We present Magma, a write-optimized high data density key-value storage engine used in the Couchbase NoSQL distributed document database. Today's write-heavy data-intensive applications like ad-serving, internet-of-things, messaging, and online gaming, generate massive amounts of data. As a result, the requirement for storing and retrieving large volumes of data has grown rapidly. Distributed databases that can scale out horizontally by adding more nodes can be used to serve the requirements of these internet-scale applications. To maintain a reasonable cost of ownership, we need to improve storage efficiency in handling large data volumes per node, such that we don't have to rely on adding more nodes. Our current generation storage engine, Couchstore is based on a log-structured append-only copy-on-write B+Tree architecture. To make substantial improvements to support higher data density and write throughput, we needed a storage engine architecture that lowers write amplification and avoids compaction operations that rewrite the whole database files periodically. We introduce Magma, a hybrid key-value storage engine that combines LSM Trees and a segmented log approach from log-structured file systems. We present a novel approach to performing garbage collection of stale document versions avoiding index lookup during log segment compaction. This is the key to achieving storage efficiency for Magma and eliminates the need for random I/Os during compaction. Magma offers significantly lower write amplification, scalable incremental compaction, and lower space amplification while not regressing the read amplification. Through the efficiency improvements, we improved the single machine data density supported by the Couchbase Server by 3.3x and lowered the memory requirement by 10x, thereby reducing the total cost of ownership up to 10x. Our evaluation results show that Magma outperforms Couchstore and RocksDB in write-heavy workloads.

Couchbase Server is a rethinking of the database given the current set of realities. Memory today is much cheaper than disks were when traditional databases were designed back in the 1970's, and networks are much faster and much more reliable than ever before. Application agility is also an extremely important requirement. Today's Couchbase Server is a memory- and network-centric, shared-nothing, auto-partitioned, and distributed NoSQL database system that offers both key-based and secondary index-based data access paths as well as API- and query-based data access capabilities. This is a major change from Couchbase's roots; in its early days, its focus was entirely on high performance and highly available key-value (memcache) based caching. Customer needs and competitive pressures in the evolving non-relational database market also accelerated this change. This paper describes the architectural changes needed to address the requirements posed by next-generation database applications. In addition, it details the implementation of such an architecture using Couchbase Server and explains the evolution of Couchbase Server from its early roots to its present form. Particular attention is paid to how today's Couchbase Server cluster architecture is influenced by the memory-first, high-performance, and scalability demands of typical customer deployments. Key features include a layer-consolidated cache, a consistency-controllable interplay between updates, indexes, and queries, and a unique "multi-dimensional" approach to cluster scaling. The paper closes with a look at future plans for supporting semi-structured operational data analytics in addition to today's more OLTP-like, front-facing use cases.

Database consistency and recoverability require guaranteeing write atomicity for one or more pages. However, contemporary database systems consider write operations non-atomic. Thus, many database storage engines have traditionally relied on either journaling or copy-on-write approaches for atomic propagation of updated pages to the storage. This reliance achieves write atomicity at the cost of various write amplifications such as redundant writes, tree-wandering, and compaction. This write amplification results in reduced performance and, for flash storage, accelerates device wear-out. In this paper, we propose a flash storage interface, SHARE. Being able to explicitly remap the address mapping inside flash storage using SHARE interface enables host-side database storage engines to achieve write atomicity without causing write amplification. We have implemented SHARE on a real SSD board, OpenSSD, and modified MySQL/InnoDB and Couchbase NoSQL storage engines to make them compatible with the extended SHARE interface. Our experimental results show that this SHARE-based MySQL/InnoDB and Couchbase configurations can significantly boost database performance. In particular, the inevitable and costly Couchbase compaction process can complete without copying any data pages.

We present Nitro, a high-performance in-memory key-value storage engine used in Couchbase 4.5 Global Secondary Indexes. The Nitro storage engine is well suited for the recent hardware trends like large amounts of memory and many CPU cores. The storage engine leverages latch-free data structures and tries to achieve linear scalability for the index read-write operations. The Nitro storage engine offers concurrent readers and writers, lightweight database snapshots, stable scan, backup and recovery operations. We integrated Nitro into the Couchbase Global Secondary Indexes (GSI) and observed significant improvement in performance compared to our disk oriented storage engine configured with the same amount of memory for buffer cache. On a 32 core machine, we observed an end-to-end GSI server insertion throughput of 1,650,000 entries/sec and index update throughput of 822,000 entries/sec. A single instance of Nitro data structure running on a 40 core machine achieved a peak insertion throughput of 4 million index entries/sec and entry lookup throughput of 10 million lookups/sec.

Indexing key-value data on persistent storage is an important factor for NoSQL databases. Most key-value storage engines use tree-like structures for data indexing, but their performance and space overhead rapidly get worse as the key length becomes longer. This also affects the merge or compaction cost which is critical to the overall throughput. In this paper, we present ForestDB, a key-value storage engine for a single node of large-scale NoSQL databases. ForestDB uses a new hybrid indexing scheme called HB+-trie, which is a disk-based trie-like structure combined with B+-trees. It allows for efficient indexing and retrieval of arbitrary length string keys with relatively low disk accesses over tree-like structures, even though the keys are very long and randomly distributed in the key space. Our evaluation results show that ForestDB significantly outperforms the current key-value storage engine of Couchbase Server [1], LevelDB [2], and RocksDB [3], in terms of both the number of operations per second and the amount of disk writes per update operation.