Written by Tim Moreton

In my previous post, we looked at how Acunu Analytics maintains OLAP-like cubes as event data is collected, enabling it to deliver instant queries on event high velocity streams.

One common use case that we see is aggregating results over a hierarchy of some aspect of the event data. In this post, I'm going to explore how Acunu makes this very simple, and provide some insight into how it works under the hood.

One of the tasks I set myself for our latest release was to refresh the slide deck we use to explain—by way of an example—how Acunu Analytics manages to respond to queries so fast and why it can keep the response time down, regardless of how data volumes grow. Even with a fully-indexed datastore, the time to query does increase as the volume grows. And it wouldn't be 'big data' if that didn't happen.

This blog post is the first in a series of posts where we'll look under the hood at how Acunu Analytics works. We'll take a tour some of the building blocks which you can put together to get rich, instant analytics on streaming data. 

This week we've officially announced Acunu Analytics for Apache Cassandra.

Real Time is an overloaded phrase. It means different things in different areas of the software industry, and its surge in use in the Big Data arena hasn't helped users understand what choices are available in an already noisy space. 

Part of what makes Cassandra such a great data-store is its distribution. Modeled after Amazon's Dynamo, it's a symmetrical system that is operationally simple, linearly scalable, with no single points of failure. However, despite its roots, Cassandra's implementation eschewed one important experience-driven design element of Dynamo, virtual nodes.

To understand why this is important, it helps to understand the basics of this style of distribution.

It's a well known fact that buses arrive in threes, right? Well I thought I’d look for this effect in some real data. 

Estimating the demand for a new service is tough. Even predicting the future demand for an existing service turns out to be difficult since demand tends to be unpredictable and changes over time.

I had the pleasure recently to spend several hours talking to some people in the City about Big Data.  These were a mixture of senior technology people from a large professional services firm attempting to figure out what Big Data might mean to them.

A web browser's main job is to render HTML. However, as we will see in this article, for certain rendering tasks it can be painfully slow at performing this primary function.