SDDC could be part of the answer, if only we can get there
Working in a data centre or computer room in a large enterprise is not a job for those seeking peace and tranquillity. This is confirmed in a recent report by Freeform Dynamics which unsurprisingly showed that only one in five respondents to a European survey were experiencing a reduction in pressure on their IT infrastructure and operational processes. Almost half said such pressures were growing steadily, while nearly a third reported the pressure was growing rapidly. With new service demands from business a regular requirement, budgets tight, and skilled staff often difficult to find or keep, the case for automation in many data centre systems and operations appears to be almost iron-clad, but is it taking place?
The survey, which targeted IT professionals in organisations with over 2,500 staff, found that the need for automating IT tasks is now very widely accepted in a number of key areas (Figure 1).
The responses show that automation is now seen by between two-thirds and three-quarters of those surveyed as being worthwhile or highly valuable to help alleviate operations overheads and delays. This marks something of a shift in attitudes over the past decade as the underlying technologies that drive IT automation have evolved. But while automating IT processes is widely recognised as valuable, the survey also indicates that the penetration of automation into IT operations still lags some way behind (Figure 2).
The technology to automate individual components of the core IT infrastructure has been available for some time, and the management tools to handle routine operations have developed rapidly, especially to keep up with the deployment of virtualised systems. But the research highlights several reasons for the relatively low penetration of automation into mainstream operations (Figure 3).
Top of the practical considerations that limit the use of automation comes the very complexity that many organisations find in their data centres. While automating any individual element might appear straightforward in isolation, most organisations have complex assembled systems with differences baked in, even when they are based on similar underlying kit.
Unless you are a web scale service provider, it is likely that your data centre will be running multiple sets of hardware, perhaps with different operating systems, or at least at divergent release levels. And these in turn will be hosting a diverse portfolio of applications and services, each with a host of valid business reasons for why they are more important than everything else.
Add the fact that there is still a lot of uncertainty about how mission-capable the technology actually is, and it is little wonder that automation is still very early in its adoption. And then there are all the usual challenges that face IT such as lacking the time, resources and budget to act, and of course always having “higher priority” issues to tackle first.
With strong enough business drivers, these can be tackled. But there are human concerns that have been burned into data centre professionals’ bones over decades. These include not trusting that advanced solutions are ready for production deployment until they have been very extensively tested, and these could prove to be more subtle and more difficult obstacles to conquer.
It will be interesting to see how quickly Software Defined Data Centre (SDDC) solutions can overcome these inhibitions. SDDC is one of the newer technology initiatives developed by infrastructure vendors to help make IT systems more responsive and better able to manage the business pressures, and the early signs are good (Figure 4).
The survey results show that almost a quarter of organisations are already adopting SDDC solutions, while similar numbers have already built it into their plans. Indeed, only one in ten currently think that SDDC will probably not be useful to them. Taken together these results look very promising, but it is important to add a couple of caveats.
The first concerns the fact that SDDC is still very early in its development, and while one in four organisations may be adopting it (or elements of it), the clear majority are still not using it across their entire data centre estate. Instead the growth in usage is likely to follow well-established patterns – use and test it in well-defined, controlled environments first, then extend usage only when you are happy it really works and know where you can trust it to function effectively. This avoids undue risk and, just as importantly, it allows standards to mature to support very complex environments and gives data centre managers time to modify key operational processes to keep the SDDC functioning over years.
The second is that SDDC has rapidly become one of those IT terms that has been taken over by vendor marketing budgets and is now used to describe an eclectic, wildly diverse range of products, some of them real and others mere vapourware. In some ways the term has become almost as badly abused as “Cloud” was a decade ago. The level of noise that SDDC is creating amongst business executives as they read potentially over-simplified marketing material makes the job of data centre professionals even more challenging.
The Bottom Line
The automation of data centres is a work in progress, but it is one that is now advancing on several fronts, from the underlying systems hardware to management frameworks and stacks. The adoption of SDDC may even prove to be swifter than traditional experience of advanced technologies may suggest. Many of the foundations of SDDC can be found in software stacks that already enjoy very widespread adoption in data centres, most notably VMware Cloud Foundation, Microsoft Azure and OpenStack.
Article originally published on DCS UK
Tony is an IT operations guru. As an ex-IT manager with an insatiable thirst for knowledge, his extensive vendor briefing agenda makes him one of the most well informed analysts in the industry, particularly on the diversity of solutions and approaches available to tackle key operational requirements. If you are a vendor talking about a new offering, be very careful about describing it to Tony as ‘unique’, because if it isn’t, he’ll probably know.
Have You Read This?
Generative AI Checkpoint
From Barcode Scanning to Smart Data Capture
Beyond the Barcode: Smart Data Capture
The Evolving Role of Converged Infrastructure in Modern IT
Evaluating the Potential of Hyper-Converged Storage
Kubernetes as an enterprise multi-cloud enabler
A CX perspective on the Contact Centre
Automation of SAP Master Data Management