Spanner: Google's Globally-Distributed Database

By J. Corbett, Joseph Dean, Michael Epstein, Andrew Fikes, Christopher Frost, JJ Furman, Sanjay Ghemawat, Andrey Gubarev, Christopher Heiser, Peter H. Hochschild and 16 more. USENIX Symposium on Operating Systems Design and Implementation, 2012

1,654 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Hardware and Architecture

View Paper: https://www.semanticscholar.org/paper/082bc77513862f8d709322916f44d6fe2f2d06d7

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Hardware and Architecture

Citations in the corpus

Stronger Semantics for Low-Latency Geo-Replicated Storage (2013, positive): “The implementers of Sinfonia [1], TxCache [46], HBase [ 32], and Spanner [22], also recognized the importance of limited transactions.”
TAO: Facebook's Distributed Data Store for the Social Graph (2013, background): “Span-ner, the next generation globally distributed database developed at Google after Megastore, introduces the concept of a time API that exposes time uncertainty and leverages that to improve commit throughput and provide snapshot isolation for reads [8].”
OmniLedger: A Secure, Scale-Out, Decentralized Ledger via Sharding (2018, background): “The proven and obvious approach to building “scale-out” databases, whose capacity scales horizontally with the number of participants, is by sharding [14], or partitioning the state into multiple shards that are handled in parallel by different subsets of participating validators.”
In-Memory Big Data Management and Processing: A Survey (2015, neutral): “More recently, a new type of relational databases, called NewSQL (e.g., Google Spanner [8], H-Store [36]) has emerged.”
Big Data Service Architecture: A Survey (2020, methodology): “The current widely used NewSQL databases are: Spanner, MemSQL [31-32].”
Towards Scaling Blockchain Systems via Sharding (2018, background): “rmshardingtorefertothecombination of replication and partitioning as shown in Figure 1a. This architecture is adopted by recent distributed database systems to achieve fault tolerance and scalability [22]. Each partition is replicated over multiple replicas, and its content is kept consistent by con”
BLOCKBENCH: A Framework for Analyzing Private Blockchains (2017, methodology): “Current transactional, distributed databases [46, 14] employ classic concurrency control techniques such as two-phase commit to ensure ACID.”
Low Latency Geo-distributed Data Analytics (2015, background): “Google Spanner [28] is an instance of a distributed database deployed at scale.”
Machine Learning Operations (MLOps): Overview, Definition, and Architecture (2022, neutral): “file systems [41], [42], or distributed databases [43], [44] combined with parallel and distributed data processing algorithms (e.”
No compromises: distributed transactions with consistency, availability, and performance (2015, methodology): “[9] CHANG, F., DEAN, J., GHEMAWAT, S., HSIEH, W. C., WALLACH, D. A., BURROWS, M., CHANDRA, T., FIKES, A., AND GRUBER, R. E. Bigtable: A distributed storage system for structured data.”
Tendermint: Byzantine Fault Tolerance in the Age of Blockchains (2016, background): “In practice, synchronous solutions are only used where the message latency is under extremely well defined control, for instance between controllers on an airplane [49], or between datacenters utilizing synchronized atomic clocks [23].”
Staring into the Abyss: An Evaluation of Concurrency Control with One Thousand Cores (2014, methodology): “In distributed systems, synchronization is accomplished using software protocols [31] or external clocks [9].”
Hashing for Distributed Data (2015, background): “However, because more and more large scale datasets have emerged in real-world applications, the data is often distributed across different locations, such as distributed databases (Corbett et al., 2013), images/videos over the networks, etc.”
Toward Scalable Systems for Big Data Analytics: A Technology Tutorial (2014, background): “• Spanner [157] is the first system to distribute data on a global scale and support externally consistent distributed transactions.”
CockroachDB: The Resilient Geo-Distributed SQL Database (2020, background): “Spanner [4, 17] is a SQL system that provides the strongest isolation level, strict serializability [30].”
The Past, Present and Future of Cyber-Physical Systems: A Focus on Models (2015, methodology): “A technique similar to Ptides was independently developed at Google for managing distributed databases in the Google Spanner system [66].”