Bigtable news from Google Cloud NEXT in SF, March 8–10

[Misha Brukman]

[ bcc: google-cloud-bigtable-announce ]

Hi everyone,

During Google Cloud NEXT, March 8–10 in San Francisco, we launched several new Bigtable features:

• BigQuery / Bigtable federated queries (beta): you can connect BigQuery to Bigtable and run SQL queries on your data in Bigtable, without having to write a Dataflow or Dataproc job. We presented the details of the integration with a demo together with our partner iCharts, to demonstrate visualization via the BigQuery bridge. Notably, iCharts only integrates with BigQuery, not Bigtable, and it seamlessly works with the BigQuery / Bigtable integration. Try out the BigQuery / Bigtable integration, whether directly via SQL from BigQuery, or via BigQuery-integrated tools and let us know what you think.
  
  
• Bigtable development instances: for a lower-cost getting started experience, or functional/system testing which does not require high-throughput but requires full API compatibility and storage durability. Together with the Bigtable emulator, it provides an additional option for development and testing workflows. If there's anything else you'd like to see from Bigtable to improve dev/test workflows, please let us know.

We also presented several case studies of Bigtable customer use cases, Bigtable architecture, data model, and schema design:

• Bigtable in action, focusing on schema design and use cases:
• Server Density discussed storing and querying time series with OpenTSDB + Bigtable
• CCRi discussed geospatial indexing with GeoMesa + Bigtable
• Bigtable, Dataflow and Dataproc at WePay and Outbound
• How eBay built ShopBot with billions of listings, using Bigtable & other Google Cloud services
• How Brightcove uses Bigtable for time series data

We have also added a few more integrations with Bigtable:

• Geospatial + temporal data: GeoMesa
  
• Graph databases: JanusGraph and HGraphDB, both API-compatible with Apache TinkerPop
• ... any other integrations you think we should be working on?

As always, we look forward to your thoughts and feedback!

Thanks,

Misha