Analytics October 26, 2017 • By Yves Mulkers

Top 3 Takeaways From AnalyticsX – Faster, Better, Harder Integration

3 min read

The right platform allows faster, better, harder integration

In various discussions and interviews with analytics practitioners and SAS experts at Analytics Experience, I was reminded again that having a good platform for your data is of major importance to enterprise grade data management and sustainability. With the upcoming new features in SAS, due to be announced in December, SAS will continue to offer opportunity for collaboration with open source by embracing R and Python. These tools will allow for fast prototyping and turn around.

Integration of the ‘sandbox’ models into a full data lifecycle platform like Viya will be a key strength moving forward. Not only is it an end-to-end platform from data intake to data management, data modelling, analytics modelling, data lineage and visualisation, but compared to the open source tools is it offers an MPP (Massive Parallel Processing) engine.

Do you need a lot of Unicorns or a team with complementary competencies?

There still is a shortage of data science and data management skills. If you look at job advertisements, recruiters are looking for unicorns, and hope to find that one person mastering all the much needed skills and business expertise.

As it came clear from various discussions and presentations, we need a ‘team’ to deliver insights from data. A team that crosses the silos across domains and expertise; business and technical people together in the boat.

A great presentation “The Purple Theory” was the nice “red” wire, explaining the flow in traditional projects – where business takes the requirements, and hands it over to the ‘tech / data science’ team for the modelling and building. At the final step in the chain, the solution is handed back to business where the tech part is put into a business context.

The New Way of Working

In the new set-up, tech people were involved in business discussions, acquiring a better understanding as to how their models are impacting business.

Business people were exposed to tech matters, building better understanding of the difficulties and constraints the tech part is confronted with.

Purple People The Theory

IoT is hotter than hot, and needs the right Analytics power.

SAS IoT Truck Analytics On The EdgeEvent Streaming Processing and ‘edge analytics’ (where analytics are applied directly to data at source), are vital in generating insights from IoT generated data.
SAS has jumped on the IoT bandwagon as it knows its strengths. It has managed to work its MPP platform to handle 400,000 events per second. Partnerships with Intel, who are taking the next step in traditional instruction based processing, bring GPU (Graphical Processing Unit) to the table as the new data crunching workhorse, as well as the recently announced the ‘Neural Compute Engine‘. These strong connections will make analytics key in the IoT era.

 

What we’ve learned

  • Analytics leads to Insights, meaning faster, better decisions that build business value
  • Don’t build your models without also making sure you’re building business value
  • Co-create analytics solutions with business experts
  • Make sure you get management buy-in from the top down
  • Choose the right platform for the right task – i.e. Agile and Quick vs Flexible and Sustainable
  • The joy of data: make people love data by including them in the journey
  • SAS is built to scale, offering an end-to-end solution from data intake over ingestion and presentation, supporting open source and MPP.

Yves Mulkers

Yves Mulkers is the founder of 7wData and a widely followed voice in the data and AI community. He curates the 7wData and AI Beat newsletters, reaching hundreds of thousands of data and AI professionals, and writes on data strategy, analytics, AI, and the evolving data ecosystem.

Related Articles