Pebble

Pebble is a key-value store written in Go and developed by Cockroach Labs. It is based on an unfinished Go port of LevelDB. Furthermore, additional features and optimizations are inspired by parts of RocksDB. Pebble is specifically created to suit the needs of CockroachDB - some features from RocksDB are modified, while others are completely left out. A few examples of modified features include using offsets instead of pointers in the skiplist, and changing the rate of flushes to match user writes.

History

The project started in 2018 as a replacement for RocksDB running inside of CockroachDB.

Concurrency Control

Deterministic Concurrency Control

Pebble achieves concurrency during its commit pipeline, by using a 'commit queue'. In order to commit a batch, Pebble must first write the batch to its WAL and then add it to a memtable. The commit queue allows Pebble to synchronize writes to the WAL first. Then batches can be concurrently added to their memtables.

Data Model

Key/Value

Pebble uses two types of keys - user keys suited for the user API and 'Internal Keys' which are composed of the user key, a sequence number, and a value signifying the key type. The Internal Key type is used, since the LSM tree requires versioning with keys.

Foreign Keys

Not Supported

Indexes

Log-Structured Merge Tree

Pebble uses a LSM tree to store writes. All records are stored in batches, which in turn are stored in in-memory memtables and sstables. To support indexing, there is a separate skiplist implementation based on RocksDB. It stores offsets to the records within their batch in order to perform reads. Furthermore, batches themselves are transformed into levels of the LSM tree (with temp sequence numbers set to be the most recent sequence).

Isolation Levels

Snapshot Isolation

Since Pebble uses a LSM tree with a sequence number per log, this provides a very easy way to generate snapshots of the database by sequence number.

Logging

Physiological Logging

Pebble uses a write-ahead logging (WAL) infrastructure. Batches are first written to the WAL, and then inserted into a memtable. Pebble purposefully maps each WAL file to a memtable.

Query Execution

Tuple-at-a-Time Model

Query Interface

Custom API

Pebble emulates the custom RocksDB API for querying keys/values.

Storage Architecture

Disk-oriented

Pebble primarily uses sstables on disk which store various things such as internal keys/values and range tombstone deletions. It also has memtables, which temporarily store committed records before they are written to a sstable.

Storage Organization

Log-structured

System Architecture

Embedded

Embeddings

People Also Viewed