We invited Tonny Bastiaans as the European offering manager SAS on Power System at IBM to publish his view on how to combine hardware and software to make analytics seem. Tonny started more than 20 years ago as an engineer on RS6000 systems (the predecessor of Power). He moved via several roles to his current role. With his technical background, he is bridging the gap between technic and business and is transforming technical value to business value.

So let's hand the word over to Tonny

In modern business, we are seeing more and more use of analytical tools, and the dependency on the outcome of analytics becomes ever bigger. Where in the past analytics was a luxury, now companies are totally reliant on the outcome of the analytics run on a daily or even hourly basis. We see that companies using analytics in an effective way have a competitive advantage with a faster time to market and the ability to foresee and react to trends more quickly.

By effectively using analytics, companies are able to get better insights and make decisions based on data, enabling them to invest money and resources where they make a difference. Modern analytics is also crucial when meeting compliance and regulatory requirements. It is a key asset used to prove that a company is compliant and is essential to staying in business.

For life science companies, Food and Drug Administration and Pharmaceuticals and Medical Devices Agency approvals are essential when bringing new products to the market and maintaining compliance. In the financial sector, analytics is key in many parts of the business. For example, fraud detection is impossible to imagine without analysing a huge amount of data. And a fast, reliable analysis of transactions is important for the success of your fraud detection system. Noncompliance in both of these sectors can lead to very high fines.

Analytics needs a sound foundation

These examples show the importance of fast, reliable and always available analytics. But if we look at how modern companies are deploying analytics, we often see that the focus is on the analytics tools and not on the underlying hardware.

The mantra “good is good enough” is often heard when talking about the infrastructure. But if we have a deeper conversation with these customers, we see that these are the companies struggling with issues around their analytics platforms. For example, the performance and throughput of their analytics platforms are often insufficient for their needs. They try to solve this problem by adding more hardware. But the result is that the throughput does not increase, and the performance issue remains. They have only created a bigger problem by having more servers to manage.

These companies must also react to changing business demands. If the workload grows beyond their existing capacity, they must endure a drawn-out process of buying, installing and deploying new hardware, processes which can take weeks or even months. And the business demand can totally change during that time.

Finally, companies are struggling with the reliability of their hardware. This not only includes the traditional reliability associated with failing hardware components and the need to repair them, but also reliability from a security point of view. Ransomware and other attacks by hackers are becoming more and more frequent and a real concern. So the security of the IT environment requires more and more focus and is therefore high on the company agenda, particularly for those working with a lot of confidential or competitive, sensitive data.

Infrastructure for performance

IBM Power Servers solve these issues by delivering an infrastructure that provides the necessary performance. Next to that, a flexible solution is important. And given that IBM also designs its servers with security and reliability in mind, they deliver one of the most secure infrastructures on the market, with five-nines uptime. Also, if we look at the ModelOps framework developed by SAS, we see the same importance for infrastructure. ModelOps is how analytical models cycle from the data science team to the IT production team in a regular cadence of deployment and updates.

Considering all of the facts above, we can state that Power servers deliver a foundation for analytic workloads that results in more effective data scientists, more and faster data research, and a secure and reliable environment. And we at IBM are not alone in this story that Power Servers deliver the best foundation. SAS is underpinning this powerful combination of SAS and IBM infrastructure to optimize the analytics life cycle. With this foundation in place, you have the perfect start to develop your analytics in a way that helps you do great things and have that competitive advantage you are looking for.