RediSearch Memory Footprint Comparisons

Blog

https://joeywhelan.blogspot.com/2023/02/redis-hash-vs-json-memory-footprints.html

Summary

Tool for comparing the Redis object and index memory footprints for various scenarios.

Architecture

Features

• Calculates memory footprint of Redis objects and search indices for various configurable scenarios

Prerequisites

• Python
• Redis installation with Search + JSON modules

Installation

1. Clone this repo.

2. Go to doc-memory-footprint folder.

cd doc-memory-footprint

3. Install Python requirements (either in a virtual env or global)

pip install -r requirements.txt

Usage

Options

• --url. Redis connection string. Default = redis://localhost:6379
• --nkeys. Number of keys to be generated for each test. Default = 10,000.
• --nfields. Number fields created for each key. Default = 10.
• --textsize. Number of random characters in each text or tag field. Default = 10.
• --numericsize. Number of random digits in each numeric field. Default = 10.
• --format. Output format for the table of consolidated results. Options are text, html or markdown. Default = text.

Execution

python3 mptest.py --nkeys 1000000 --nfields 15 --textsize 10 --numericsize 10 --format markdown

Output

Sample output for the test above.

Consolidated Results - Num Keys:1000000, Num Fields:15, Text Field Size:10, Numeric Field Size:10

Index Structure	Object Size(b)	Index Size(mb)
Hash Text Unsorted	269	199.15
Hash Text Sorted	269	637.84
Hash Tag Unsorted	269	133.34
Hash Tag Sorted	269	572.03
Hash Numeric Unsorted	245	192.3
Hash Numeric Sorted	245	535.62
JSON Text Unsorted	290	199.13
JSON Text Sorted	290	637.82
JSON Tag Unsorted	290	133.32
JSON Tag Sorted	290	572.02
JSON Numeric Unsorted	190	192.3
JSON Numeric Sorted	190	535.62