What’s in store for the cloud in 2024? We look at three trends that give us a good idea.

TL;DR

Managing cloud spend is a top challenge, overcoming security for the first time in years.
Gen AI is driving the cloud industry into new avenues.
Rising cloud costs make organizations think twice before migrating.

Keep reading to find out what happened in the cloud in 2023 and what this tells us about the coming year.

💸 Paying A Fair Price In The Cloud

The cloud industry was no stranger to the whirlwind of the energy crisis, inflation, and supply chain issues in 2023. There’s no denying: cloud services have become more expensive, yet companies are still wasting them.

Top data points:

Spot instances have become more expensive, as shown in CAST AI research and another study.
The US Producer Price Index (PPI) showed that while the cost of computers and servers is dropping by 3.9% month-over-month, cloud services are getting more expensive, by 2.3% since Q3 2022.
Azure pricing for all cloud services increased by 9-15% in the UK and Europe from April 2023.
Gartner forecasted worldwide spending on public cloud services in 2023 to reach $591 billion, a 20.7% uptick from 2022’s $490 billion.
82% of respondents to a recent survey said that their top cloud challenge is managing cloud spend. Self-estimated wasted cloud spend was 28%.

Over half of corporate cloud spend commitments are unused, amounting to more than $300 billion in untapped credits, according to Infosys.

🤔 To Cloud Or Not To Cloud

The rapidly growing cloud costs and uncontrollable waste have contributed to a lively discussion about how the cloud is no longer a slam dunk. It’s also become clear that companies should think twice before migrating applications to the cloud without preparing them first.

Top data points:

X shifted its workloads to on-prem and reduced the monthly cloud costs by 60%, DHH announced. But what chunk of these cost savings has now become CapEx instead of OpeEx? And what about the salaries of all the on-prem experts?

On the other hand, giants like Uber are moving to the cloud, as the Google Cloud CEO publicly confirmed the signing of a long-term contract with Uber.

🤖 The Impact Of Gen AI On The Cloud

ChatGPT accelerated the growth of AI 10x last year, prompting companies both large and small to embark on a wild goose chase for GPU chips.

Combine this sudden spike in demand with supply chain issues, and you get yourself into a situation where people take quite extraordinary measures to get their hands on GPU.

Top data points:

a16z has seen companies spend more than 80% of their total capital raised on compute resources.
Gen AI budgets can grow 3.4x larger than anticipated. The bill for training and running LLMs can reach $700,000 a day, as OpenAI’s infrastructure is on Azure.
Gen AI data center and infrastructure costs are projected to reach $76 billion by 2028.
Tech giants are pouring billions of dollars into building gen AI capabilities and ensuring the availability of cloud capacity (they spent $42 billion in Q3 2023 alone).

Let’s not forget about the environmental cost of running LLMs.

👇 Brought to you by CAST AI

CAST AI is a Kubernetes cost optimization platform that cuts your cloud spend by 50% or more. The platform automatically optimizes clusters, leading to improved performance, reliability, and a boost in DevOps and engineer productivity.

Best,

Allen from CAST AI