Architecture of OzSearch

Architecture_OZ_Search

Status Quo of BeCurio Technology

Core Technology

  1. Search Engine indexing Structure and Algorithm
  2. String Management related data structure and algorithm
  3. Memory / resource efficiency enhancement library
  4. Big Data in Memory Related technology (based on Search Engine Technology)
statusQuo-180x185
statusQuo_900x445

Index Structure of OzSearch

Through Shared Index structure, saving by more than 50% and
guaranteeing memory based search for big data.
Index_Structure-_OzSearch

Shared index Structure

  • Small index DB →Decrease OP load → Speedy search
  • The bigger the data size is, the more the resource can be saved
Shared-index-Structure

Operation Algorithm and inverted file

Operation-Algorithm-and-inverted-file

Memory data reduction technology

Memory (data type simplification + byte + slimed data ST) – I/O{usage frequency grouping + data reduction}
Memory-data-reduction-technology

Memory Reduction: Data Structure and Capacity

Memory-Reduction

Inverted file size & capacity comparison

Inverted-file-size

Comparison test with the same data set

Test Procedure

  1. Data download
  2. Not bi-gram but full n-gram indexing without compression
  3. Comparison factors
    • Indexing time, index size, search speed
    • Number of searched items
    • Retrieval of the target item

Data Source Wikipedia

Data size : 10,000,000 (50GB text)
Avg. Record length : 400 bytes
Data format : Title
Comparison-test-with-the-same-data-set_button

Comparison Table with Solr

Comparison-test-with-the-same-data-set