CloudOps is critical to cloud adoption. It is no longer an afterthought rather the front and centre of enterprise strategy. We talk to industry experts to discuss trends, challenges, and the future of CloudOps
Organisations are moving to the cloud in a big way, hybrid or multi-cloud. But the secret sauce to success is managing complexity, which is setting up an action plan for long-term cloud operations – from problem management to monitoring, security, and performance – an effective CloudOps strategy.
“As with other IT projects, successfully implementing CloudOps requires careful attention to three pillars – people, processes, and technology. From a technical perspective, two components need to be considered. First, the cloud management platform must enable the automation of continuous operations and facilitate autoscaling through self-provisioning. Equally important, you must have unified observability tools that deliver visibility into all transactions in the cloud and integrate tightly with Cloud Management Platform via the actionable insights,” says Charbel Khneisser, Vice President Solutions Engineering – EMEA, Riverbed.
While most organisations that migrated to cloud-based platforms talk about using artificial intelligence and machine learning, and big data, they do not focus on cloud operations. Bridging the Cloud Transformaton Gap report found that 62 per cent cited complexity and abundance of choice as a hindrance when planning a digital transformation that leverages the cloud.
In an AppDynamics report, Agents of Transformation 2021: The Rise of Full-Stack Observability, 86 per cent of technologists in the UAE (compared to 77 per cent globally) are experiencing greater levels of complexity as a result of the acceleration of cloud computing initiatives caused by their company’s response to the pandemic.
The biggest challenge affecting the future of CloudOps is skills and talent, says Fred Crehan, Regional Director – Middle East, Confluent. “The massively competitive market for talent is not going to change, and supply issues will continue. This is especially true in the Middle East where a survey of CEOs by PwC found that 70 per cent said the availability of essential digital skills already poses a threat to business.”
There is far more to cloud operations than many companies realise. Successful cloud migrations require a detailed understanding of the current capabilities of cloud resources rather than trying to adapt existing operational approaches to fit into the cloud platform, says Gregg Ostrowski, Executive CTO, AppDynamics.
“It’s important to introduce best practices that streamline shared tasks, monitoring, and management between physical and cloud-based systems to avoid the two teams becoming more siloed.”
We talk to industry experts Khneisser, Ostrowski, and Crehan to discuss trends, challenges, and the future of CloudOps.
Excerpts from the interview:
What are trends and challenges affecting the future of CloudOps?
Khneisser: CloudOps is a relatively new term for CIOs. While it is the extension of the principles of ITOps into the cloud, it isn’t limited to just this. It is the unification of processes and ITOps into cloud environments, with the intent of better serving business users. This lack of awareness of this wider scope of CloudOps has resulted in adoption being hampered by a lack of specific skill sets.
Operating cloud environments is fundamentally different from managing on-prem systems. ITOps skills don’t translate directly over to the cloud, and similarly, the most common cloud qualifications – such as cloud architect – don’t typically cover the foundational ITOps skill sets.Budget is another challenge as the move to the cloud requires a shift in mindset from CapEx-based planning to consumption-based planning. And finally, security threats specific to cloud deployments need to be considered when leveraging CloudOps.
CloudOps delivers more visibility into the cloud, and the flip side of collecting more data is that you could run the risk of exposing this to malicious actors. Of course, organisations that handle this well will stand to strengthen their security postures.
Ostrowski: CloudOps is the culmination of DevOps and traditional ITOps applied to cloud-based architectures. Successful cloud migrations require a detailed understanding of the current capabilities of cloud resources rather than trying to adapt existing operational approaches to fit into the cloud platform. However, transitioning to serverless functions comes with its own set of obstacles. When the underlying infrastructure is based on traditional operations, it can be a challenge for CloudOps to merge seamlessly into existing DevOps systems. It’s important to introduce best practices that streamline shared tasks, monitoring, and management between physical and cloud-based systems to avoid the two teams becoming more siloed.
With an on-site data centre, there is a physical server to be worked on. CloudOps relies on dashboards and tools that automate much of the monitoring process. It also means that alerting and monitoring need to be cloud-optimised.
A successful CloudOps transformation requires a company-wide culture shift where every aspect of the organisation is invested in using the same tools, following the same practices, and working together under the same leadership.
The benefits of cloud services can bring great advancement to roll out new capabilities and scalability during peak usage of your digital services however managing the cost of cloud spending can also be required. Ensuring workloads are adequately sized to run optimally and scale when needed also needs to be managed to ensure costs align with the desired business outcomes.
Crehan: The biggest challenge affecting the future of CloudOps is the same for the whole tech industry – skills and talent. The massively competitive market for talent will not change, and supply issues will continue. This is especially true in the Middle East, where a survey of CEOs by PwC found that 70 per cent said the availability of essential digital skills already poses a threat to business.
Enterprises talk about how they need to lower the dependency on unique technical skills and deploy them to the highest value tasks. This means engineers are not managing data infrastructure when they could be building and focusing on CloudOps, like creating standard procedures about how things should operate or designing new migrated systems.
The other big challenge is the rise of security threats; cybersecurity attacks continue to make headlines daily, and experts are continuously searching for trends to help better predict and prepare for the next attack. UAE organisations would be well advised to address the security concerns around CloudOps as cybersecurity failure is ranked as the greatest risk to the country, according to the World Economic Forum’s The Global Risks Report 2022.
Are most enterprises clueless as to how to properly implement CloudOps?
Khneisser: As with other IT projects, implementing CloudOps requires careful attention to three pillars – people, processes, and technology. From a technical perspective, two components need to be considered. First, the cloud management platform must enable the automation of continuous operations and facilitate autoscaling through self-provisioning. Equally important, you must have unified observability tools that deliver visibility into all transactions in the cloud and integrate tightly with the Cloud Management Platform via actionable insights.
Ostrowski: Different organisations are at different stages of their cloud journey, and therefore the sophistication of CloudOps within IT teams will vary. However, what is consistent across most enterprises is the phenomenal increase in cloud deployment over the last two years and the subsequent challenges this has caused. CloudOps can play an important part in getting technologists’ visibility across their entire IT estate.
Crehan: In terms of adequately implementing CloudOps, one of the key discussion points is how to manage hybrid cloud strategies. While the pandemic truly shifted everything into high gear, accelerating the shift to the cloud even more, as companies overnight had to adopt an entirely new way of working to ensure business continuity, most businesses are still utilising a combination of on-prem and cloud environments. To get the hybrid cloud right, enterprises need to understand that front and centre facilitates the data exchange between on-prem infrastructure and the cloud. Yet, this is what is often overlooked by many enterprises.
Many new applications might be built in the cloud, but they still mostly need to connect to historical data sitting in data centres, and major issues arise when data isn’t shared efficiently between different systems. Historically, the approach to resolving this challenge was to introduce a direct linkage between systems that need to communicate. While this may sound like the answer, it is typically only manageable for a limited number of links and chaotic when scaled up to every office and home.
Ultimately, solving the data exchange problem requires a new approach and one that can successfully evolve and optimise hybrid cloud models over time. That innovation lies in handling data in a new way – one that supports collecting a continuous flow of data from across the business, between apps, databases, SaaS layers, and cloud providers. Like the human body’s nervous system, there needs to be a data platform and instant link between all of a company’s footprints – one that ties together different parts of the business, uniting all the applications into a coherent network that can respond intelligently in real-time.
What are some of the best practices to help organisations get started with CloudOps?
Ostrowski: A successful CloudOps strategy isn’t a one-size-fits-all approach and can’t be accomplished in a single step. The ultimate goal is to integrate several teams with different focuses and skillsets into one fully integrated operation. Achieving that requires an assessment of existing strengths and weaknesses among team members and communication methods and decisions about leadership, tools, processes, and budget management.
It will help if you create a plan for not only the initial transitional period but also for identifying future optimisation opportunities. Many organisations are creating new roles for CloudOps and cloud infrastructure management; new opportunities may arise depending on the skills assessments.
And it’s also about creating that culture of collaboration, letting people work at their pace towards a common goal but in a structured environment. If teams fail to work together and make usage more complex, there will be less transparency and a lack of overall cohesiveness. You don’t want to create more restrictions. Instead, clearly define and implement necessary guidelines.
What’s the future of CloudOps?
Khneisser: The momentum we see around CloudOps is that it addresses two primary challenges organisations face with the cloud – control and visibility. Today, the cloud will no longer be a choice but rather a necessary component of the IT environment of the modern enterprise. Cloud delivers scalability and agility and paves the way for organisations to adopt the latest technologies such as AI. With CloudOps being critical to cloud adoption, in the future, we will see it being leveraged to provide more automation, analytics, machine learning, and AI capabilities and thus facilitate the self-provisioning and self-healing of cloud environments.
Ostrowski: The rapid digital transformation that occurred as organisations responded to the pandemic shows no relenting. Instead, IT teams need to adjust to the new normal of rapid innovation across their organisation. Cloud will play a huge part in this as IT teams rely on the cloud to scale their operations at pace. And CloudOps will be an important part of this.
But unless they find a way to gain greater visibility and insight into all IT environments and cut through ‘data noise,’ they will be unable to drive the rapid, sustainable digital transformation their organisations need. Without genuine full-stack observability, technologists don’t stand a chance of being able to quickly identify and fix technology issues before they impact end-users and the business.
Technologists need to be able to monitor all technical areas across their IT stack, including within cloud environments and directly link technology performance to the end-user experience and business outcomes to prioritise actions and focus on what matters to the business.
Crehan: The future of CloudOps is clear as studies estimate worldwide end-user spending on public cloud services will grow 23.1 per cent this year to $332.3 billion. However, enterprises need to clearly understand how they can best optimise their capabilities to boost growth and revenue and know how they can overcome and be prepared for the challenges and complexities that the cloud brings. This requires a new mindset to overhaul how they think about data and how they can set it into motion and ensure CloudOps is no longer an afterthought but the front and centre of enterprise strategy.
If you liked reading this, you might like our other stories
Small Is The New Big
Can Edge AI Power Modern Data Analytics?
