feat: add OceanBase UI config settings (#691)

Signed-off-by: min.tian <min.tian.cn@gmail.com>

Author: alwayslove2013

vectordb_bench/backend/clients/oceanbase/config.py

@@ -1,6 +1,6 @@
 from typing import TypedDict
 
-from pydantic import BaseModel, SecretStr, validator
+from pydantic import BaseModel, SecretStr
 
 from ..api import DBCaseConfig, DBConfig, IndexType, MetricType
 
@@ -31,14 +31,6 @@ def to_dict(self) -> OceanBaseConfigDict:
             "database": self.database,
         }
 
-    @validator("*")
-    def not_empty_field(cls, v: any, field: any):
-        if field.name in ["password", "host", "db_label"]:
-            return v
-        if isinstance(v, str | SecretStr) and len(v) == 0:
-            raise ValueError("Empty string!")
-        return v
-
 
 class OceanBaseIndexConfig(BaseModel):
     index: IndexType
@@ -83,7 +75,7 @@ def search_param(self) -> dict:
 
 
 class OceanBaseIVFConfig(OceanBaseIndexConfig, DBCaseConfig):
-    m: int
+    m: int | None = None
     sample_per_nlist: int
     nbits: int | None = None
     nlist: int

vectordb_bench/frontend/components/run_test/generateTasks.py

@@ -6,12 +6,17 @@ def generate_tasks(activedDbList: list[DB], dbConfigs, activedCaseList: list[Cas
     tasks = []
     for db in activedDbList:
         for case in activedCaseList:
+            cfg = {key.value: value for key, value in allCaseConfigs[db][case].items()}
+            # Many DBCaseConfig models require an `index` field, while the UI stores the selection under `IndexType`.
+            # Passing both keeps backwards-compatibility (extra fields are ignored) and enables strict models (e.g. OceanBase).
+            if CaseConfigParamType.IndexType in allCaseConfigs[db][case] and "index" not in cfg:
+                cfg["index"] = allCaseConfigs[db][case][CaseConfigParamType.IndexType]
             task = TaskConfig(
                 db=db.value,
                 db_config=dbConfigs[db],
                 case_config=case,
                 db_case_config=db.case_config_cls(allCaseConfigs[db][case].get(CaseConfigParamType.IndexType, None))(
-                    **{key.value: value for key, value in allCaseConfigs[db][case].items()}
+                    **cfg
                 ),
             )
             tasks.append(task)

vectordb_bench/frontend/config/dbCaseConfigs.py

@@ -423,6 +423,22 @@ class CaseConfigInput(BaseModel):
     },
 )
 
+CaseConfigParamInput_IndexType_OceanBase = CaseConfigInput(
+    label=CaseConfigParamType.IndexType,
+    inputType=InputType.Option,
+    inputHelp="Select OceanBase index type",
+    inputConfig={
+        "options": [
+            IndexType.HNSW.value,
+            IndexType.HNSW_SQ.value,
+            IndexType.HNSW_BQ.value,
+            IndexType.IVFFlat.value,
+            IndexType.IVFSQ8.value,
+            IndexType.IVFPQ.value,
+        ],
+    },
+)
+
 CaseConfigParamInput_IndexType_PgDiskANN = CaseConfigInput(
     label=CaseConfigParamType.IndexType,
     inputHelp="Select Index Type",
@@ -685,6 +701,20 @@ class CaseConfigInput(BaseModel):
     isDisplayed=lambda config: config.get(CaseConfigParamType.IndexType, None) == IndexType.HNSW.value,
 )
 
+CaseConfigParamInput_m_OceanBase = CaseConfigInput(
+    label=CaseConfigParamType.m,
+    displayLabel="m",
+    inputHelp="HNSW graph degree (m) for OceanBase HNSW/HNSW_SQ/HNSW_BQ",
+    inputType=InputType.Number,
+    inputConfig={
+        "min": 4,
+        "max": 128,
+        "value": 16,
+    },
+    isDisplayed=lambda config: config.get(CaseConfigParamType.IndexType, None)
+    in [IndexType.HNSW.value, IndexType.HNSW_SQ.value, IndexType.HNSW_BQ.value],
+)
+
 
 CaseConfigParamInput_EFConstruction_Milvus = CaseConfigInput(
     label=CaseConfigParamType.EFConstruction,
@@ -703,6 +733,20 @@ class CaseConfigInput(BaseModel):
     ],
 )
 
+CaseConfigParamInput_EFConstruction_OceanBase = CaseConfigInput(
+    label=CaseConfigParamType.EFConstruction,
+    displayLabel="efConstruction",
+    inputHelp="HNSW efConstruction for OceanBase HNSW/HNSW_SQ/HNSW_BQ",
+    inputType=InputType.Number,
+    inputConfig={
+        "min": 8,
+        "max": 65535,
+        "value": 256,
+    },
+    isDisplayed=lambda config: config.get(CaseConfigParamType.IndexType, None)
+    in [IndexType.HNSW.value, IndexType.HNSW_SQ.value, IndexType.HNSW_BQ.value],
+)
+
 CaseConfigParamInput_SQType = CaseConfigInput(
     label=CaseConfigParamType.sq_type,
     inputType=InputType.Option,
@@ -862,6 +906,48 @@ class CaseConfigInput(BaseModel):
     isDisplayed=lambda config: config.get(CaseConfigParamType.IndexType, None) == IndexType.HNSW.value,
 )
 
+CaseConfigParamInput_ef_search_OceanBase = CaseConfigInput(
+    label=CaseConfigParamType.ef_search,
+    displayLabel="ef_search",
+    inputHelp="HNSW ef_search (session var ob_hnsw_ef_search) for OceanBase",
+    inputType=InputType.Number,
+    inputConfig={
+        "min": 1,
+        "max": 65535,
+        "value": 100,
+    },
+    isDisplayed=lambda config: config.get(CaseConfigParamType.IndexType, None)
+    in [IndexType.HNSW.value, IndexType.HNSW_SQ.value, IndexType.HNSW_BQ.value],
+)
+
+CaseConfigParamInput_sample_per_nlist_OceanBase = CaseConfigInput(
+    label=CaseConfigParamType.sample_per_nlist,
+    displayLabel="sample_per_nlist",
+    inputHelp="OceanBase IVF training sample multiplier (total samples = sample_per_nlist * nlist)",
+    inputType=InputType.Number,
+    inputConfig={
+        "min": 1,
+        "max": 1000000,
+        "value": 256,
+    },
+    isDisplayed=lambda config: config.get(CaseConfigParamType.IndexType, None)
+    in [IndexType.IVFFlat.value, IndexType.IVFSQ8.value, IndexType.IVFPQ.value],
+)
+
+CaseConfigParamInput_ivf_nprobes_OceanBase = CaseConfigInput(
+    label=CaseConfigParamType.ivf_nprobes,
+    displayLabel="ivf_nprobes",
+    inputHelp="OceanBase IVF search probes (session var ob_ivf_nprobes)",
+    inputType=InputType.Number,
+    inputConfig={
+        "min": 1,
+        "max": 65535,
+        "value": 10,
+    },
+    isDisplayed=lambda config: config.get(CaseConfigParamType.IndexType, None)
+    in [IndexType.IVFFlat.value, IndexType.IVFSQ8.value, IndexType.IVFPQ.value],
+)
+
 CaseConfigParamInput_EFConstruction_PgVector = CaseConfigInput(
     label=CaseConfigParamType.ef_construction,
     inputType=InputType.Number,
@@ -2302,6 +2388,19 @@ class CaseConfigInput(BaseModel):
     CaseConfigParamInput_VectorSearchBeamSize_CockroachDB,
 ]
 
+OceanBaseLoadConfig = [
+    CaseConfigParamInput_IndexType_OceanBase,
+    CaseConfigParamInput_m_OceanBase,
+    CaseConfigParamInput_EFConstruction_OceanBase,
+    CaseConfigParamInput_ef_search_OceanBase,
+    CaseConfigParamInput_Nlist,
+    CaseConfigParamInput_sample_per_nlist_OceanBase,
+    CaseConfigParamInput_Nbits_PQ,
+    CaseConfigParamInput_M_PQ,
+    CaseConfigParamInput_ivf_nprobes_OceanBase,
+]
+OceanBasePerformanceConfig = OceanBaseLoadConfig
+
 MariaDBLoadingConfig = [
     CaseConfigParamInput_IndexType_MariaDB,
     CaseConfigParamInput_StorageEngine_MariaDB,
@@ -2616,6 +2715,10 @@ class CaseConfigInput(BaseModel):
         CaseLabel.Load: CockroachDBLoadingConfig,
         CaseLabel.Performance: CockroachDBPerformanceConfig,
     },
+    DB.OceanBase: {
+        CaseLabel.Load: OceanBaseLoadConfig,
+        CaseLabel.Performance: OceanBasePerformanceConfig,
+    },
 }
 
 

vectordb_bench/models.py

@@ -137,6 +137,10 @@ class CaseConfigParamType(Enum):
     confidence_interval = "confidence_interval"
     clip = "clip"
 
+    # OceanBase IVF parameters
+    sample_per_nlist = "sample_per_nlist"
+    ivf_nprobes = "ivf_nprobes"
+
     # CockroachDB parameters
     min_partition_size = "min_partition_size"
     max_partition_size = "max_partition_size"