The basics
Hyperledger Fabric is an open source enterprise-grade permissioned distributed ledger technology (DLT) platform, designed for use in enterprise contexts, that delivers some key differentiating capabilities over other popular distributed ledger or blockchain platforms.
Fabric has a highly modular and configurable architecture, enabling innovation, versatility and optimization for a broad range of industry use cases including banking, finance, insurance, healthcare, human resources, supply chain and even digital music delivery.
Fabric is the first distributed ledger platform to support smart contracts authored in general-purpose programming languages such as Java, Go and Node.js, rather than constrained domain-specific languages (DSL). This means that most enterprises already have the skill set needed to develop smart contracts, and no additional training to learn a new language or DSL is needed.
The Fabric platform is also permissioned, meaning that, unlike with a public permissionless network, the participants are known to each other, rather than anonymous and therefore fully untrusted. This means that while the participants may not fully trust one another (they may, for example, be competitors in the same industry), a network can be operated under a governance model that is built off of what trust does exist between participants, such as a legal agreement or framework for handling disputes.
Fabric can leverage consensus protocols that do not require a native cryptocurrency to incent costly mining or to fuel smart contract execution. Avoidance of a cryptocurrency reduces some significant risk/attack vectors, and absence of cryptographic mining operations means that the platform can be deployed with roughly the same operational cost as any other distributed system.
The combination of these differentiating design features makes Fabric one of the better performing platforms available today both in terms of transaction processing and transaction confirmation latency, and it enables privacy and confidentiality of transactions and the smart contracts (what Fabric calls "chaincode") that implement them.
Consensus mechanism
Fabric currently offers a CFT (crash fault-tolerant) ordering service implementation based on the etcd library of the Raft protocol.
The Raft protocol is the go-to ordering service choice for production networks, the Fabric implementation of the established Raft protocol uses a "leader and follower" model, in which a leader is dynamically elected among the ordering nodes in a channel (this collection of nodes is known as the "consenter set"), and that leader replicates messages to the follower nodes. Because the system can sustain the loss of nodes, including leader nodes, as long as there is a majority of ordering nodes (what's known as a "quorum") remaining, Raft is said to be "crash fault tolerant" (CFT). In other words, if there are three nodes in a channel, it can withstand the loss of one node (leaving two remaining). If you have five nodes in a channel, you can lose two nodes (leaving three remaining nodes). This feature of a Raft ordering service is a factor in the establishment of a high availability strategy for your ordering service. Additionally, in a production environment, you would want to spread these nodes across data centers and even locations. For example, by putting one node in three different data centers. That way, if a data center or entire location becomes unavailable, the nodes in the other data centers continue to operate.
More information can be found on the official Hyperledge Fabric documentation website.