Tuning the Index Parameters

The parameters of the HSNW index can be tuned using the index_params shown below:

vectorstore = VectorStore(path, 
                          index_params = {"threshold": <DEFAULT_-1>, # Index turned off by default
                                          "distance_metric":<DEFAULT_COS>,
                                          "additional_params": {
                                              "efConstruction": <DEFAULT_200>,
                                              "M": <DEFAULT_16>}})

Further information about the impact of the index parameters .

Caching of Embeddings and Index

Either of the following operations caches the embeddings are on disk and the index in RAM:

  • The index is created

  • The first vector search is executed after the Vector Store is loaded

Since the first query caches critical information, subsequent queries will execute much faster compared to the first query. Since the cache is invalidated after the Vector Store is loaded or initialized, the optimal access pattern is not to re-load the Vector Store each search, unless you believe it was updated by another client.

The embeddings are cached on disk in the following locations:

Mac: /tmp/....

Linux: /var/folders/

Caching of Other Tensors

Tensors containing other information such as text and metadata are also cached in memory when they are used in queries. As a result, the first query that utilized this data will be slowest, with subsequent queries running much faster.

If the data size exceeds the cache size, it will be re-fetched with every query, thus reducing query performance. The default cache size is 2 MB, and you may increase the cache size using the parameter below:

vectorstore = VectorStore(path, memory_cache_size = <cache_in_MB))

🏢
can be found here