Go53 DNS Data Storage Design

1. In-Memory Data Structure

• Each DNS zone (e.g. "example.com.") is a key in a map.
• Each zone value is a map of record types ("A", "MX", "SOA", "NS", etc).
• Each record type is a map (or array) of record names to their data struct.

Go structure:

map[string]map[string]map[string]any
// zone      // type      // record name -> struct

Example (JSON):

{
  "A": {
    "www.example.com.": { "name": "www.example.com.", "ip": "192.0.2.5", "ttl": 3600 }
  },
  "SOA": {
    "example.com.": {
      "name": "example.com.",
      "ns": "ns1.example.com.",
      "mbox": "admin.example.com.",
      "serial": 2024060101,
      "refresh": 3600,
      "retry": 600,
      "expire": 86400,
      "minttl": 3600,
      "ttl": 3600
    }
  },
  "MX": {
    "example.com.": {
      "name": "example.com.",
      "preference": 10,
      "mx": "mx1.example.com.",
      "ttl": 3600
    }
  },
  "NS": {
    "example.com.": [
      { "name": "example.com.", "ns": "ns1.example.com.", "ttl": 3600 },
      { "name": "example.com.", "ns": "ns2.example.com.", "ttl": 3600 }
    ]
  }
}

2. Backend: BadgerDB Key-Value Storage

• Key: zone name (e.g., "example.com.")
• Value: full JSON serialization of the zone’s records as above

Write Flow

• On record add/delete/update:
  1. Update in-memory cache[zone]
  2. Marshal cache[zone] as JSON
  3. db.Set(zoneName, jsonBlob)

Read Flow

• At startup or reload:
  1. For each key (zone) in BadgerDB:
  2. Read value
  3. Unmarshal JSON into cache[zone]

3. Loading/Saving Data

• In-memory: Data is loaded into the cache map on startup or after modification.
• BadgerDB: Zones are persisted as full JSON blobs keyed by zone name.

4. Summary

• In memory:
  • map[zone] = map[type] = map[name] = record
  • JSON serializable; easy to persist/load
• BadgerDB:
  • Key = zone name, Value = full zone record JSON