Peta Cube
Aggregates data at any granularity on OLAP cubes in Hadoop. An easy web interface manages, builds, monitors and queries them, so cube maintenance does not become a specialist job.
PetaBolt
OLAP on Hadoop. The query layer we stand up when the data is too big for the tools already in the building.
Architecture
Four components
Each does one job, and together they present a single query surface to whatever is asking.
Peta Table
An operational data store integrating data from multiple sources on Hadoop — the layer where enterprise systems and external feeds are reconciled before anything is aggregated.
Peta View
Interactive query at sub-second latency, over billions of rows, with no scalability limitation and no coding required.
Peta API
RESTful interfaces for concurrent access from R, SQL and ODBC, so models and dashboards read the same numbers.
Operating characteristics
What it is like to run
The numbers that matter once it is in production, rather than the ones that look good in a pitch.
400x
Faster reporting
10bn+
Rows queried
- Extreme-scale OLAP engine designed to query billions of rows on Hadoop
- Scale-out architecture supporting thousands of concurrent users, highly available
- Decreased data latency, which is what makes real-time analytics real
- Streaming data support down to minute-level latency
- Access to all your data, rather than the subset that fitted the old warehouse
- Seamless integration with BI tools — Tableau, Business Objects, Excel
- Explore, slice, dice and drill down on massive cubes
Foundation
Enterprise-grade Hadoop underneath
The platform delivers on the promise of Hadoop for mission-critical and real-time production use — dependability, ease of use and world-record speed across Hadoop, NoSQL, database and streaming workloads in one unified big data platform.
Where it is already running
Financial services, retail, media, healthcare, manufacturing, telecommunications and government organisations, and leading Fortune 100 and Web 2.0 companies.
Stand it up against your data
We will run PetaBolt against a slice of your warehouse and show you the query times on your own volumes.