Skip to main content

Permissioned Distributed Ledger based on Ethereum

Project description

HydraChain
==========

[![Join the chat at https://gitter.im/HydraChain/hydrachain](https://badges.gitter.im/Join%20Chat.svg)](https://gitter.im/HydraChain/hydrachain?utm_source=badge&utm_medium=badge&utm_campaign=pr-badge&utm_content=badge)
[![Build Status](https://travis-ci.org/HydraChain/hydrachain.svg?branch=master)](https://travis-ci.org/HydraChain/hydrachain)

HydraChain is an extension of the [Ethereum](https://ethereum.org/) platform which adds support for creating [*Permissioned Distributed Ledgers*](http://www.ofnumbers.com/2015/04/06/consensus-as-a-service-a-brief-report-on-the-emergence-of-permissioned-distributed-ledger-systems/). Its primary domain of application are [*private chain* or *consortium chain*](https://blog.ethereum.org/2015/08/07/on-public-and-private-blockchains/) setups.

Features
--------

**Full Compatibility to the Ethereum Protocol**

HydraChain is 100% compatible on an API and contract level. Existing tool chains to develop and deploy *Smart Contracts* and *ÐApps* can easily be reused.

**Accountable Validators**

The main difference is the byzantine fault tolerant consensus protocol ([*detailed here*](https://github.com/HydraChain/hydrachain/blob/master/hc_consensus_explained.md)) which does not depend on proof-of-work. Instead it relies on a registered and accountable set of validators which propose and validate the order of transactions.

**Instant Finality**

New blocks are negotiated by the validators. A quorum by the validators which signs the block is required, before it is added to the chain. Thus there will be *no forks or reverts.* Once a block is committed, the state is final.

The protocol allows for *sub second block times*. New blocks are only created in the presence of pending transactions though.


**Native Contracts**

HydraChain provides an infrastructure to develop smart contracts in the Python high level language. Benefits are significantly reduced development times and better debugging capabilities. As the Ethereum Virtual Machine is bypassed, native contract execution is also way faster.
Native Contracts support the ABI and are inter-operable with EVM based contracts written in the Solidity or Serpent languages and can co-exist on the same chain. The constraint, that all validators need to run a node configured with the same set of native contracts is well manageable in private chain settings.

**Customizability**

Many aspects of the system can be freely configured to fit custom needs. For example transaction fees, gas limits, genesis allocation, block time etc. can easily be adjusted.

**Easy Deployment**

Setting up a test network can be done with almost zero configuration. [Dockerfile templates](https://github.com/HydraChain/hydrachain/tree/master/docker) are available.

**Open Source**

The core software is open source and available under the permissive [MIT license](https://en.wikipedia.org/wiki/MIT_License).

**Commercial Support**

Consulting, support plans and custom development are offered by [brainbot technologies](http://www.brainbot.com) and a network of partners.

Upcoming Features
-----------------
*Note: We are happy to align our roadmap with the priorities of our users. If you have a certain requirement or prioritization, feel free to [file an issue](https://github.com/HydraChain/hydrachain/issues) or directly [contact us](mailto:heiko.hees@brainbot.com).*

**Documentation**

We are working on a comprehensive set of documentation which covers various deployment scenarios. This will be accompanied by a range of example contracts with a focus on use cases of the financial industry.

**Proof of Identity - KYC/AML**

An extension to ensure that all transactions in the system are between registered participants only. The goal is full audibility while preserving as much privacy as possible.

**Selective State Sharing**

Non-validating users of the system which must not know complete state (e.g. all transactions), are still able to verify the results of transaction and the state of contracts they interact with.

**Chain Inter-Operability**

Multi-chain setups can solve scalability and privacy requirements.
As the term *Hydra* in the name already hints, that the software will support to run a node which concurrently participates in multiple chains. Next to other applications, this allows to support cross chain asset transfers as a native feature.


Setup & Invocation
------

**Install dependencies on Ubuntu**

> sudo apt-get install libssl-dev libffi-dev libtool python-dev autoconf automake

**Installation**

> git clone https://github.com/HydraChain/hydrachain
> cd hydrachain
> python setup.py develop


**Multiple nodes in a single process**

> hydrachain -d datadir runmultiple --num_validators=3 --seed=42


**Multiple nodes in multiple processes on the same machine**

> hydrachain -d <datadir> rundummy --num_validators=3 --seed=42 --node_num=0

Example how to run three nodes in three terminals:

> hydrachain -d datadir0 rundummy --node_num=0
> hydrachain -d datadir1 rundummy --node_num=1
> hydrachain -d datadir2 rundummy --node_num=2

The `rundummy` command automatically configures a setup for `num_validator` nodes (instances of the application) which are running on the same machine. The node id of each instance can be specified by `--node_num=<int>` and `--seed=<int>` can be used to configure a different set of keys for all nodes.

**Multiple nodes in a single docker daemon**

See the [Dockerfile templates](https://github.com/HydraChain/hydrachain/tree/master/docker)



Status: Work in Progress
------------------------
- 08.03.2016 - v0.3.0 - Fix versioning scheme.
- 08.03.2016 - v0.2.0 - Add support for min\_block\_time/multiple tx/block.
- 16.01.2016 - v0.1.10 - Automate docker image creation
- 27.11.2015 - v0.1.0 - Automate PyPI release process
- 26.11.2015 - v0.0.7 - Various fixes; first PyPI release
- 21.11.2015 - v0.0.6 - Various fixes
- 21.10.2015 - v0.0.4 - Run multiple node instances in the same process
- 16.10.2015 - v0.0.3 - Update docker configuration
- 18.09.2015 - Added zero config docker compose files
- 09.09.2015 - Initial release, work in progress.

Project details


Download files

Download the file for your platform. If you're not sure which to choose, learn more about installing packages.

Source Distribution

hydrachain-0.3.2.tar.gz (43.2 kB view details)

Uploaded Source

Built Distribution

hydrachain-0.3.2-py2.py3-none-any.whl (50.2 kB view details)

Uploaded Python 2 Python 3

File details

Details for the file hydrachain-0.3.2.tar.gz.

File metadata

  • Download URL: hydrachain-0.3.2.tar.gz
  • Upload date:
  • Size: 43.2 kB
  • Tags: Source
  • Uploaded using Trusted Publishing? No

File hashes

Hashes for hydrachain-0.3.2.tar.gz
Algorithm Hash digest
SHA256 17bc152d4d8da0f35b027f5cef52f941840b63cff62ba329fa334a48fe10c0c2
MD5 a5f73b5c8c8a2a3d42627a54ba282311
BLAKE2b-256 96034ca8ea447a1ca3eceb286b9efebbc5a1b5934cd782ad3531b0fb1d0cba46

See more details on using hashes here.

File details

Details for the file hydrachain-0.3.2-py2.py3-none-any.whl.

File metadata

File hashes

Hashes for hydrachain-0.3.2-py2.py3-none-any.whl
Algorithm Hash digest
SHA256 192e2f94974c696b269bec8919304d773a92cdb85ceee8cd1d346840371fff65
MD5 f1c10e385a60b8bb005b2a9692267e6c
BLAKE2b-256 aea6846038e6ead168d6361eb1d81611eb8e125d9c5733fe3285beff91e7c7a9

See more details on using hashes here.

Supported by

AWS AWS Cloud computing and Security Sponsor Datadog Datadog Monitoring Fastly Fastly CDN Google Google Download Analytics Microsoft Microsoft PSF Sponsor Pingdom Pingdom Monitoring Sentry Sentry Error logging StatusPage StatusPage Status page