minusonedb - Platform

We offer a hosted platform appropriate for most data-intensive tasks, especially user analytics of all kinds. This platform includes a complete set of tools for loading and managing data, managing the operational lifecycle of deployed environments and integrating with pre-existing solutions.

Single-tenant deployment provides for complete risk isolation. It is easy to manage, scalable and fast. New data environments are deployed and operational in minutes. Our services help you understand your data and design your schema. Events are captured at any speed. Large datasets are loaded easily in hours. Queries are executed in seconds.

Our platform provides four storage technologies that can be combined for a variety of solutions and use cases:

Data Lake

Efficient, organized storage includes schema management, declarative data transformation, transformation via user software and key/value data retrieval. Data Lakes can be built from our Archive or your own data.

Data Store

Indexed search-based storage used for retrieval of arbitrary data via query, aggregation and custom metrics. Data Stores support near real-time indexing and very fast query performance.

Session Store

Fast storage with high update concurrency appropriate for data modified in flight and shared application-layer data.

What makes us better?

Any scale, always fast

minusonedb handles Terabytes to Petabytes of data in near flat-time for both writes and reads.

Simplify your operations

Deployment, management, reconfiguration and termination of environments and resources are all performed via scriptable services or push-button automation.

Choose your own features

Different storage technologies can be combined for solutions as diverse as event capture, log aggregation, pattern detection and trend analysis. Data can be from any domain including social media, user analytics, location data or performance metrics of any kind.

Predictable costs

Pay for the infrastructure you use by the minute. Our pricing is designed to be transparent and reasonable.

What do I need?

Environment

You will want one environment per project. A project is usually defined by a set of data that "goes together" and a group of people with access to that data. In many cases most (or all) data can be hosted in a single environment but in cases of sensitive (or third-party) data sometimes multiple environments are required to preserve access boundaries.

Data processing servers

Data Processing servers provide all input and output to the environment. You will need more servers in write-heavy and read-heavy use cases. These servers are elastically scalable to handle increased traffic. One Data Processing server supports:

2200-2500 docs/second written (depending on request size)

1-2 concurrent queries/exports

Data stores

Data Stores are deployed in clusters that are sharded to index larger datasets and replicated for redundancy and query throughput. For most analytics use cases 1-2 replicas are sufficient (test/stage vs production use cases). Boost data stores are typically only deployed during the initial creation of an index for batch indexing performance. Standard data stores provide this performance. Boost data stores provide approximately 5x indexing performance.

1 shard supports 450 GB of indexed data and 4500 documents written/second

1 replica supports 2-4 concurrent queries

Session stores

Session Stores are also deployed in clusters. Generally 2 replicas is recommended.

1 shard supports 450,000 concurrent documents and 45,000 updates/second