Google Cloud Bigtable is a managed sparse table store provided by Google Cloud. It scales to billions of rows, thousands of columns and petabytes of data. It suits workloads that need high-throughput, low-latency reads and writes by a single key, and it can also be used for stream or MapReduce-style processing.
Bigtable is a proprietary managed service. It runs only on Google Cloud and there is no self-managed edition, so buyers should weigh the dependence on one provider. The service is not versioned in the way that downloadable software is, and Google handles upgrades, so there is no release number or date to track.
Pricing is not published, and no free tier or commercial edition was confirmed. Organisations should consult Google Cloud for current costs and should model them against the volume of data and the throughput they expect, since sparse wide tables at large scale can lead to considerable storage and node charges.
The data model is a wide-column design in which data is addressed by row key, column and timestamp, so the choice of row key largely decides how well queries perform. It is therefore best suited to teams that know their access patterns in advance.