)]}'
{
  "commit": "8698ad6bf55d3968033cdacbe8894030cd923731",
  "tree": "c9ef579f4fa5b09233e21d3dcf907458f057ee0d",
  "parents": [
    "7f44a9d89fff117b4455ad7332007697422f2d1a"
  ],
  "author": {
    "name": "Yongqiang YANG",
    "email": "dataroaring@gmail.com",
    "time": "Thu Mar 12 18:40:11 2026 -0700"
  },
  "committer": {
    "name": "Yongqiang YANG",
    "email": "dataroaring@gmail.com",
    "time": "Thu Mar 12 18:40:11 2026 -0700"
  },
  "message": "[opt](memory) Intern cluster ID strings in CloudReplica\n\nAdd a static intern pool for cluster ID strings to eliminate millions\nof duplicate String instances across CloudReplica objects. Each replica\nstores cluster IDs as map keys; without interning, Gson deserialization\ncreates a separate String instance per replica (~40-70 bytes each).\n\n- Add ConcurrentHashMap-based intern pool (heap-safe, unlike String.intern)\n- Intern strings at all write points (updateClusterToPrimaryBe/Secondary)\n- Intern keys during gsonPostProcess for deserialized maps\n- For 1M replicas with 3 clusters: saves ~40-70 MB of duplicate Strings\n\nCo-Authored-By: Claude Opus 4.6 \u003cnoreply@anthropic.com\u003e\n",
  "tree_diff": [
    {
      "type": "modify",
      "old_id": "b5f119cc803ef38b469939dc3b99bc49cbdef49a",
      "old_mode": 33188,
      "old_path": "fe/fe-core/src/main/java/org/apache/doris/cloud/catalog/CloudReplica.java",
      "new_id": "6948ef2f49c45b060f3dc2d95db8c2e97c99dd91",
      "new_mode": 33188,
      "new_path": "fe/fe-core/src/main/java/org/apache/doris/cloud/catalog/CloudReplica.java"
    }
  ]
}