Navigating Cloud Costs and Carbon Impact: The Role of Gen AI and Advanced Analytics

The Growing Challenge of Cloud Costs and Environmental Impact
 

The Surge in Worldwide End-User Spend

The public cloud is set to experience a remarkable surge in worldwide end-user spend, projected to reach a staggering $675.4 billion by the end of 2024, as per Gartner. This substantial growth is a testament to the appeal of cloud computing, offering scalability and flexibility for organisations.

Wasted Cloud Spend and Environmental Concerns

Despite its benefits, recent surveys have unveiled a concerning trend: up to 30% of cloud spend is wasted due to the prevalent pay-as-you-go model, lack of visibility, overprovisioning, and unplanned costs. Simultaneously, the environmental impact of cloud computing has become a pressing concern, particularly in the face of climate change and rapidly evolving ESG regulatory landscape.

Corporate Optimism and Cost Reduction Priorities

Deloitte’s latest quarterly CFO survey has revealed that 55% of CFOs have identified corporate cost reduction as a top priority. This underscores the urgency for businesses to address the escalating costs and environmental implications associated with cloud usage.

The Impact of Gen AI on Cloud Workloads

Transformative Capabilities and Cost Challenges

Gen AI has captured the imagination of businesses with its inherent ability to generate new content and provide valuable insights. However, this transformative technology also presents challenges, particularly in driving up costs, energy consumption, and carbon emissions.

Environmental Implications of Gen AI

As cloud adoption has soared to drive digital transformation, the cost and carbon impact of Gen AI have simultaneously risen. The need to integrate cost reductions with low carbon emissions has become a critical consideration, as highlighted by Amazon’s CTO Werner Vogels at last year’s AWS re:Invent.

Balancing Benefits and Carbon Impact

The power of Gen AI to fundamentally change the business landscape in the next five years is undeniable. However, the models produced by Gen AI consume significant computational power for operations, virtualisation, and data transfers, contributing to increased cloud spend, energy consumption, and carbon emissions.

Leveraging Advanced Analytics for Cloud Cost and Carbon Optimisation

In response to the escalating cloud costs and environmental concerns, Deloitte has taken a proactive approach to address these challenges through advanced analytics and data-driven strategies.

1. Rapid Diagnostic Tests and Optimisation Reviews

We are committed to controlling the costs and emissions of clients’ existing data and Gen AI workloads by conducting rapid diagnostic tests and optimisation reviews. These comprehensive assessments culminate in remediation plans that are tailored to each client's specific needs and objectives.

2. Cost-Aware and Carbon-Aware Design Patterns

Our approach involves the development of cost-aware and carbon-aware design patterns, which are designed to optimise cloud workloads while simultaneously reducing environmental impact. By integrating these design patterns, clients can achieve a balance between operational efficiency and sustainability.

3. Granular Cloud Consumption Monitoring Schemes

To gain a deeper understanding of cloud usage patterns and identify areas for improvement, we set up granular cloud consumption monitoring schemes. These schemes provide valuable insights into usage trends, underutilisation, and excess capacity, enabling clients to make informed decisions about their cloud resources.

4. Accelerators for Cost and Carbon Optimisation

We have developed several accelerators to quickly identify and quantify opportunities for cost and carbon optimisation. These accelerators include a detailed hypothesis library covering key services across all major cloud platforms, coupled with data analytics expertise to rapidly validate hypotheses for quick wins.

5. Savings and Emissions Calculator

We are building a savings and emissions calculator provides actionable insights on potential areas that can be evaluated for cost and carbon reductions. This tool enables clients to assess the impact of proposed changes and make informed decisions about optimising their cloud workloads.

6. Investment in In-House Tools and Development

We are continuously developing in-house tools to support clients in their cloud cost and carbon optimisation efforts. These include a KPI and dashboards template, as well as a Gen AI prompt analyser, which will be designed to provide clients with the necessary resources to monitor, analyse, and optimise their Gen AI and Data Workloads on cloud environment.