Database service for the #realtimeweb - New Features

Adieu 2015! It has been a great year for the Appbase crew.

We started 2015 on a high note by joining the Techstars Cloud program. January - April were the most sceszplus! months and we did more strategy in those months than in the entire previous year.

After speaking with many of our users, we knew it was time to ditch our proprietary APIs and adopt a more transparent system. At the same time, we were seeing good usage for our Elasticsearch modules.

Using Elasticsearch as our primary data store, we redesigned our entire backend stack with performance as the key goal.

Thus SCALR was born. We ditched the v1, v2 ... numbering in favor of a more robust word based versioning (in lieu of Android's Key Lime Pie, and Ubuntu's Wily Werewolf). In building SCALR, we listened to our users and abandoned being a graph data store and instead adopted a JSON document data store (like MongoDB, Elastic, Couchbase, ...).

SCALR's been getting some great adoption. Our API usage figure (as of Jan 10, '16) is 101,977,442 - over 100 million API calls have been made from 1000+ apps built by developers. On our end, we have been cranking up supporting feature sets as we focus on user growth and engagement:

Our first supporting lib for SCALR has been in Javascript, we have had 14 releases for the lib since launching it in September '15. It is 100% compatible with Elasticsearch and bakes in some of the best seen realtime querying capabilities.

Towards the end of November, we added app collaboration - making it easy to share app data with teammates and clients, and an ability to create multiple security keys (much crucial for using Appbase in production).

As we changed our data model from graph to a document store, we ditched our old data browser interface and have adopted a new one that's more in line with the document store and can be used transparently with Elasticsearch (without Appbase as well).

Elasticsearch rolled out v2.0 in beta in November. As of today, all new apps on Appbase support the latest ES v2.1 and older apps (v1.6) can be updated to v2.1 via an upgrade request.

Our next library for supports Golang. Go has nice concurrency routines and is seeing rapid adoption among the systems software and infrastructure community (hint: Docker is built on Go).

We believe 2016 is a great time to carry our mission of making data streams driven apps forward. Some of the exciting things we are already working towards:

  1. Using Kafka as a frontend for traffic shaping and providing data durability + backups,
  2. Running on your own cloud provider or data centre,
  3. Improving our lib support in Javascript to cater to React, Meteor and Angular's communities,
  4. Data importer tools for MongoDB, ElasticSearch, PostgresSQL, Kafka / Kinesis,
  5. Visual Query Composer for ease of query creation without understanding Elasticsearch's APIs.
Author image
Founder at Appbase, read my musings on the db world.