The all-rounder

Mr. Pleier, the current crisis has hit the economy hard, all over the world. What do you think is the most important lesson for corporate IT?

Markus Pleier: This crisis and its dimensions took us all by surprise. For such an event, it is not possible to develop plans in advance that have been formulated down to the smallest detail, which then only have to be pulled out of the drawer in the event of an incident and which cover all eventualities.

That is simply not possible. Nevertheless, something can always be done better, of course. If there is a general lesson to be learned, it can be summed up in one term: Resilience.

I would rather have expected the word digitization.

Pleier: Increased digitization is certainly one of the important and correct answers to the question of how companies and entire economies can best emerge from the crisis and return to their growth path.

But the fragility of our previous structures, which was revealed in the crisis, must first be eliminated. We need a new stable basis on which to build, on which we can then proceed with digitization.

But we have to be careful and clear up a misunderstanding from the outset: Stability would be the wrong term if it meant rigid structures. That is why I prefer to speak of resilience.

Were our previous structures too rigid?

Pleier: Yes, and this is true for many areas. A tree that bends in the wind does not break and at most loses a few branches. Supply chains that were not branched out and regionally diversified enough suffered the most from the crisis.

For the single strand in the chain was not a flexible branch or trunk, but a column of concrete that was bound to break when shaken, causing the entire building to collapse.

We have seen this in Germany, for example, with protective equipment for medical staff. Companies that did not have sufficient capacity and flexibility to set up high-performance home office workstations had to contend with productivity losses and increased security risks, as less effective encryption requires less computing power.

Or they had to move to the public cloud, perhaps violating internal or external policies and regulations.

What could the companies have done differently?

Pleier: Resilience means withstanding external shocks without the core breaking, while parts may well be temporarily impaired or even fail.

So it's all about flexibility, redundancy and resilience, in the sense that partial failures don't put the whole at risk and, moreover, the level of resilience doesn't drop.

It also means that failed components can be repaired or replaced elsewhere, preferably automatically. Regionally diversified supply chains could stand in for each other so that total failure is avoided.

IT environments that are non-dependent, distributed, and redundant can flexibly control and adjust resource allocation for different workloads so that they can always fulfill their core mission of supporting the respective business priorities.

With the right infrastructure, enterprises can provide stability that is also flexible, resilience and redundancy that is also affordable, extensibility and scalability that prevents new dependencies and vulnerabilities.

That sounds very challenging.

Pleier: That is correct. But above all, I want to encourage IT to learn the right lessons. After all, crisis is not synonymous with chaos, collapse and finger-pointing.

Crisis, in the original sense of the word, means a moment of decision, admittedly not an unimportant, incidental one that has no major impact. Rather, it is a decision that leads to a more stable future, which in turn is built on a more resilient foundation.

They demand long-term thinking. However, companies currently tend to have short-term priorities.

Pleier: You are right about that. And of course that's the way it has to be. But the decision-makers also know that short-term priorities must not be at the expense of long-term goals.

What good are short-term cost savings if they take away from true digitization that leads to more automation and productivity gains? From a business perspective, the best way to save has always been to invest wisely.

That's what SAP's existing customers had in mind with the transformation to S/4! Now they have postponed this important digitization project for the time being.

Pleier: That's all too understandable. In fact, the IT teams have other worries at the moment. After all, transformation is a project that takes time and ties up a lot of human resources.

Right now, the priority is called teleworking, which has come to stay in the crisis. At the same time, IT departments must do their part to reduce costs.

But this brings us back to the moment of decision already mentioned: Are these short-term projects and activities planned and realized in such a way that they lay the foundation today for subsequent long-term initiatives and even simplify and accelerate them?

What's your guess?

Pleier: If the crisis shows one thing, it's that infrastructure matters. In good times, we don't notice it, it's there and it works somehow, but we don't really take any notice of it.

That is then a matter for the infrastructure teams. The fact that we become aware of the concrete design of an infrastructure beyond the circle of specialists is the consequence of an exceptional situation.

To take a positive example beyond IT: Our healthcare system has held up. It was decentralized and flexible enough to further increase the already impressive number of intensive care beds in a short time.

This has enabled the functioning of the overall system to be maintained at all times. Overall, the healthcare system has proven to be remarkably resilient.

And how can this now be applied to IT?

Pleier: In everything that IT teams do and plan to do in the long term, they should make sure that their infrastructure is resilient. Because only if it is, short-term priorities and effects can be kept in line with long-term strategic goals. What's more, they can even fuel them.

Could you please be more specific about that?

Pleier: Very much so. Expanding the number of home workstations is not only a challenge in terms of performance, but also a question of security. Patches must therefore be applied as quickly as possible.

In traditional infrastructure environments, this is not so easy. Due to the large number of dependencies on a specific hardware driver or operating system version of a storage resource, for example, the intervals between individual updates can be very long, in many cases too long to thwart the increasingly sophisticated attacks of cybercriminals.

Wouldn't it be better to have an infrastructure that automatically applies security updates at the push of a button?

But isn't that a specific problem that can be solved in the short term and has no particular long-term impact?

Pleier: I don't mean. I deliberately chose the example of updates because it's a constant construction site for SAP teams. Of course, an infrastructure for telecommuting can be set up relatively quickly, without a connection to the SAP environment and probably by teams other than those responsible for the SAP applications and databases.

But here is the moment of decision: Wouldn't now be the time to introduce an infrastructure that not only solves the update problem for the home office, but also for the central SAP environment?

Perhaps even a common infrastructure for a wide variety of workloads, including SAP? One that works for SAP NetWeaver and AnyDB landscapes as well as for SAP Hana environments?

That sounds a bit like the egg-laying jack-in-the-box ...

Pleier: You're not wrong there. But please don't forget: A revolution took place in the infrastructure sector years ago.

Since then, infrastructure has been software, the direct link to hardware has been cut, creating two separate layers and merging the individual infrastructure components such as computing power, storage, network and hypervisor into functionalities of one and the same software platform.

Many companies and IT teams have not really noticed this until today because the revolution happened in the data centers of the cloud providers.

Isn't the public cloud then the resilient component in the infrastructure mix?

Pleier: Hyperscalers have certainly helped many companies a great deal in the current crisis to overcome internal bottlenecks. The short-term effect is certainly extremely positive.

But in the long term, this creates new complexities and dependencies. Different and incompatible technology stacks require more specialized knowledge and more coordination, thus leading to more complexity and new dependencies.

New vulnerabilities arise that endanger the functioning of the overall system. Just think what happens if a data cable is accidentally cut. That's not what I call resilient.

What would be an alternative?

Pleier: The answer is obvious: bring the cloud to your data center! This is achieved by means of infrastructure software that is completely independent of the underlying hardware, can be managed centrally and automatically performs all administrative tasks at the push of a button.

It is fully virtualized and redundant. If a hardware component or a virtual machine fails, this does not lead to interruptions at the application level, nor is any data lost.

A wide variety of workloads can then be run on such a software platform: SAP applications, NetWeaver middleware, databases from various vendors including Sybase, Microsoft SQL Server, DB2, PostgreSQL, MySQL and Oracle, of course Hana, test, development, quality assurance and production environments for SAP, VDI, third-party applications, etc.

From the point of view of the SAP teams, however, this contradicts pure doctrine. Many are also likely to have doubts about whether an SAP production system runs with sufficient performance in a virtualized environment.

Pleier: We are aware of these concerns, but they are unfounded. By keeping the data accessed via a virtual machine in close proximity to it, and in memory to boot, we minimize network traffic. Bare Metal is probably no faster.

We call this data locality, which is just one of the principles built into our hyperconverged infrastructure software, so to speak (see Bas Raayman's article on page 46). Our software platform brings its own hypervisor right along with it.

AHV is included in our licenses and subscriptions at no extra charge and is certified by SAP for SAP environments based on NetWeaver and Hana. Customers can be confident that their Suite 7, as well as S/4, will run smoothly on Nutanix. It also saves masses of licensing costs.

But apart from performance and reliability, the cloud has other advantages to offer. What does that look like?

Pleier: Let's start with the topic of scalability. Our software predicts resource requirements and sounds the alarm when bottlenecks are imminent. Then nodes, i.e. combinations of memory, CPU and network, can simply be added.

We support a very wide range of manufacturers for the necessary hardware. These include Dell EMC, Lenovo, Cisco, HPE and Fujitsu. In addition, we also offer the nodes in a subscription model, so that no upfront investment is required.

In this way, financial expenditure can be shifted from capital to operating costs. This is also common practice in the cloud. Finally, we offer numerous seamless connections to hyperscalers, should peak loads only be cushioned with the help of external resources.

This is a true hybrid approach that removes the boundaries between the technology stacks and makes them invisible to IT because all the necessary adjustments, for example, when moving a workload from your own data center to the public cloud, AWS, GCP or Microsoft Azure, are handled automatically by our software.

Doesn't that then lead to new dependencies again, this time on your software?

Pleier: There is a difference between usage and dependency. Neither the underlying hardware nor the workloads are hardwired to our software.

You can port your environment to other platforms at any time. Our customers not only have greater degrees of freedom in their choice of hardware, but also in their choice of hypervisor and public cloud.

In addition, the licenses and subscriptions that customers purchase from us are also portable, i.e. from their own data center to the cloud and back. We offer these degrees of freedom and choice for as long as customers use our software.

Let's return to the aspect of short-term versus long-term. What does your software offer SAP existing customers in this respect?

Pleier: With the help of our software platform, companies can realize digital workplaces in large numbers from order to deployment within a few days.

However, the same platform can also be used to run a Hana database, a NetWeaver environment or SAP Business Suite and S/4. Our platform is certified by SAP for all these environments.

But the most important thing is the automation of administrative tasks and the support of the DevOps concept. That's what matters most in the long run.

Why is this so important?

Pleier: If digitization is to succeed, IT must become more agile. The principle of "never touch a running system" can then no longer apply, and wasting time on routine tasks must come to an end.

Companies are restructuring, buying and selling in response to the current crisis, but also to technological innovation and the resulting competition from start-ups.

It cannot be that it takes weeks or months until the necessary infrastructure is ready to integrate the application landscape of purchased companies or to provide new database environments. Developers can no longer be hampered in their work and productivity because the operations team can't keep up.

Our platform, for example, can be used to create blueprints for SAP development and test environments. Developers can then click on such a blueprint in the self-service model and the environment is ready in seconds or minutes.

In addition, data from the production system can be cloned within minutes for test purposes. This increases the reliability of test results by whole orders of magnitude. This can trigger a quantum leap in productivity.

However, many existing SAP customers are quite happy with their running environments. Why should they port them to your platform?

Pleier: For two reasons. First, they benefit from greater agility and productivity thanks to increased automation. Second, they reduce their total cost of ownership.

They can save on licenses, power and cooling, space requirements and administration costs; overall, infrastructure costs are reduced by up to 60 percent, according to one study.

Customers can also deploy a complete SAP landscape with development, test, QA and production systems for a new business unit or acquired subsidiary at the push of a button.

You can migrate faster from existing databases to SAP Hana. You can consolidate and harmonize existing landscapes more easily in preparation for the big transformation to S/4.

Furthermore, they can apply updates at the push of a button without having to worry about dependencies at the infrastructure level or having to sacrifice weekends for planned system downtimes.

And: We offer a "back" button, should something not work as desired or planned.

Don't you provoke the resistance of many specialists precisely because of these simplifications?

Pleier: Of course, we always have to convince our customers. Yes, we make them less dependent on specialized knowledge and specialists, but we do not devalue their skills and know-how.

Because everything our software does can be controlled via command lines and viewed at the code level. So IT teams never give up control or have to trust some kind of black box.

But we radically simplify daily work, even for non-specialists. In this way, we increase the resilience of the entire IT. If a specialist for this or that hardware or database is on vacation or sick, this does not lead to painful downtime with the help of our platform. But this is not the end of the resilience argument.

But?

Pleier: Let's not think in the short term, but in the medium and long term. In the next ten years, a large proportion of the baby boomers will retire. It will be difficult, if not impossible, to replace all the positions that become vacant with young colleagues.

We also need our IT professionals today and in the future, especially for application development and support. We can only counter demographic change through higher productivity and higher value-added contributions from the next generation if we want to maintain the growth prospects of companies in the long term and secure our prosperity as a whole. Our software can also make a contribution to this.

Thank you very much for the interview.