Cloud Hosting

Discover expert insights from Drew Firment, chief cloud strategist at Pluralsight, on optimizing cloud and AI integration for transformative business innovation. Learn essential strategies for success in the ever-evolving tech landscape.

I recently came across an Insider article lamenting techs broken promises: streaming bundles are just as expensive as cable, Uber and Lyft cost as much as taxis, and, notably, the cloud is no longer cost-effective. Ive been hearing this opinion about the cloud for some time now, and in the past year, I have seen large companies like Dropbox and HEY repatriate most (if not all) of their IT workflows back to on-premise solutions.

While cloud computing is an easy scapegoat for techs empty promises, services like Uber and Netflix wouldnt be possible without it. Cloud computing allows me to call an Uber to pick me up at a predetermined location anywhere in the world at the click of a button. Netflix prevents me from walking to my local Blockbuster to rent a movie for a similar cost. How is that not delivering on its value proposition? And yes, big organizations like Dropbox and HEY are moving back to on-prem largely due to massive economies of scale when it comes to storage costsbut the irony is that they wouldve never gotten so large without leveraging cloud computing to get through the early stages of scaling in the first place.

The cause of higher cloud cost is often related to operational inefficiencies tied to supply chain or human capital. When it comes to the cloud, much of the value isnt realized simply because people arent trained to use it efficiently, hence the growing trend of cloud cost optimization, or FinOps.

These same challenges with economies of scale and operational inefficiencies will arise as companies adopt AI. If organizations think cloud computing is expensive, wait until they try running many machine learning algorithms at scale! Most enterprises go through standard cloud adoption patterns, which will be mirrored in AI adoption. So, what can we learn from cloud computing adoption that can help us accelerate the adoption of AI more quickly, with less pain and less cost?

Many companies need help with cloud adoption because they need the foundational purpose, practices, and skill sets to do cloud computing effectively. Without these fundamentals, organizations do the same things with new tools and expect magic. Plus, as companies adopt AI, theyll first need to have mastered the basics of cloud computing and data before seeing AIs actual value. Below are four questions technology leaders should consider when making decisions about their cloud infrastructure and AI adoption.

See More: The Cloud Revolution: Adapting to Changing Realities

One of the biggest mistakes organizations make with cloud computing is using it tactically to drive cost control versus using it as a strategic lever for innovation. Ultimately, cloud computing aims to help companies create customer value. However, most organizations still need help tying cloud strategy to organizational outcomes.

Many businesses need to avoid repeating these mistakes with AI by jumping on the bandwagon without a clear and compelling use case tied to customer value. To avoid going down a path that will fail to deliver meaningful returns to the business and its customers, companies must ensure they know what cloud and AI mean for them in the context of their businesses.

Many enterprises jumped into cloud adoption with the foundational principles of DevOps or Agile firmly entrenched within their organization and culture. And while you can fake Agile, you cant fake cloud computing. As a result, many organizations simply lifted and shifted their apps and legacy practices from on-prem to the cloud. The results? Underwhelming.

Rather than take accountability when outdated practices dont achieve the desired results, many companies blame cloud computing services and threaten repatriation to the comfort of their on-prem caves.

Similarly, before expecting to achieve the desired results with the promise of AI, similar foundations must be firmly established within your organization. For starters, its nearly impossible to do AI without cloud computing, especially given the intensity and scale of GPU (graphics processing unit) requirements. Organizations cant expect to move forward with AI adoption without a solid cloud architecture to serve as their infrastructures foundation. Additionally, its just as important that enterprises establish a strong data foundation that includes collection, cleaning, and distribution. Without a strong backbone of high data quality, the language models generated by AI will reflect garbage in, garbage out.

See More: AI-Powered Cloud Security: More Resilience and Adaptability

After years of enterprises doing lift and shift migrationswherein an exact copy of an application is rehosted from one IT environment to anotherits no surprise that cloud cost optimization, or FinOps, is now trending.

While lift and shift migrations will earn you some early wins, the long-term costs of unoptimized cloud architectures could be more sustainable. Many organizations are now forced to pay the piper with reinvestments in cloud-native architecture combined with efforts to control cloud bills that mimic legacy infrastructure.

And if organizations are struggling with the cost of cloud computing, theyre in for a rude awakening once they let their developers loose to run machine learning models. Given the high cost of AI, it will be crucial for enterprises to learn from cloud computing and follow the best practices for optimizing models and establishing governance. This is the only way to control the consumption of intensive and expensive compute utilization.

Many enterprises needed to see the breadth and depth of the clouds impact on their entire organizations. As a result, companies still need to improve their cloud computing skills today. Recent researchOpens a new window found the largest cloud skills gaps exist in data, analytics, and storage (42%), followed by security and governance (37%).

Clearly, a critical mass of cloud fluency is essential for fostering a sustainable transition to the new operating model. Cloud is a culture, and theres a language to learn to participate. The same goes for AI.

For enterprises to succeed with AI, skills development must start early and often. Workforce development needs to stretch beyond technologists understanding of machine learningit should also include business leaders and other stakeholders in the value chain. Ultimately, the goal is to connect artificial intelligence with a companys most strategic advantage, human intelligence. The only way to do that effectively is through continuous education and upskilling.

The debate surrounding cloud computings effectiveness mirrors the broader discourse on the promises and pitfalls of innovation. Its crucial to recognize that cloud repatriation is not a declaration of the clouds failure but a reflection of the complex nature of getting it right.

Cloud computing came with big promises to transform our lives and undeniably has. Yet, challenges persist due to operational inefficiencies and the need for proper training. This same narrative echoes in artificial intelligence, where similar hurdles mar the road to innovation.

Both cloud and AI require a strategic approach grounded in customer value, foundational principles, cost considerations, and, most importantly, a workforce trained on technical aspects and broader business context. By learning from the clouds journey, organizations can pave a smoother path for the integration of AI, ensuring that these powerful technologies are harnessed to their full potential.

Are you leveraging the full potential of Cloud and AI integration for your business? Let us know on FacebookOpens a new window , XOpens a new window , and LinkedInOpens a new window . Wed love to hear from you!

Image Source: Shutterstock

Read the original here:
How Cloud Can Help Us Master AI - Spiceworks News and Insights