Home/ Questions/Q 7520219
In Process

The Archive Base Latest Questions

Editorial Team
  • 0
Asked: 2026-05-30T02:01:48+00:00

I’m going to be storing some specification data as JSON strings (height, weight etc)

  • 0

I’m going to be storing some specification data as JSON strings (height, weight etc) — one per product. The data will be accessed very infrequently (once initially and then only again if I change an algorithm and need to update the summary metrics). My plan is to store a reference to the JSON string’s location in a MySQL database, along with the rest of the product data.

Each blob is around 6k and, assuming I’m going to store 10m of them, my tests indicate that the size of a MySQL table with these 10m rows would be ~70GB (including an id column with an index). Storing them in S3 would work out slightly cheaper but it’s low in both cases.

Are there any disadvantage to putting them in MySQL. Would the one very large table have an impact on the performance of the rest of the server?

Share

Leave an answer

You must login to add an answer.

Need An Account,

1 Answer

  1. Editorial Team

    I recommend, you

    • Compress them: This is the ideal scenario to trade stoage space against decompression CPU cycles. JSON typically is highly compressible, say 2:1 as a minimum going up to 10:1
    • Store them directly in a separate MyISAM table: This way it’s guaranteed not to interfere with your other tables
    • 0