Subscribe

Becoming more reflective of a digital society

Automation of the datacentre isn’t a new trend, but it’s one that will continue to gain momentum.
Kirsten Doyle
By Kirsten Doyle, ITWeb contributor.
Johannesburg, 11 Feb 2020
Ocea Garriock, Oracle South Africa.
Ocea Garriock, Oracle South Africa.

Datacentres are at the heart of many large organisation’s operations, providing services to internal users and outside customers alike. But they’re under pressure, as they’re evolving rapidly, becoming increasingly complex, and generating gargantuan amounts of data, much of which will reside in the cloud. Not being available 24/7/365 simply isn’t an option for datacentres. Such downtime is hugely costly to the business, so as datacentres become more advanced, the tools that manage them need to be more advanced too.

Enter automation, a trend that’s taking ICTs by storm. As the amount of data grows, and alongside it the speed of business, traditional approaches to managing and monitoring datacentres often results in more alerts that can be dealt with manually. Automation can make these operations practically autonomous, managing and executing virtually all routine workflows and processes within a datacentre, including scheduling, monitoring, maintenance, and application delivery, and it can do this without human intervention. David Odayar, datacentre architecture lead, Comstor, at Westcon-Comstor Sub-Saharan Africa, says the datacentre of the future will be automated. However, while the temptation to let technology drive the business will always remain, decision-makers must focus on how datacentre solutions can complete tasks and deliver results. “Certainly, automation can address many pain points as they relate to the performance of mundane tasks. It eliminates human errors, removes inefficiencies, reduces time, and saves costs.”

In the case of the last, says Odayar, there are inherent time-savings to be had when repetitive tasks are removed from the employees’ core responsibilities, which enables them to focus on delivering more strategic value. “The repeatability of many of these tasks will see automation become more intelligent by maturing its decision-making capabilities over time. Inevitably, it will result in greater confidence in the use of these and other related technologies such as cognitive intelligence.”

Automation, or what we refer to as autonomous systems, introduces a new level of efficiency and reliability to business by utilising the latest advances in computing power, AI and machine learning to self-manage, adds Ocea Garriock, solution engineering director, Oracle South Africa. “This capability essentially eliminates downtime and reputational damage due to lapses in security updates and error-plagued maintenance processes that still require a human clicking a mouse. System self-awareness and proactivity in handling issues – while running – are critical.”

Patching up the gaps

Garriock says research reveals that the vast majority (85%) of security breaches occur where patches were already available, but not yet applied. As autonomous systems automatically apply patches regularly, this security risk is reduced. “In the case of datacentres specifically, embedded autonomous capabilities also mean continual monitoring of workflows and scalable allocation of resources as needed to optimally handle application delivery. In summary, true automation in the datacentre means self-driving, self-securing and self-repairing, while autonomously scaling resources.”

Automation can address many pain points as they relate to the performance of mundane tasks. It eliminates human errors, removes inefficiencies, reduces time, and saves costs.

David Odayar, Westcon-Comstor Sub-Saharan Africa

Datacentre automation is building infrastructure from code, called Immutable Scripted Infrastructure, with which we can do the job of an entire IT department, adds Dave Joyce, MD at Nimble Technologies. “At the touch of a button, we can set up a full datacentre environment in minutes. The main benefits are that the infrastructure is repeatable, there are no mistakes and it’s more cost-effective than a traditional datacentre environment. Essentially, it’s about the repeatable de-risking of building environments in which you deploy industry-critical applications, and it gives you a tighter handle on your datacentre environment.”

We have to look at ways to enable the developers and operations engineers to easily deploy, configure and maintain an application or service within the datacentre, whether this is an on-premise or cloud datacentre, says Robert Graham, technical team lead at Obsidian. “You would want to be able to do this in a trusted and predictable way, utilising tools that will allow you to define your infrastructure as code. With infrastructure as code, you can continuously test your environment before it’s deployed or upgraded. Without infrastructure as code, you stand the risk of creating a snowflake environment, which requires a great number of person-hours to do manual tasks and can lead to human error and fatigue. With automation, you free up your people to focus on building great solutions instead of wasting time and effort on troubleshooting and manually building environments.”

However, when employees hear the words ‘freeing up people’, too often they fear that this means job losses are on the cards. Odayar says this isn’t the case. “With the automation of tasks, staff can be redeployed to areas more critical to the strategic success of the organisation. It also enables the company to upskill employees to reflect the digital requirements of the connected business. In fact, this upskilling will be essential to mitigate against the high potential for job losses in the future.”

He adds that a McKinsey Global Institute study stated that 30% of work tasks across 60% of occupations could be automated, and by 2030, between 400 and 800 million people may need to find new work due to the growth of automation. It estimates that this could result in job losses as high as 44% in many countries. “Despite this, automation will create more jobs than it destroys. Much like the recent emergence of data scientists and other positions, automation will see the emergence of skills and positions previously unimagined.”

The evolving IT professional

Joyce believes that automation will decimate traditional roles within the IT department, and we’re already seeing CIOs’ roles changing quite dramatically. “Currently, about 60% of any CIO in South Africa’s job is to look after infrastructure, but it should be about finding ways of using technology better in the business, rather than running legacy systems. There will be massive job losses. A lot of the traditional roles will disappear, but there will be capacity and ability to use people in more creative ways. So, at least some of the jobs will be replaced by new roles.”

This begs the question, of whether there’s anything that shouldn’t be automated within the datacentre. “There’s very little that shouldn’t be automated,” says Joyce. “There’s no reason to do anything manually anymore. If a server goes out, you could press a button and bring up a new instance of it. Why would you have to log in and try to problem-solve the issue? There’s a whole class of problems that you actually don’t have to deal with anymore. What can’t be automated is the design and philosophy behind it – things like compliance, or your various security stances. Those things have to come from a skilled and knowledgeable individual.”

At the touch of a button, we can set up a full datacentre environment in minutes.

Dave Joyce, Nimble Technologies

Graham adds: “The aim of automation should always be to make your engineer’s life more enjoyable. Saying this, some engineers aren’t for automation as they feel they’re losing control. This dichotomy brings us to the point where one has to decide what’s better for the organisation. Look at the modern datacentre and compare it with the ones from a decade or two ago. Some of the ‘old’ datacentres already had some form of automation by means of robotic arms swapping backup tapes in and out during the daily backup process. Even this form of automation has become redundant due to today's fast network and internet capabilities and more affordable storage. This just shows us that automation within the datacentre will be an ever-evolving process, and as technology will evolve, so will our datacentres.”

Mapping the route

Where should an organisation begin its datacentre automation journey? Firstly, identify a few people who will form part of your automation team, says Graham. “Secondly, identify a small component that can be automated that will give you a quick win. This will build confidence in the team and the rest of the organisation that automation is an achievable option. It can be anything from automating the deployment of infrastructure for a small project utilising a tool like Terraform by Hashicorp or deploying one piece of software such as the backup tool to your development environment with a tool like CHEF, Ansible, Puppet or Saltstack.

“It’s of utmost importance that the automation team works closely with the existing infrastructure and development team to identify other tasks and processes that can be automated within your datacentre. Slowly but surely, more people will start to see the success of automation. From a development point of view, look at DevOps tools that will give your team the ability to continuously test and deploy their code. Tools like Gitlab, Jenkins and Bamboo are great for this.”

In Odayar’s view, a company should consider starting on the infrastructure side. “By automating the processes across the computing, network, and storage layers in physical and virtual environments, the business will start experiencing a new-found agility to produce results faster. There’s no perfect way to embrace automation as companies have their own reasons for adopting it. Some choose to automate because they don’t like to script, others adopt this technology when processes become too cumbersome and complex, and others do so when the transfer of knowledge becomes too challenging.”

Garriock agrees: “Automation can begin at an infrastructure level with automated infrastructure management solutions to monitor both hardware and software. Then cloud automation solutions can provide fast provisioning, monitoring and orchestration. Automation can then move up the stack with the deployment of Autonomous Linux and Autonomous Databases, where real value comes in terms of elimination of administration costs, reduced security risk and increased innovation. Hybrid technologies offer the best of both the cloud and the traditional datacentre. Companies could employ several strategies for a hybrid approach such as renting datacentre space, or moving some applications to cloud IaaS or SaaS, or deploying private cloud.”

For Joyce, the starting point depends on what the organisation’s requirements are. “But you should start with something small that you can get completely automated and hands-free. What will emerge from this is a methodology that will make it easier to bring in other elements. Don’t start with your most mission-critical technology, but rather something that can stand alone – and then get it all automated. If you have something that you’re developing, start automating the process of getting it deployed in a way that’s hands-free and repeatable.”

So what will automation in the datacentre look like in the future? There are many ways automation can be used to change a traditional approach into one that is more reflective of a digital society. The future of the datacentre will therefore hinge on its ability to adopt and employ the right technologies at the right time. These will help shape the not-so-distant future and pave the way for organisational success, says Odayar.

In the future, there will be much more automation than we’re seeing today. “Businesses can’t justify doing things the way they are at the moment. There are companies at the leading edge right now, but in five years’ time, a lot of organisations will be doing what the leaders are doing at present. There’s very little that can’t be automated, so we’ll definitely be seeing more and more datacentre automation. There will be no need to have any physical interaction with the datacentre anymore,” says Joyce.

Graham agrees: “Automation will keep on improving as technology improves. The future datacentre will rely strongly on automation to continuously improve the service a company delivers, whether this is for an on-premise or cloud-based datacentre. The important thing is to start your automation journey as soon as possible, for this will allow you to clear out some technical debt.”

Share