Skip to content
A decentralized, trusted, high performance, SQL database with blockchain features
Go C C++ Java Python Assembly Other
Branch: develop
Clone or download

Latest commit

laodouya Merge pull request #369 from CovenantSQL/feature/gomod
Change go package tool from dep to go mod.
Latest commit 79a9b93 Sep 6, 2019


Type Name Latest commit message Commit time
Failed to load latest commit information.
.github Update stale.yml Apr 29, 2019
api Fix golint Jun 20, 2019
bin Fix golint Jun 20, 2019
blockproducer Use once reset instead of once assignmet in test case Jul 4, 2019
chainbus Add cloudflare key Dec 5, 2018
client Minor fix Aug 8, 2019
cmd Print DSN after bp accepted transaction. Aug 8, 2019
conf Add testnet w node_c config. Aug 6, 2019
consistent Fix Consistent.count May 13, 2019
crypto Resolve crypto/secp256k1 go-ethereum symbol conflict Jun 24, 2019
docker Fix missing gocovmerge binary. Sep 5, 2019
kayak Use goroutine specified rpc caller to avoid concurrent issue Jul 26, 2019
keys add cpu_count, loadavg metric. normarlize crucial metrics to "mem_ava… Aug 1, 2018
logo Add wechat qr code in readme (#351) May 28, 2019
merkle Format godoc Feb 25, 2019
metric Fix golint Jun 20, 2019
naconn Add tcp dial timeout May 30, 2019
pow/cpuminer Minor fix Apr 1, 2019
proto Refine metrics May 30, 2019
route Merge branch 'develop' into feature/adapter Aug 6, 2019
sqlchain Minor fix Aug 7, 2019
storage Fix tests for go-sqlite 3.28.0 will return table name and column name… Jun 19, 2019
test Add dstat in stable test, reduce concurrency count. Jul 30, 2019
twopc Fix ETLS permission test Apr 2, 2019
types Fix account type IsValid bug. May 23, 2019
utils Add utils once test case Jul 4, 2019
worker Use raw expvar.Int for dbcount Aug 7, 2019
xenomint Add convertQueryAndBuildArgs test for VIRTUAL TABLE and MATCH Jun 18, 2019
.dockerignore Add keystore in conf to docker ignore list Mar 20, 2019
.gitignore Remove make dep receipt. Sep 5, 2019
.gitlab-ci.yml Bump compatibility CI test version to 0.8.0. Aug 7, 2019
.gitmodules use readonly GNTE submodule Aug 2, 2018
.travis.yml Update docker/ci go version to 1.13 Sep 5, 2019 Changelog for v0.8.0 Aug 7, 2019
LICENSE update license to Apache2.0 May 5, 2018
Makefile Remove make dep receipt. Sep 5, 2019 Add wechat qr code in readme (#351) May 28, 2019 Fix typo May 31, 2019 Merge remote-tracking branch 'origin/develop' into feature/beta_upstream Mar 27, 2019 Update cleanup db script Mar 22, 2019
codecov.yml Combine delete path func. Mar 5, 2019
docker-compose.yml Rename docker observer to explorer. May 24, 2019
fmtGoDoc.sed Format comment Aug 29, 2018 Add imports format script and format all codes Feb 25, 2019 Keep HashStablePack latest before generate code Jan 21, 2019
go.mod Change go dep to go mod. Sep 5, 2019
go.sum Change go dep to go mod. Sep 5, 2019

Go Report Card Coverage Build Status License GoDoc homebrew


CovenantSQL(CQL) is a Byzantine Fault Tolerant relational database built on SQLite:

  • ServerLess: Free, High Availabile, Auto Sync Database Service for Serverless App
  • SQL: Most SQL-92 support.
  • Decentralize: Running on Open Internet without Central Coordination.
  • Privacy: Access with Granted Permission and Encryption Pass.
  • Immutable: Query History in CQL is Immutable and Trackable.
  • Permission: Column Level ACL and SQL Pattern Whitelist.

What is CQL?

  • Open source alternative of Amazon QLDB
  • Low cost DBaaS
  • Just like filecoin + IPFS is the decentralized file system, CQL is the decentralized database

Quick Start

CQL client supports:

  • macOS X 10.9+
  • Linux 2.6.23+ (x86, x86_64, armeabi-v7a, arm64-v8a)
Developer Guide


  • 🍺 Homebrew users can just run:

    brew install cql
  • non-Homebrew users can run:

    sudo bash -c 'curl -L "" | tar xzv -C /usr/local/bin/ --strip-components=1'


  • Just run:

    sudo bash -c 'curl -L "" | tar xzv -C /usr/local/bin/ --strip-components=1'

To continue: TestNet Quickstart


CovenantSQL Testnet is already released, have a try.

How CQL works

3 Layers Arch

CovenantSQL 3 Layer design

  • Layer 1: Global Consensus Layer (the main chain, the middle ring in the architecture diagram):
    • There will only be one main chain throughout the network.
    • Mainly responsible for database Miner and the user’s contract matching, transaction settlement, anti-cheating, shard chain lock hash and other global consensus matters.
  • Layer 2: SQL Consensus Layer (shard chain, rings on both sides):
    • Each database will have its own separate shard chain.
    • Mainly responsible for: the signature, delivery and consistency of the various Transactions of the database. The data history of the permanent traceability is mainly implemented here, and the hash lock is performed in the main chain.
  • Layer 3: Datastore Layer (database engine with SQL-92 support):
    • Each Database has its own independent distributed engine.
    • Mainly responsible for: database storage & encryption, query processing & signature, efficient indexing.
For more details

Consensus Algorithm

CQL supports 2 kinds of consensus algorithm:

  1. DPoS (Delegated Proof-of-Stake) is applied in Eventually consistency mode database and also Layer 1 (Global Consensus Layer) in BlockProducer. CQL miners pack all SQL queries and its signatures by the client into blocks thus form a blockchain. We named the algorithm Xenomint.
  2. BFT-Raft (Byzantine Fault-Toleranted Raft)bft-raft is applied in Strong consistency mode database. We named our implementation Kayak. The CQL miner leader does a Two-Phase Commit with Kayak to support Transaction.transaction

CQL database consistency mode and node count can be selected in database creation with command cql create '{"UseEventualConsistency": true, "Node": 3}'


Ethereum Hyperledger Fabric Amazon QLDB CovenantSQL
Dev language Solidity, ewasm Chaincode (Go, NodeJS) ? Python, Golang, Java, PHP, NodeJS, MatLab
Dev Pattern Smart Contract Chaincode SQL SQL
Open Source Y Y N Y
Nodes for HA 3 15 ? 3
Column Level ACL N Y ? Y
Data Format File Key-value Document Filefuse, Key-value, Structured
Storage Encryption N API Y Y
Data Desensitization N N N Y
Multi-tenant DIY DIY N Y
Throughput (1s delay) 15~10 tx/s 3500 tx/s ? 11065 tx/s (Eventually Consistency)
1866 tx/s (Strong Consistency)
Consistency Delay 2~6 min < 1 s ? < 10 ms
Secure for Open Internet Y N Only in AWS Y
Consensus PoW + PoS(Casper) CFT ? DPoS (Eventually Consistency)
BFT-Raft (Strong Consistency)


  • BFT-Raft: A CQL leader offline needs CQL Block Producer to decide whether to wait for leader online for data integrity or promote a follower node for availability. This part is still under construction and any advice is welcome.

  • Transaction: Talking about ACID, CQL has full "Consistency, Isolation, Durability" and a limited Atomicity support. That is even under strong consistency mode, CQL transaction is only supported on the leader node. If you want to do "read v, v++, write v back" parallelly and atomically, then the only way is "read v from the leader, v++, write v back to leader"

  • FUSE: CQL has a simple FUSE support adopted from CockroachDB. The performance is not very ideal and still has some issues. But it can pass fio test like:

    fio --debug=io --loops=1 --size=8m --filename=../mnt/fiotest.tmp --stonewall --direct=1 --name=Seqread --bs=128k --rw=read --name=Seqwrite --bs=128k --rw=write --name=4krandread --bs=4k --rw=randread --name=4krandwrite --bs=4k --rw=randwrite

Network Stack

DH-RPC := TLS - Cert + DHT

Layer Implementation
RPC net/rpc
Naming Consistent Secure DHT
Pooling Session Pool
Multiplex smux
Transport Security Enhanced TLS
Network TCP or KCP for optional later

Test Tools we use


Our team members published:

that inspired us:


Strong consistency bench result (2 miners, 8 core aws c5.2xlarge): CovenantSQL bench

As seen above, the concurrency pressure on the database increased gradually in the first 5 hours, and the write TPS also increased. When the TPS no longer grows, the concurrent pressure is maintained for 100 hours.


Use cases

Traditional App & ĐApp use cases

Traditional App

Privacy data

If you are a developper of password management tools just like 1Password or LastPass. You can use CQL as the database to take benefits:

  1. Serverless: no need to deploy a server to store your user's password for sync which is the hot potato.
  2. Security: CQL handles all the encryption work. Decentralized data storage gives more confidence to your users.
  3. Regulation: CQL naturally comply with GDPR.

IoT storage

CQL miners are deployed globally, IoT node can write to nearest CQL miner directly.

  1. Cheaper: Without passing all the traffic through a gateway, you can save a large bandwidth fee. And, CQL is a shared economic database which makes storage cheaper.
  2. Faster: CQL consensus protocol is designed for Internet where network latency is unavoidable.

Open data service

For example, you are the most detailed Bitcoin OHLC data maintainer. You can directly expose an online SQL interface to your customers to meet a wide range of query needs.

  1. CQL can limit specific SQL query statements to meet the needs while also balancing data security;
  2. CQL can record SQL query records on the blockchain, which is very convenient for customers to check their bills for long-tail customers and billing, like this
  3. For customers with high performance requirements, Slave nodes can be deployed at the customer to meet the needs of customers with low latency queries while enabling almost real-time data updates.

Secure storage

Thanks to the CQL data history is immutable, CQL can be used as a storage for sensitive operational logs to prevent hacking and erasure access logs.


Storing data on Bitcoin or Ethereum is quite expensive ($4305 / MB on Ethereum 2018-05-15). Programming is very complicated due to the lack of support for structured data storage. CQL gives you a low-cost structured SQL database and also provides more room for ĐApp to exchange data with real-world.



You can’t perform that action at this time.