Three information infrastructure myths debunked
The Information Management keynote session at IBM Insight 2014 brought new product offerings, memorable stories and answers to some common information infrastructure myths. Beth Smith, general manager of IBM Information Management, accompanied by special guest Grant Imahara of former Mythbusters fame, and a slew of IBMers led the audience on a systematic journey to debunk three specific misconceptions about information infrastructure one by one.
Myth #1: New technologies will replace everything we know today
Adam Kocoloski, distinguished engineer and chief technology officer for IBM Information Management, and Ritika Gunnar, director of product management for Pure Data Analytics at IBM, took the stage to discuss the first myth and how it applies to six key areas of Information Management:
While it’s true that data warehousing has undergone major changes in the past decade, new offerings from IBM, such as the newly announced PureData for Analytics N3001 “Mako,” defined in the keynote as “an even smarter simple and fast datawarehouse appliance," accommodate new technology, notably with cloud agility. With dashDB comes appliance simplicity, security and powerful in-memory capabilities for high performance data warehousing. This is just one example of new technology not completely replacing current systems, but rather incorporated to help them run more effectively.
NoSQL and Cloud
Adam and Ritika discussed these two technologies together. Cloudant IBM NoSQL databases allow for rapid iterative development and are horizontally scalable and designed for high availability. The keynote described Cloudant as “the only fully-managed NoSQL database-as-a-service.”
It was brought up in the keynote that a lot of people think Hadoop would make existing skills obsolete and even require new ones. However, Adam and Ritika demonstrated an IBM BigSheets iOS app which they defined as “a spreadsheet-like capability in IBM’s Hadoop distribution, InfoSphere BigInsights.” As a result, it is easy to say that this myth is busted, as SQL and spreadsheet skills are valuable when working with Hadoop.
The common misconception brought up in the keynote suggested that “newer technologies will offer mixed workload (transactional and analytical) capabilities that are not offered on current solutions.” With the new DB2 “Cancun” release, however, IBM has merged the thinking that row based data is fastest for transactional data and column data is best for analytics. This release optimizes performance and runs analytics “35x to 73x faster” and with some queries running “more than 1400x faster.”
New technology will replace mainframes; or will it? According to Adam and Ritika, this couldn’t be further from the truth. Citing statistics such as “80 percent of the world’s corporate data resides or originates on mainframes” and “mainframes process roughly 30 billion business transactions per day,” they made the case that mainframes are actually becoming a “strategic platform for business critical analytics, mobile access and private cloud environments.” They also discussed IBM’s System Z and its ability to have accelerated analytics “up to 2000 times” utilizing the IBM DB2 Analytic Accelerator. In addition, Hadoop has come to the mainframe with the launch of BigInsights for Linux on System Z, which promises to allow for the user to “simply point and click on a z/OS data source and the new System Z Connector for Hadoop puts it in to the Hadoop cluster ready to analyze.
Myth #2: Universal access to relevant data—not in my lifetime
Rob Thomas, VP of product development at IBM, took the stage to discuss the second myth. He spoke about the need for a “data refinery,” which is, as he describes, “a facility that ingests raw data, transforms it into clean, relevant data and makes it easily available to everyone who needs it.” IBM’s data refinery solution is a new product called IBM DataWorks. Rob promised that “with IBM DataWorks, you can deliver trusted data as a service, saving time and building confidence to drive your business.” Access to clean and relevant data for everyone who needs it busts this myth wide open.
Myth #3: Even if I could get real-time insights, I couldn’t use them
Ben Snyman, VP of product management for Joy Global, and John Choi, director of product management at IBM, took the stage to discuss the third and final myth. The two had a discussion on sensors and roof collapses at Joy Global mining operations and how IBM solutions could help. John mentioned The IBM Stream Computing platform and how it “continuously integrates and analyzes data in motion to deliver real-time analytics and enhanced, more accurate business models and cognitive systems.” He further went on to say that, “with context analytics built in, IBM enables organizations to detect insights (risks and opportunities) in high velocity data which can only be detected and acted on at a moment’s notice. High velocity flows of data from real-time sources such as the Internet of Things and social data remain largely un-navigated.” This platform allows you to harness that data thereby busting this myth.
Session replays and more with InsightGO
For more in-depth analysis regarding Information Infrastructure tune into InsightGO for a full replay of the Information Management Keynote.