How it works
At its heart, ripple is a distributed database. The current state is replicated on all ripple nodes. The system provides a way for nodes to verify and reach consensus on the correct view of the database.
This state can be maintained without requiring nodes to trust each other.
Built on top of this distributed database is what we call the Ledger. This Ledger contains a list of Accounts and the balance of each account. The owner of an account can send credits to any other account. There are 100 billion Ripple credits that exist in the Ledger. No new credits can ever be created.
A ripple Account is an entry in the Ledger. People typically have one account that stores their Ripple Balance, IOUs and trust grants to and from other accounts. Each account has an address and a private key. Anyone that knows an account's private key can authorize transactions from that account.
Unique Node List
The Unique Node List (UNL) is the key to the process of keeping the network in sync. Each Node has their own UNL. Unique here means nodes not run by the same group or individual. A node chooses to accept a particular Ledger based on how many of the unique nodes on its list have validated that ledger.
The network requires a small fee for all transactions in order to prevent certain attacks. All fees are paid in Ripple Credits. The fees are redistributed evenly to all the Ripple account holders in proportion to their account balance (i.e. destroyed).
One Ledger Cycle happens every 2-10 seconds.
Ledgers are really hash trees. This allows servers to only fetch the small portion of the ledger that changes between ledger updates. ripple also uses a Skip List to quickly verify that a past Ledger was in fact a parent of the current Ledger.