Skip to content

Open Source alternative to Algolia + Pinecone and an Easier-to-Use alternative to ElasticSearch ⚡ 🔍 ✨ Fast, typo tolerant, in-memory fuzzy Search Engine for building delightful search experiences

Throttle collection loading to prevent CPU saturation.

kishorenc0577df5e91bd
diff --git a/include/collection_manager.h b/include/collection_manager.h
index 7a5be35c..8ba4dcd8 100644
--- a/include/collection_manager.h
+++ b/include/collection_manager.h
@@ -32,10 +32,12 @@ private:
static constexpr const char* COLLECTION_NUM_MEMORY_SHARDS = "num_memory_shards";
std::string bootstrap_auth_key;
- std::string bootstrap_search_only_auth_key;
float max_memory_ratio;
+ /// During load, sleep `LOAD_THROTTLE_PERCENT`% of time taken for indexing to prevent CPU saturation
+ float LOAD_THROTTLE_PERCENT = 10;
+
CollectionManager();
~CollectionManager() = default;
diff --git a/src/auth_manager.cpp b/src/auth_manager.cpp
index af860ed3..3df010b2 100644
--- a/src/auth_manager.cpp
+++ b/src/auth_manager.cpp
@@ -4,6 +4,7 @@
constexpr const char* AuthManager::DOCUMENTS_SEARCH_ACTION;
Option<bool> AuthManager::init(Store *store) {
+ // This function must be idempotent, i.e. when called multiple times, must produce the same state without leaks
LOG(INFO) << "AuthManager::init()";
this->store = store;
diff --git a/src/collection_manager.cpp b/src/collection_manager.cpp
index ca73391d..e548659f 100644
--- a/src/collection_manager.cpp
+++ b/src/collection_manager.cpp
@@ -67,6 +67,7 @@ void CollectionManager::init(Store *store,
}
Option<bool> CollectionManager::load(const size_t init_batch_size) {
+ // This function must be idempotent, i.e. when called multiple times, must produce the same state without leaks
LOG(INFO) << "CollectionManager::load()";
Option<bool> auth_init_op = auth_manager.init(store);
@@ -119,6 +120,12 @@ Option<bool> CollectionManager::load(const size_t init_batch_size) {
uint32_t collection_next_seq_id = next_seq_id_status == StoreStatus::NOT_FOUND ? 0 :
StringUtils::deserialize_uint32_t(collection_next_seq_id_str);
+ if(get_collection(this_collection_name)) {
+ // To maintain idempotency, if the collection already exists in-memory, drop it from memory
+ LOG(WARNING) << "Dropping duplicate collection " << this_collection_name << " before loading it again.";
+ drop_collection(this_collection_name, false);
+ }
+
Collection* collection = init_collection(collection_meta, collection_next_seq_id);
LOG(INFO) << "Loading collection " << collection->get_name();
@@ -146,9 +153,14 @@ Option<bool> CollectionManager::load(const size_t init_batch_size) {
iter_batch.push_back(std::vector<index_record>());
}
- int num_docs_read = 0;
+ size_t num_found_docs = 0;
+ size_t num_valid_docs = 0;
+ size_t num_indexed_docs = 0;
+
+ auto begin = std::chrono::high_resolution_clock::now();
while(iter->Valid() && iter->key().starts_with(seq_id_prefix)) {
+ num_found_docs++;
const uint32_t seq_id = Collection::get_seq_id_from_key(iter->key().ToString());
nlohmann::json document;
@@ -160,8 +172,7 @@ Option<bool> CollectionManager::load(const size_t init_batch_size) {
return Option<bool>(false, "Bad JSON.");
}
- num_docs_read++;
-
+ num_valid_docs++;
iter_batch[seq_id % collection->get_num_memory_shards()].emplace_back(index_record(0, seq_id, document));
// Peek and check for last record right here so that we handle batched indexing correctly
@@ -169,7 +180,10 @@ Option<bool> CollectionManager::load(const size_t init_batch_size) {
iter->Next();
bool last_record = !(iter->Valid() && iter->key().starts_with(seq_id_prefix));
- if((num_docs_read % init_batch_size == 0) || last_record) {
+ // batch must match atleast the number of shards
+ const size_t batch_size = std::max(init_batch_size, collection->get_num_memory_shards());
+
+ if((num_valid_docs % batch_size == 0) || last_record) {
std::vector<size_t> indexed_counts;
indexed_counts.reserve(iter_batch.size());
@@ -186,12 +200,25 @@ Option<bool> CollectionManager::load(const size_t init_batch_size) {
}
}
iter_batch[i].clear();
+ num_indexed_docs += num_indexed;
}
}
+
+ auto time_millis = std::chrono::duration_cast<std::chrono::milliseconds>(
+ std::chrono::high_resolution_clock::now() - begin).count();
+
+ auto throttle_time_millis = uint64_t((LOAD_THROTTLE_PERCENT/100) * time_millis);
+
+ if(throttle_time_millis != 0) {
+ // we throttle only when we have accumulated atleast 1 ms worth of throttling time
+ begin = std::chrono::high_resolution_clock::now();
+ std::this_thread::sleep_for(std::chrono::milliseconds(throttle_time_millis));
+ }
}
add_to_collections(collection);
- LOG(INFO) << "Loaded " << num_docs_read << " documents into collection " << collection->get_name();
+ LOG(INFO) << "Indexed " << num_indexed_docs << "/" << num_found_docs
+ << " documents into collection " << collection->get_name();
}
std::string symlink_prefix_key = std::string(SYMLINK_PREFIX) + "_";
@@ -369,7 +396,6 @@ Option<bool> CollectionManager::drop_collection(const std::string& collection_na
collection_id_names.erase(collection->get_collection_id());
delete collection;
- collection = nullptr;
return Option<bool>(true);
}