Big Data is supposed to level the business playing field, but up until now it has not. Because even though it’s cheaper to store data in Hadoop or to work with open source NoSQL in theory, it’s too expensive in a more practical sense because it’s too hard and the talent isn’t available.
Vendors aim to change that — whether it’s by getting NoSQL technologies into the hands of more developers or by making big data analytics more accessible. We feature some of them today.
Developers Love Couchbase Mobile
We know developers love MongoDB, but what about its NoSQL rival Couchbase?
If downloads by 100,000 mobile developers signal a sweetly pumping heart, then the open source NoSQL vendor might just be doing something right. Couchbase’s mobile offering, appropriately named Couchbase Mobile, is just 13 months old and it has gained traction quickly.
And that momentum is likely to continue because late last week Couchbase revealed that it has built a new generation of mobile apps that work with or without a network connection.
Bob Wiederhold, CEO at Couchbase said that “the day is quickly arriving when mobile apps will only meet user expectations by delivering the same experience offline as they do online”.
He is seeing a significant shift in the market as app developers and large enterprises adopt this mindset and only build apps that work regardless of connectivity.
“The reason for Couchbase Mobile’s accelerated success is simple: it makes it easy for developers to build the new generation of mobile apps. And our performance and ease of use advantage make Couchbase uniquely positioned to take advantage of this opportunity,” he added.
Couchbase rolled out Couchbase Mobile 1.1 last week, which adds change notification capabilities to its cloud database. It leverages change notifications, or web hooks, so that developers don’t have to write custom code to poll for changes within either the cloud or device database.
As a result, when data is changed in either database, external services and systems are notified and can run the appropriate application logic. What’s the benefit? It saves developers time by reducing the amount of code they need to write to maintain data integrity in both the cloud and the device, which in turn helps them get their applications to market faster.
Can Datawatch Make Everyone a Big Data Expert?
When a solution provider says it can make everyone a big data expert, they don’t literally mean “everyone." In most cases they’re talking about data workers or business analysts who have experience working with data.
That being said, Datawatch released Datawatch Monarch version 13 today and it goes a long way. It’s the latest release of its flagship data preparation solution and has some useful, enterprise grade functionalities. According to Datawatch it provides:
- The ability to grab data from almost any source, including multi-structured documents such as PDFs, text and spool files, as well as all major relational databases, Salesforce.com, web pages, CSVs and more.
- Point and click ease- 80 prebuilt data manipulation functions
- A facility to use existing reports in formats like .PDF or .TXT to automatically bring data in for analysis.
- An ability to unlock data in customer invoices, purchase orders, third party research and virtually any other kind of document. It inspects documents and extracts all the relevant data into a tabular, spreadsheet-like display.
- Drag-and-drop web pages so that content from a web page can become a data source, so that you can add important data like competitive pricing or market research that is only available via the web.
- Export to your favorite tool –Prepared data can be saved in a variety of native BI formats so you can immediately visualize results not just in Datawatch Designer, but in other popular BI and advanced analytics tools such as Tableau, Qlik, IBM Watson Analytics, Tibco Spotfire, SAS, SAP Lumira, Dell Statistica, Excel and many others.
Data prep can be a tedious part of a data worker’s job and if a tool can do the same quality of work as a human, then the potential value-add is substantial.
Platfora Remedies Big Data Disillusionment
You can’t do big data discovery with yesterday’s tools, that’s why Platfora geared its platform specifically with Hadoop and Spark in mind.
After all, companies who need to derive insights via multi-channel customer analytics, Internet of Things (IoT) pattern analysis and Advanced Persistent Threat (APT) cybersecurity analytics need to work with massive and interlinked datasets; desktop BI tools weren’t built for that. At least that’s the argument that Ben Werther, CEO of Platfora makes.
“Disillusionment with big data usually goes hand-in-hand with attempts to force fit old IT practices and common desktop BI products, but when business analysts are given purpose-built tools designed for, and native to, the world of big data, magic happens and remarkable insights quickly follow,” he said.
Last Friday Platfora introduced Platfora 4.5, which aims to make it easier for data scientists and the business to work together and to derive big data insights in record time.
This release introduces Platfora’s Aurora Visualization Engine, a fully HTML5-compliant engine that provides intricate and vibrant visualizations without sacrificing the ability to work with petabytes of raw or structured data. It promises faster performance, an ability to interact with data on the fly, an enhanced ability to freely explore big data to find patterns and insights at drag-and-drop speed and more.
The new release also provides improvements in Platfora's scale-out query engine which yields 10x performance improvement for queries—especially those that involve fine grain analysis of billions of rows; richer dynamic segmentation that makes it easier to analyze customer populations and behavior; and enhanced sharing and security.
Cubes on Hadoop?
Praveen Kankariya, founder and CEO of Kyvos Insights thinks business users want to use OLAP (online analytical processing) to gain big data insights. So why don’t they do that, you might be asking. Because OLAP has performance and scalability constraints. In other words, it’s a showstopper.
Or at least it was. This morning Kyvos Insights emerged from stealth and revealed Kyvos, a solution that brings a new model of OLAP to big data. With it users can visually create and analyze “cubes on Hadoop,” or in other words, explore and analyze big data interactively, working directly on Hadoop.
What might the wins for the business user look like? Self-service analytics; an ability to develop insights from all data, both structured and unstructured, regardless of size and granularity with a simple, drag and drop interface; access to big data via the tools of their choosing ranging from Excel to Tableau; big data query results in seconds and more.
Wrapping Up the Pieces
Companies see the value in leveraging big data, but there are challenges to overcome. Business users recognize this, so in many cases they don’t try. Some of those who do try stumble and drop out of the game. It’s unfortunate because there are big wins to be had and without them some companies may fail to thrive, let alone survive.
The good news is that big data and big data analytics keep getting easier to adopt and more user friendly. Those featured in today’s Big Data Bits are prime examples.