August 17, 2015 - Priyank Patel | Big Data Ecosystem

Big data, instant gratification – with Arcadia Instant.

Big data starts with Hadoop. But just because it’s big doesn’t mean you shouldn’t jump right in. That’s why we introduced Arcadia Instant. It’s the quickest, easiest way for you to connect directly to Hadoop, an all-in-one tool that unifies data discovery, visual analytics and business intelligence.

Also, it’s free. That should make it even easier.

Unified Visual Analytics and BI for Big Data

With Arcadia Instant, you build interactive visualizations using all your data. You can mix and match Hadoop data with other sources, like flat files and/or traditional or cloud databases. There’s a rich palette of intuitive authoring tools so that you can point and click your way from connecting and modeling, through visual analysis, and complete insights that include all your data sources.

Self-starting: connect and browse data-in-place.

Arcadia Instant starts with connecting you right to your Hadoop cluster. Just point and login; it works with whatever distro you’re using – Cloudera, Hortonworks, MapR, Amazon EMR, or even the Apache bits you pulled yourself. The connection explorer lets you quickly browse the data stored in those tables. Got some data in a flat file outside HDFS? It’s just as easy to pull that in.

Just like that, you’re set to get going on some datasets to work with right out of the box – out of the box, but not out of the data source. With Arcadia Instant, the data stays in its source repository; no need to extract your data back out to a warehouse and set up appliances or data marts. You can leave your data where it is and work with it right from there. No need to extract or refresh it again and again to keep your work from going stale, or trying to keep track of who has which part of the data in which extract from and where they put what.

When you connect to your Hadoop cluster, the the natural flexibility it offers in dealing with different data structures gives you immediate access to the list of all the tables it has, which Arcadia Instant presents as datasets. A dataset is simply a logical representation that displays a table of the data from any source you connected to; use these datasets as inputs to build your visualizations.

Arcadia Data lets you explore all the data stored within your Hadoop and non-Hadoop repositories with an intuitive easy-to-use connection explorer. Point and click to look at all of the data within your data sources so you can find the valuable data faster. Once you find the data you were looking for, create a dataset from it and start exploring it using Arcadia’s visual exploration tool.

Arcadia Instant lets you take a look at exactly how the source tables are set up, and to change the mapping of source to presentation. That way, you control the way data is labeled or named, to make it easier to understand, for yourself and the people who will consume your visualizations (you can change any of this later, too). Then, you can create any number datasets from any data source you can connect to.

Get it? Got it. Model it, slice and dice it.

Once you have a catalog of the tables that are in your source data repositories, you’ll want to organize them so they work consistently over time. As you’ve no doubt experienced, tables in your source data don’t always come with exactly the field names and structure you might have expected. Arcadia simplifies management of this critical metadata to ensure you can quickly and easily explore and analyze all of your data across all your sources.

Something you’ll find useful is the ability to join tables within any one of your sources, using Arcadia’s straightforward modeling interface.

Within the scope of a single visual application that includes data from multiple sources, you can take values from one data source, and use them to filter data from another data source. What’s more, you can present multiple sources in a single visual application, and create joins in run-time (they don’t need to be pre-materialized). That’s right: you can slice, dice, cross-filter and drill through across data sources – like Hadoop, MySQL and CSV files, right then and there.

You can also create segments within your dataset. Segments are any grouping that you’ll use again and again within that dataset – such as only looking at states on the East Coast, or selecting for cereal brands whose average calorie count per serving has risen in the last 5 years. In other words, you can define any subset of values in one or more dimensions, without carving out new tables. It gives you a simple and fast way to compare results and show trends. Given that big data is – well, big – the more specific you can get about underlying causes and trends and variations, the more your data pays off for you.

Show your work: Visualize and Compose data.

Once you’ve organized datasets and segments, it’s time to do some visual analysis and exploration. Build a bunch of these interactive visuals, and you can bundle them into a visual data application – like a dashboard, wired live into your source data.

Start by picking a visual type and feeding it to an existing dataset. Here’s an example of TV watching patterns over time, using the Calendar Heatmap visual. To include data in the chart, just drag any dimension or measure from the fields tab on the left, and drop it into the shelves on top; that define the visualization itself.

It’s also easy to slice and dice data within a visualization to make it easy to compare data across dimensions or measures in your dataset, so you can discover and reveal relationships.

 

A great way to display more than just 2 dimensions in your data is to pick different color palettes and ranges for measures, dimensions, segments, etc. You’ll find the color-picker that Arcadia provides almost as easy as using crayons. Almost.

 

Stand and deliver – and fear no interaction.

Arcadia Instant

Arcadia Instant Presentation Canvas

With all data connected and a visualization for each key aspect of your analysis, it’s time to bring to all of it together in one dashboard to that tells the whole story. Again, simple: take the visualizations you built, drag and drop them onto the canvas, and you’re set.

You’re probably thinking, “That looks like a dashboard”. But looks aren’t everything; unlike many others, Arcadia Instant has a dashboard that drills down live into all your data, big and otherwise, with no data left behind. We think you’ll find that useful, and here’s why.

Anyone who’s ever presented data in a meeting to the Big Boss knows how it goes: no matter how much you’ve slaved over a hot laptop for hours, some Clever Individual in the group pipes up and asks that you “drill down and show something else”, or “change the segments”, since “it’s all in the Hadoop cluster.”

That’s what makes an Arcadia Instant data app really useful: the data is already inside. Did Big Boss nod thoughtfully at Clever Individual? No worries: just click on any one of your visualizations and drill down to the source data. Sure, you could export the data in the chart on the spot into a CSV and mail it to Clever Individual so he or she can go off and play with some data by themselves.

Or … if you really want to score points for being a team player, just send the link to download Arcadia Instant to any Clever Individual you like. They can download it for free and build their own visualizations. They might not be as good as you (yet), but it will be good to have followers.


Related Posts