Merge branch 'branch-0.5.0' into 'branch-0.5.0'

# Conflicts:
#   cpp/src/core/unittest/test_ivf.cpp
#   cpp/unittest/scheduler/test_resource.cpp

Former-commit-id: 92f072d94703b39712b5a61c03cbd144bfadfcd9

ci/jenkinsfile/dev_test.groovy

@@ -2,10 +2,9 @@ timeout(time: 30, unit: 'MINUTES') {
     try {
         dir ("${PROJECT_NAME}_test") {
             checkout([$class: 'GitSCM', branches: [[name: "${SEMVER}"]], doGenerateSubmoduleConfigurations: false, extensions: [], submoduleCfg: [], userRemoteConfigs: [[credentialsId: "${params.GIT_USER}", url: "git@192.168.1.105:Test/milvus_test.git", name: 'origin', refspec: "+refs/heads/${SEMVER}:refs/remotes/origin/${SEMVER}"]]])
-            sh 'python3 -m pip install -r requirements.txt'
+            sh 'python3 -m pip install -r requirements.txt -i http://pypi.douban.com/simple --trusted-host pypi.douban.com'
             sh "pytest . --alluredir=\"test_out/dev/single/sqlite\" --level=1 --ip ${env.JOB_NAME}-${env.BUILD_NUMBER}-milvus-gpu-engine.milvus-1.svc.cluster.local"
         }
-
         // mysql database backend test
         load "${env.WORKSPACE}/ci/jenkinsfile/cleanup_dev.groovy"
 

ci/jenkinsfile/dev_test_all.groovy

@@ -2,7 +2,7 @@ timeout(time: 60, unit: 'MINUTES') {
     try {
         dir ("${PROJECT_NAME}_test") {
             checkout([$class: 'GitSCM', branches: [[name: "${SEMVER}"]], doGenerateSubmoduleConfigurations: false, extensions: [], submoduleCfg: [], userRemoteConfigs: [[credentialsId: "${params.GIT_USER}", url: "git@192.168.1.105:Test/milvus_test.git", name: 'origin', refspec: "+refs/heads/${SEMVER}:refs/remotes/origin/${SEMVER}"]]])
-            sh 'python3 -m pip install -r requirements.txt'
+            sh 'python3 -m pip install -r requirements.txt -i http://pypi.douban.com/simple --trusted-host pypi.douban.com'
             sh "pytest . --alluredir=\"test_out/dev/single/sqlite\" --ip ${env.JOB_NAME}-${env.BUILD_NUMBER}-milvus-gpu-engine.milvus-1.svc.cluster.local"
         }
 

cpp/CHANGELOG.md

@@ -36,6 +36,7 @@ Please mark all change in change log and use the ticket from JIRA.
 - MS-619 - Add optimizer class in scheduler
 - MS-614 - Preload table at startup
 - MS-626 - Refactor DataObj to support cache any type data
+- MS-648 - Improve unittest
 
 ## New Feature
 - MS-627 - Integrate new index: IVFSQHybrid

cpp/cmake/ThirdPartyPackages.cmake

@@ -143,7 +143,7 @@ if(USE_JFROG_CACHE STREQUAL "ON")
     if(NOT DEFINED JFROG_ARTFACTORY_URL)
         message(FATAL_ERROR "JFROG_ARTFACTORY_URL is not set")
     endif()
-    set(JFROG_ARTFACTORY_CACHE_URL "${JFROG_ARTFACTORY_URL}/generic-local/milvus/thirdparty/cache/${CMAKE_OS_NAME}/${MILVUS_BUILD_ARCH}/${BUILD_TYPE}")
+    set(JFROG_ARTFACTORY_CACHE_URL "${JFROG_ARTFACTORY_URL}/milvus/thirdparty/cache/${CMAKE_OS_NAME}/${MILVUS_BUILD_ARCH}/${BUILD_TYPE}")
     if(DEFINED ENV{JFROG_USER_NAME})
         set(JFROG_USER_NAME "$ENV{JFROG_USER_NAME}")
     endif()

cpp/src/core/cmake/ThirdPartyPackagesCore.cmake

@@ -123,7 +123,7 @@ if(NOT DEFINED USE_JFROG_CACHE)
     set(USE_JFROG_CACHE "OFF")
 endif()
 if(USE_JFROG_CACHE STREQUAL "ON")
-    set(JFROG_ARTFACTORY_CACHE_URL "${JFROG_ARTFACTORY_URL}/generic-local/milvus/thirdparty/cache/${CMAKE_OS_NAME}/${KNOWHERE_BUILD_ARCH}/${BUILD_TYPE}")
+    set(JFROG_ARTFACTORY_CACHE_URL "${JFROG_ARTFACTORY_URL}/milvus/thirdparty/cache/${CMAKE_OS_NAME}/${KNOWHERE_BUILD_ARCH}/${BUILD_TYPE}")
     set(THIRDPARTY_PACKAGE_CACHE "${THIRDPARTY_DIR}/cache")
 endif()
 

cpp/src/core/knowhere/knowhere/index/vector_index/IndexGPUIVFSQ.cpp

@@ -71,4 +71,13 @@ GPUIVFSQ::CopyGpuToCpu(const Config& config) {
     return std::make_shared<IVFSQ>(new_index);
 }
 
+void
+GPUIVFSQ::search_impl(int64_t n, const float* data, int64_t k, float* distances, int64_t* labels, const Config& cfg) {
+#ifdef CUSTOMIZATION
+    GPUIVF::search_impl(n, data, k, distances, labels, cfg);
+#else
+    IVF::search_impl(n, data, k, distances, labels, cfg);
+#endif
+}
+
 }  // namespace knowhere

cpp/src/core/knowhere/knowhere/index/vector_index/IndexGPUIVFSQ.h

@@ -38,6 +38,10 @@ class GPUIVFSQ : public GPUIVF {
 
     VectorIndexPtr
     CopyGpuToCpu(const Config& config) override;
+
+ protected:
+    void
+    search_impl(int64_t n, const float* data, int64_t k, float* distances, int64_t* labels, const Config& cfg) override;
 };
 
 }  // namespace knowhere

cpp/src/core/unittest/test_ivf.cpp

@@ -154,8 +154,8 @@ class IVFTest : public DataGen, public TestWithParam<::std::tuple<std::string, P
 INSTANTIATE_TEST_CASE_P(IVFParameters, IVFTest,
                         Values(std::make_tuple("IVF", ParameterType::ivf),
                                std::make_tuple("GPUIVF", ParameterType::ivf),
-                               //                            std::make_tuple("IVFPQ", ParameterType::ivfpq),
-                               //                            std::make_tuple("GPUIVFPQ", ParameterType::ivfpq),
+                               std::make_tuple("IVFPQ", ParameterType::ivfpq),
+                               std::make_tuple("GPUIVFPQ", ParameterType::ivfpq),
                                std::make_tuple("IVFSQ", ParameterType::ivfsq),
 #ifdef CUSTOMIZATION
                                std::make_tuple("IVFSQHybrid", ParameterType::ivfsq),
@@ -240,6 +240,7 @@ TEST_P(IVFTest, hybrid) {
         auto result = hybrid_1_idx->Search(query_dataset, conf);
         AssertAnns(result, nq, conf->k);
         PrintResult(result, nq, k);
+        hybrid_1_idx->UnsetQuantizer();
     }
 
     {
@@ -438,6 +439,7 @@ TEST_P(IVFTest, clone_test) {
     }
 }
 
+#ifdef CUSTOMIZATION
 TEST_P(IVFTest, seal_test) {
     // FaissGpuResourceMgr::GetInstance().InitDevice(device_id);
 
@@ -472,6 +474,7 @@ TEST_P(IVFTest, seal_test) {
     auto with_seal = tc.RecordSection("With seal");
     ASSERT_GE(without_seal, with_seal);
 }
+#endif
 
 class GPURESTEST : public DataGen, public ::testing::Test {
  protected:
@@ -637,7 +640,7 @@ TEST_F(GPURESTEST, copyandsearch) {
     // search and copy at the same time
     printf("==================\n");
 
-    index_type = "GPUIVFSQ";
+    index_type = "GPUIVF";
     index_ = IndexFactory(index_type);
 
     auto conf = std::make_shared<knowhere::IVFSQCfg>();
@@ -699,7 +702,7 @@ TEST_F(GPURESTEST, copyandsearch) {
 }
 
 TEST_F(GPURESTEST, TrainAndSearch) {
-    index_type = "GPUIVFSQ";
+    index_type = "GPUIVF";
     index_ = IndexFactory(index_type);
 
     auto conf = std::make_shared<knowhere::IVFSQCfg>();

cpp/src/core/unittest/test_kdt.cpp

@@ -36,6 +36,7 @@ class KDTTest : public DataGen, public ::testing::Test {
  protected:
     void
     SetUp() override {
+        Generate(96, 1000, 10);
         index_ = std::make_shared<knowhere::CPUKDTRNG>();
 
         auto tempconf = std::make_shared<knowhere::KDTCfg>();

cpp/src/core/unittest/test_nsg/test_nsg.cpp

@@ -38,17 +38,17 @@ class NSGInterfaceTest : public DataGen, public ::testing::Test {
     SetUp() override {
         // Init_with_default();
         knowhere::FaissGpuResourceMgr::GetInstance().InitDevice(DEVICE_ID, 1024 * 1024 * 200, 1024 * 1024 * 600, 2);
-        Generate(256, 1000000, 1);
+        Generate(256, 1000000 / 100, 1);
         index_ = std::make_shared<knowhere::NSG>();
 
         auto tmp_conf = std::make_shared<knowhere::NSGCfg>();
         tmp_conf->gpu_id = DEVICE_ID;
-        tmp_conf->knng = 100;
-        tmp_conf->nprobe = 32;
-        tmp_conf->nlist = 16384;
-        tmp_conf->search_length = 60;
-        tmp_conf->out_degree = 70;
-        tmp_conf->candidate_pool_size = 500;
+        tmp_conf->knng = 20;
+        tmp_conf->nprobe = 8;
+        tmp_conf->nlist = 163;
+        tmp_conf->search_length = 40;
+        tmp_conf->out_degree = 30;
+        tmp_conf->candidate_pool_size = 100;
         tmp_conf->metric_type = knowhere::METRICTYPE::L2;
         train_conf = tmp_conf;
 

cpp/unittest/db/test_db.cpp

@@ -297,6 +297,7 @@ TEST_F(DBTest, SEARCH_TEST) {
         ASSERT_TRUE(stat.ok());
     }
 
+#ifdef CUSTOMIZATION
     //test FAISS_IVFSQ8H optimizer
     index.engine_type_ = (int)milvus::engine::EngineType::FAISS_IVFSQ8H;
     db_->CreateIndex(TABLE_NAME, index); // wait until build index finish
@@ -314,9 +315,7 @@ TEST_F(DBTest, SEARCH_TEST) {
         stat = db_->Query(TABLE_NAME, file_ids, k, nq, 10, xq.data(), dates, results);
         ASSERT_TRUE(stat.ok());
     }
-
-
-    // TODO(lxj): add groundTruth assert
+#endif
 }
 
 TEST_F(DBTest, PRELOADTABLE_TEST) {

cpp/unittest/scheduler/test_resource.cpp

@@ -184,7 +184,7 @@ class ResourceAdvanceTest : public testing::Test {
 };
 
 TEST_F(ResourceAdvanceTest, DISK_RESOURCE_TEST) {
-    const uint64_t NUM = 2;
+    const uint64_t NUM = max_once_load;
     std::vector<std::shared_ptr<TestTask>> tasks;
     TableFileSchemaPtr dummy = nullptr;
     for (uint64_t i = 0; i < NUM; ++i) {

cpp/unittest/wrapper/test_wrapper.cpp

@@ -188,7 +188,7 @@ INSTANTIATE_TEST_CASE_P(WrapperParam, KnowhereWrapperTest,
                                             10,
                                             10),
                             std::make_tuple(milvus::engine::IndexType::FAISS_IVFSQ8_CPU, "Default", DIM, NB, 10, 10),
-                            std::make_tuple(milvus::engine::IndexType::FAISS_IVFSQ8_GPU, "Default", DIM, NB, 10, 10),
+//                            std::make_tuple(milvus::engine::IndexType::FAISS_IVFSQ8_GPU, "Default", DIM, NB, 10, 10),
                             std::make_tuple(milvus::engine::IndexType::FAISS_IVFSQ8_MIX, "Default", DIM, NB, 10, 10),
 //                            std::make_tuple(IndexType::NSG_MIX, "Default", 128, 250000, 10, 10),
 //                            std::make_tuple(IndexType::SPTAG_KDT_RNT_CPU, "Default", 128, 250000, 10, 10),