Cloud Imperium games replaces Elasticsearch for deeper player insights. Check out the case study! -->
Cloud Imperium games replaces Elasticsearch for deeper player insights. Check out the case study! -->
Start Free Trial

Chaos LakeDB

Built from the ground up to transform your cloud storage into a Live Search+SQL+GenAI Analytics Database

See it in Action

Unified Live Data Lake

Live operational and business use cases all leverage the same type of data - now you can collapse them into a single analytic database!

Observability and Security Analytics without retention limits at a fraction of the cost, plus User Insights in real-time across the organization, all enhanced with a natural language assistant powered by GenAI.

Product White Paper

Product White Paper

Get the White Paper

THIRD GENERATION DATABASE

Transform cloud storage into a live analytics database with a stateless architecture for multi-modal analytics via open APIs

UNIFIED

INTEGRATED

FORMAT

Same on Disk and in Memory

Format

STRUCTURE

Same for Search and Relational

Structure

ALGORITHMS

Generic Execution Across Access

Algorithms

ARCHITECTURE

Stateless and Auto‑Scalable

Architecture
Technical Whitepaper CTA

 

A Single Analytical Database for Live Operational and Business Use Cases

OPERATIONAL

BUSINESS

Chaos Console
OpenSearch Dashboards
Superset
Assistant
External Tools
Your App
Start with Information Science NOT specific use
Best storage service already exists... make it the analytical backend
Make it easily accessible across use cases
More than a data representation. A New Database!

Chaos Index®

Generic data representation for search and relational analytics on cloud storage

Compression through Representation

Split information into 3 different dimensions to reduce size and entropy

Data Representation

Symbols

Efficient Symbol Encoding (Text-Search)

Manifest

Efficient Scope Routing (Query Planner)

Locality

Efficient Locality of Reference (Relational)

Multiplicity of Small

Multiplicity of Small
Fast 5-20X Compression (size of GZIP)
Small Storage
Small Processing
Small Network

Performant Search+SQL

Performant Search+SQL
Unified Efficient Format
Symbols for Efficient Search (e.g. like Lucene)
Locality for Efficient Relational (e.g. like Parquet)

Optimized for Live Data

Organize data into streams of segments organized by groups, accessible by views

Auto detect and dynamically map schema and tokenize all data

Object Group (OG)

Data Segments

Live Analytics

Live Analytics
Minute Time to Glass
Seconds Query Resolution
SQL
Across Search & SQL for first or subsequent queries

No Schema Management

No Schema Management
Auto Detection and Dynamic Mapping
Ability to Apply Schema
Schema
Ability to handle flexible formats, the likes of nested fields such as JSON - "a set it and forget it ingestion model"

Cloud Storage as Backend, not Storage

Segments purposefully designed for fast performance straight from cloud storage

Same format and structure on disk as in memory

Cloud Storage Backend

Best of Cloud Storage

Best of Cloud Storage
$0.02/GB for Unlimited Retention
11 9s Durability
Cloud Storage
All the benefits of cloud storage, now as a live analytics backend

Enables Stateless

Enables Stateless
Same Format In Mem and On Disk
Fast First Read Across Access Patterns
Stateless
Enables the Chaos Fabric to be stateless and auto-scalable purely based on ingestion and querying needs

Chaos Fabric®

Serverless and Stateless Fabric

Compute used only to ingest and query data, not data persistence / cache

Serverless and Stateless Fabric

No Cost-Retention Trade-off

Cost

Compute used just for processing, not persistence and all data in cloud storage

Fast First Read

Fast Read

All queries read from cloud storage reducing first read latency penalty

Generic Compute Workers

Generic containerized workers can do any ingestion and query task

Dedicated Chaos Managed VPC

Optimized Utilization for Cost-Performance

Utilization

Any worker can be used and leased for any ingestion or query task

Scale without bottlenecks

Scale

Workers auto age-out to keep them always fresh
Workers auto-scale based on overall worker needs (Chaos Farm)

Distributed Stream Ingestion and Querying

Distributed stream ingestion Distributed query execution

Distributed Stream Ingestion and Querying

Live Ingestion
at Scale

Ingestion

Efficient live data ingestion that can auto-scale to handle any spikes

Search+SQL Analytics via Virtual Views

Analytics

Integrated Generic Query Engine Chaos Index delivers efficient Search+SQL with Virtual Views for Democratized Data Access


Chaos Farm

The Final Piece of Third Generation Database Leveraging the Best of The Cloud

How to optimize a stateless fabric's cost-performance for live analytics at scale?

Distributed Work Operating System

Orchestration and Scheduling of Distributed Work for Ingest and Query task

Distributed OS

Best-in-class Cost-Performance

Cost Performance

Distributed worker allocation for improved utilization leveraging generic workers

Improved
Experience

Improved Experience

Worker allocation prioritized to improve customer experience

Auto-Scalable
Compute

Auto-Scalable Compute Worker Pool Based on Policies and Utilization

Auto-Scalable

Leverage the best of stateless

Stateless

Ability to scale compute up in high usage periods and down in low usage ones

Intelligent Auto-Scale for Improved Cost

Auto-Scale

Ability to schedule scaling based on time / activity or dynamically based worker utilization

Centralized Farm to Support Multiple VPCs

Centralized Farm can hold compute for multiple VPCs (customers) without sharing

Centralized Farm

New Cross-VPC Compute Movement

Compute Movement

Ability to move compute across VPCs allows for centralization of compute that can securely be accessed by multiple VPCs

Reduced Time to Compute

Compute Time

Ability to access compute from another VPC allows for latency of seconds to scale rather than minutes from provider


Reduced Time, Cost & Complexity

Real-Time Analytics & Full Historical Context

  • Minute time-to-glass; Seconds query resolution
  • Auto-schema detection & dynamic mapping for easy setup & live data use cases
  • Unlimited retention without  rehydration needs

Unmatched Cost-Performance at Scale

  • Data only in cloud storage
  • Chaos Index® is 5-20x smaller than raw
  • Small data = Small compute
  • Stateless = Compute just for ingest & query, not store

Unified Live Search+ SQL+GenAI Analytics

  • Single platform across operational & business use cases
  • All data stored in customers' cloud storage with granular RBAC
  • No sharding, partitioning, schema management including of nested data
  • Auto-scaling & seamless upgrades
With ChaosSearch, we no longer have to move or transform our data. That means we don’t have to think about how we’re going to use the data before we access it. It’s just there for us to query. That helps us deliver new opportunities to the business, or discover usage patterns in our systems.
Jeff Kinsherf, SVP, Eng Services and SRE
Equifax Logo
See More Customer Stories