We've built Spire with developers in mind, and we know that for developers it’s all about the application. Spire gives you a 100% natively distributed database built on Hadoop + HBase, real-time performance, SQL and full-text search, and simplified management. With Spire, you are free to focus on your application, using familiar tools, instead of spending your time and scarce resources dealing with a complex infrastructure or trying to get a patchwork solution to work.

You've used Hadoop but need real-time

Hadoop is fantastic for storing and batch processing massive datasets. In fact, Spire is built on Hadoop. However, Hadoop is not a database and it is not real-time. If you need to power a web application or do real-time analytics with data you have processed with Hadoop, you often need to do large amounts of pre-aggregation, pre-calculation, and then dump the data to a more traditional datastore. Often, this isn’t economical or feasible. Spire builds on Hadoop with real-time SQL and MapReduce. Spire bridges the gap between Hadoop and real-time.

You've used HBase but want rich queries and search

If you’ve used HBase with Hadoop, you’re familiar with its scalability and robustness. That's why we've built Spire on Hadoop + HBase. However, you've likely also experienced the limited query capabilities of HBase's primary-key only indexing. You can take the next step with Spire. With Spire's real-time distributed indexing, SQL support and full-text search you can perform data operations — like grouping, counting, and filtering data — fast enough to power user-facing data. In addition, our JSON integration means the database integrates naturally with your application.

You've tried a document-oriented "NoSQL" database but massive scale

For the right problem, document-oriented databases, often called "NoSQL", are a lot more flexible than a conventional relational database and they can solve a number of architectural problems when dealing with datasets larger than a relational database can handle. However, document-oriented databases are not massively scalable. Sharding is the culprit. Sharded data and sharded indexes slow performance and add burdensome complexity to data at massive scales. Spire gives you a 100% natively distributed database with a fully distributed index. Spire has no shards — and no hassles.

You want to use your SQL skills with Big Data

The relational database is highly tunable and remarkably fast for moderate-sized datasets. But, when you need to scale beyond one node you sacrifice many features: indexes, rich queries, failover safety, etc. You’re quickly limited to just primary key lookups and a few simple queries. Instead of going down this painful path, Spire lets you use your data in familiar ways: real-time SQL queries. Everything you've learned about working with a database translates directly to Spire.

You've got Big Data but you don’t know where to start

You know when you've got Big Data. Your data is overwhelming your database and processing infrastructure, and it just keeps growing. From a single node to two. From two nodes to five and beyond: it just keeps growing. You need to get this fixed quickly, but where to start? Start with Spire. You'll get Hadoop scalability — the best you can get — plus real-time, SQL and enterprise manageability.