About Zilliz Cloud

Zilliz Cloud is a fully managed service on cloud for LF AI Milvus®, you can set up your Milvus cluster to unlock high-performance similarity search with no extra effort needed for infrastructure management.

Why should you choose Zilliz Cloud?

Today’s databases aren’t designed to work with vector embeddings effectively, while Milvus is born for vector embeddings, it can scale for billions of vector embeddings.

Zilliz Cloud provides an out-of-box Milvus experience based on fully managed cloud service, freeing you from the headache of database maintenance.

Zilliz Cloud can deliver surpassingly high performance of over 1000 queries per second (QPS) with very low latency for vector search.

You can choose any size and class of databases and pay for what you have actually used only.

In full consideration of your data security, Zilliz Cloud complies with the SOC 2 standard and will provide Role-Based Access Control (RBAC) for your database soon.

What are the key features of Zilliz Cloud?

  • Core Milvus® features
    • High-performance vector search
    • Low latency
    • Built-in data types filtering
    • Intelligent index auto-optimization
    • Widely adopted metrics, such as Euclidean Distance (L2) and inner product (IP)
  • Compute Units (CUs) optimized for high-performance and big-data search
  • Deployed on AWS US West, and soon on AWS US East, Google Cloud Platform, and Microsoft Azure
  • Scale as your need
  • Multiple projects and users
  • Multi-factor authentication
  • Enhanced security features
  • Enterprise support

Where can vector database help?

  • Image similarity search: Images made searchable and instantaneously return the most similar images from a massive database.
  • Video similarity search: By converting keyframes into vectors and then feeding the results into Milvus, billions of videos can be searched and recommended in near real time.
  • Audio similarity search: Quickly query massive volumes of audio data such as speech, music, sound effects, and surface similar sounds.
  • Molecular similarity search: Blazing fast similarity search, substructure search, or superstructure search for a specified molecule.
  • Recommender system: Recommend information or products based on user behaviors and needs.
  • Question answering system: Interactive digital QA chatbot that automatically answers user questions.
  • DNA sequence classification: Accurately sort out the classification of a gene in milliseconds by comparing similar DNA sequences.
  • Text search engine: Help users find the information they are looking for by comparing keywords against a database of texts.
  • Anomaly detection: Identify data points, events, and/or observations that deviate from a dataset’s normal behavior.

Zilliz Cloud vs Milvus Community edition

Deployment

Zilliz Cloud vector databaseCommunity edition
Choice of cloud providerAWS, GCP (coming soon), Azure (coming soon)No
Choice of deployment scaleMultiple deployment sizes and classesSelf-managed
StorageInfinite data storageSelf-managed
Upgrade and bug fixesEffortlessSelf-managed
Availability99.9%No guarantee

Database features

Zilliz Cloud Vector databaseCommunity edition
Vector similarity searchYesYes
Scalar data types
  • INT
  • FLOAT
  • BOOLEAN
  • VARCHAR
  • INT
  • FLOAT
  • BOOLEAN
  • VARCHAR
ANNS index
  • AUTOINDEX optimized for high-performance searches
  • AUTOINDEX optimized for big-data searches
  • FLAT
  • IVF_FLAT
  • IVF_SQ8
  • IVF_PQ
  • HNSW
  • ANNOY
Vector similarity metrics
  • Euclidean distance (L2)
  • Inner product (IP)
  • Euclidean distance (L2)
  • Inner product (IP)
  • Hamming
  • Jaccard
  • Tanimoto
  • Superstructure
  • Substructure
Hybrid searchYesYes
Time travelYesYes
Multiple usersYesYes
ClientsPython and Java SDK (fully tested)Community maintained SDKs

Administrative tools

Zilliz Cloud Vector databaseCommunity edition
GUICloud UIAttu, CLI

Security control

Zilliz Cloud Vector databaseCommunity edition
Data encryption in transitYesYes

System limits

Zilliz Cloud Vector databaseCommunity edition
Max connection65,53665,536
Max collection65,53665,536
Max vector dimensions32,76832,768
Max topK16,38416,384
Max nq16,38416,384

Request limit

Zilliz Cloud Vector databaseCommunity edition
Insert size per request32MB512 MB
Search size per request32MB512 MB
Query size per request32MB512 MB

Support & services

Zilliz Cloud Vector databaseCommunity edition
Slack and EmailYesYes
Community forumYesYes
Priority supportSLANo