Want more Insights from SAS? Subscribe to our Insights newsletter. Or check back often to get more insights on the topics you care about, including analytics, big data, data management, marketing, and risk & fraud.
A brave new world of analytics
SAS® Viya™ combines flexibility with control to power next-generation analytics
by Daniel Teachey, SAS Insights Editor
The analytics world is changing at a breakneck pace. Whether it’s big data, the rise of data scientists or dozens of other factors, analytics professionals are scrambling to keep up.
Mike Frost, Senior Manager of Product Management for the Cloud and Infrastructure team at SAS, walked us through the new world of data that organizations are facing – and how the new SAS® Viya™ architecture can help prepare you for what’s next.
What kinds of analytics challenges are organizations facing today?
The challenges fall into three areas. First, it’s about resources. Organizations simply don’t have enough people who can manage, manipulate and analyze the data that they accumulate. A key group that does this is known as data scientists. There are other professionals that perform these tasks as well, sometimes identified as citizen data scientists, statisticians, business analysts or analytics domain experts. No matter what you call them, their analytics skills and knowledge make them valuable – and difficult for organizations to attract, hire and retain. Keeping them happy often means letting them try out cool new technologies to stimulate the kind of creativity that generates new innovation.
Mike Frost, Senior Manager of Product Management for Cloud and Infrastructure, SAS
The next challenge comes from the types of data that organizations are dealing with. This goes well beyond a “big data problem” to a “bigger than big data problem.” Data is now stored in a variety of architectures, both on- and off-premises. Some of that data (but not all) is under the direct control of the organization. Data volumes continue to accelerate, driven by streaming data. Data lakes established to handle all of this data are starting to overflow. And with the demand for analysis to be performed at a sub-second rate, landing the data to analyze it is a non-starter. All of this adds up to a complex environment from an IT perspective that is getting more complex all the time.
Finally, the sheer importance of analytics is elevated. Disruptive, forward-thinking organizations often view themselves as analytics-based. They consider the value of their analyzed data as the strategic asset. This creates new challenges, however. For example, data-driven companies must be transparent about what they do with their data to avoid a perceived violation of privacy. Shielding the organization from risks means treating both the data and the analytics environment as the most valuable resources the organization has. That requires a high degree of security and governance.
You mentioned a very complex data landscape that is “bigger than big data” and what that looks like. How does that affect analytics?
We see that demand for analytics tends to start small and get bigger (and bigger) more rapidly than organizations can adapt. For example, it’s common for people who start getting answers from analytics to ask more and more questions and for those questions to quickly drown out the answers. You can quickly get to some interesting questions, like “How many more customers could we process if we shaved operations 10 seconds?” That simple question leads to a host of other questions as analysts try to find the best path for the organization.
All of this curiosity can lead to an exponential increase in the analytics at play. One week it’s 100 models, the next week it’s 500 models. On the data side, one week it’s gigabytes of data to analyze, the next week it’s petabytes. There’s also a need to incorporate the latest techniques in machine learning, cognitive computing and other advanced methods. You have to integrate data housed in traditional database structures, distributed structures like Hadoop, or even data that is in motion. Soon, organizations aren’t sure how they will make it to the end of the year, let alone plan for next year.
It becomes overwhelming if the infrastructure isn’t in place to handle diversity within the data environment and facilitate a rapid scale-up. One way that we see organizations trying to address this is to look for ways to support all of their analytic assets from a single analytic environment that can be managed, maintained and scaled as needed.
How do you feel that SAS can help customers address these needs?
This is what we do, and we’ve been doing it for a long time. SAS has been solving problems like these for 40 years now. SAS Viya is coming out at the right time to help us solve the next wave of our customers' challenges. It will serve as the foundation of the next 40 years of service to our customer base.
Is SAS Viya meant to replace customers’ existing technology investments in SAS?
The short answer: no. SAS Viya integrates with and complements existing customer investments in SAS technologies but does not replace or require additional software from SAS.
Finally, where can folks go to learn more about SAS Viya?
The SAS Viya page has more about our offerings on SAS Viya. Learn more about the new technologies and sign up to participate in our Early Preview program to try out the software before it is released. We’d love your feedback and will use it to make sure we continue to improve SAS Viya.
- Exploring the sun with big data
- Top 10 best practices for next-gen analytics
- Sensing a disturbance in the data