Alina Trabattoni’s Post

Very interesting comment on data center and it's relevance to Malaysia.

The latest report from Reuters is a stark reminder of the immense and likely underestimated growth in electrical demand, particularly from data centers powering technologies like generative AI. As U.S. electric utilities brace for this surge, it’s clear that the energy landscape is undergoing a seismic shift. This isn’t just about keeping servers running; it’s a litmus test for economic competitiveness. Countries that can provide ample, cheap, and sustainable electricity will have a distinct edge in the global market. They’ll attract more businesses, foster innovation, and lead the charge in the tech-driven economy of tomorrow. The numbers speak volumes: power use from data centers is expected to triple globally, and long-term power demand from U.S. data centers alone could reach more than 50 gigawatts by 2030. This is not just a challenge but an opportunity for nations to rethink their energy strategies and infrastructure. Investing in renewable energy sources, enhancing grid resilience, and promoting energy efficiency are no longer options but necessities. The race is on, and the prize is a thriving, sustainable economy. Let’s not underestimate the power of electricity in shaping our future. #EnergyDemand #DataCenters #SustainableEnergy #EconomicGrowth

🛜 According to the International Energy Agency, datacenters consumed 460 TWh of energy in 2022, and their consumption can pass 1000 TWh in 2026. Just for comparison, Germany, the world's 3rd largest economy, in 2022 consumed 507 TWh of energy. The good news about the data centre computational workload is that it has certain flexibility in both when and where it is executed.   🍃 Google introduced a global carbon-intelligent compute management system, which actively minimizes electricity consumption based on carbon footprint by delaying temporally flexible workloads. The core component of the system is a suite of analytical pipelines used to gather the next day’s carbon intensity forecasts, train day-ahead demand prediction models, and use risk-aware optimisation to generate the next day’s carbon-aware Virtual Capacity Curves (VCCs) for all datacenter clusters across Google’s fleet. VCCs, shown as red lines in the figure, impose hourly limits on resources available to temporally flexible workloads while preserving overall daily capacity, enabling all such workloads to be completed within a day with high probability.   🔌 Similar load flexibility management programs are important to handle the capacity constraints imposed by the grid architecture and renewable energy sources. The best-known example of these constraints is the “duck curve,” which occurs because of a fairly constant load throughout the day, which increases in the evening (see 2015 curve) and the effect of a large PV share on the energy generation, which effectively turns off other dispatchable sources of electricity like gas during the day (see curve 2023). This midday dip in net load for the conventional sources is getting lower with increasing PV share and that presents at least two challenges for grid operators. The first challenge is grid stress since the conventional sources of energy must ramp up production very quickly. The second challenge is economic. The dynamics of the duck curve can challenge the traditional economics of dispatchable power plants because the factors contributing to the curve reduce the amount of time a conventional power plant operates, which results in reduced energy revenues.   🔦 On the other hand, the duck curve creates opportunities for both energy storage and the advanced design of data centres. If you are interested in ABB applications in data centres or battery energy storage knowledge check https://lnkd.in/gmhtw98F. #datacenters #sustainability #cleanenergy #abb #gridmodernization #energystorage

Power supply is the main bottleneck for Data Centres....Data Centres will go where the power is...Better to solve this supply shortage and grid upgrade issue first...

As Amazon and Microsoft race to build new data centers in Europe, there's growing concern about meeting AI's energy demands sustainably. Companies like Amazon are striking deals to secure renewable energy. The rush to build data centers is evident, with significant investments in Sweden, Germany, France, Spain, and Finland. However, the increasing demand for clean energy is straining electricity grids, pushing tech giants to consider off-grid solutions like gas, diesel, or green hydrogen power. Data centers need immense power to operate and cool servers, with energy demand expected to double by 2026. Companies have been securing renewable energy contracts, but grid congestion remains a challenge. Some areas, like Dublin and Amsterdam, have imposed restrictions on data center growth due to grid issues. Tech giants are exploring innovative solutions, including on-site renewable projects and advanced technologies like nuclear fusion and hydrogen fuel cells. While these ideas are still developing, the goal is to find sustainable ways to power the growing demand for AI and data centers. #ai #energytransition #digitalinfrastructure #sustainability

The UK's data centre power demand is set to increase six-fold within the next decade, as per National Grid CEO John Pettigrew's insights. With the nation at a "pivotal moment," Pettigrew outlines the pressing need for an ultra-high voltage onshore transmission network to address the growing strain on an already-constrained grid. As AI and quantum computing advance, the demand will not only persist but intensify due to increasing reliance on these energy-intensive technologies. And it isn't just about the commercial sectors; residential shifts towards electric vehicles and heat pumps equally contribute to this surge. Despite the challenge, data centre companies are actively working to mitigate the impact on energy demand. Their efforts are crucial in supporting the transition to net zero and reinforcing the resilience of the UK's power infrastructure. The envisioned "super-supergrid" proposed by Pettigrew, capable of handling 800,000 volts, represents a leap towards a more strategic, interconnected electric power system. The new network would bolster energy distribution, supporting large-scale energy sources and demanding centres. Data centres, competing for grid access, now find themselves at the heart of sustainability initiatives, with the government's £960 million investment in green technologies and plans for an expedited 'connections action plan'. Overall, data centre companies are not just passive consumers of energy; they are at the forefront of pioneering solutions to reduce the environmental impact of their operations and cater to the nation's escalating power requirements. #DataCenter #EnergyDemand #Sustainability #DataCentre

A recent report by Bain & Company highlights a critical challenge for US energy companies: adapting to the AI-driven datacenter boom. With power consumption expected to outstrip supply within a few years, here are the key takeaways: Key Insights: **Rising Demand: AI model development is driving a 30% increase in datacenter capital expenditure this year. Power consumption from datacenters is projected to more than double by 2030. **Energy Generation Needs: Utilities need to boost annual energy generation by up to 26% by 2028 to meet demand. **Global Impact: Similar challenges are faced worldwide, with European datacenters also struggling to secure power. **Investment Risks: While the investment required to build additional capacity is significant, missing out on potential growth could be equally risky. **Modernization: Utility companies must modernize their operations to scale up energy resources rapidly, including upgrading power generation, transmission, and distribution lines. Environmental Considerations: **Datacenters could emit 2.5 billion tonnes of GHG by the end of the decade. **Investment in decarbonization solutions such as carbon capture and removal technologies is essential. Future Outlook: **Leveraging AI for achieving global climate goals is advocated by industry leaders like Bill Gates and Eric Schmidt. Energy companies are at a critical juncture. Adapting to this surge not only supports AI advancements but also opens avenues for energy transition investments and potentially reduces residential rates. 🌐💡 #EnergySector #AIDatacenters #Sustainability #Innovation #TechTrends #Utilities Read more from The Register here: https://lnkd.in/eJA9CBXZ

The European Commission Joint Research Centre has published a new report assessing the Energy Consumption in Data Centres and Broadband Communication Networks in the European Union (EU-27) in 2022. https://lnkd.in/db2PRGm3 Digital infrastructure – data centres and telecommunication networks – are at the heart of the digital transformation, underpinning all aspects of our increasingly digitalised and connected societies. Demand for their services is rising rapidly, raising concerns about their energy use and environmental impacts. Between 2015 and 2022, the number of internet users globally increased by 80%, internet traffic increased five-fold, while data centre workloads more than quadrupled. However, energy efficiency improvements in computation, data storage, and data transmission has helped to limit energy demand growth globally. Between 2015 and 2022, data centre and network energy demand grew at a much slower pace (+2–8% CAGR) than data traffic (+24% CAGR) or data centre workloads (+33% CAGR). Data centres and data transmission networks each accounted for 1–1.5% of global electricity use globally in 2022. Data centres in the EU used an estimated 45–65 TWh of electricity in 2022, equivalent to 1.8–2.6% of total regional electricity consumption. Data centres as a share of national electricity use range from as low as 0.4% in some countries to as high as 18% in Ireland . Telecommunication networks used an estimated 25–30 TWh of electricity, equivalent to 1–1.2% of total EU electricity use. The combined energy use of data centres and telecommunication networks in the EU was 70–95 TWh in 2022, equivalent to 2.8–3.8% of total regional electricity use Policymakers and companies must work together to improve data collection, quality and availability. While the estimates of this study represent a likely range of figures, it is critical to develop more robust estimates to better understand trends and make informed policy decisions to manage the energy and environmental impacts of digital infrastructure. Governments and companies should work together to improve data quality and availability regarding data centre energy consumption, telecommunication network energy use, as well as relevant activity indicators (e.g. connections, data traffic, data centre workloads). Data collection efforts should also seek to better understand energy use characteristics and implications of specific services and tasks such as artificial intelligence. #datacenters #energy #energyefficiency #climatechange #sustainability

#News The Big-Tech Clean Energy Crunch Is Here [ad_1] Big Tech’s appetite for energy is just about visible from the east coast of Scotland. Some 12 miles out to sea sits a wind farm, where each of the 60 giant turbines has blades roughly the length of an American football field. The utility companies behind the Moray West project had promised the site would be capable of generating enough electricity to power 1.3 million homes once completed. That was before Amazon stepped in.In January, Amazon announced it had struck a deal to claim more than half the site’s 880 megawatts of output, part of its ongoing attempt to slake its unquenchable thirst for power. As the world’s biggest companies race to build the infrastructure necessary to enable artificial intelligence, even remote Scottish wind farms are becoming indispensable.In Europe last year, $79.4 million was spent on new data center projects, according to research firm Global Data. Already in 2024, there are signs that demand is accelerating. Today Microsoft announced a $3.2 billion bet on Sweden data centers. Earlier this year, the company also said it would double its data center footprint in Germany, while also pledging a $4.3 billion data center investment for AI infrastructure in France. Amazon announced a network ...

The growing power demand of AI is well documented. What is talked about less is where that power will come from and the potential restriction power availability will have on AI growth. New data center site selection is become first and foremost about power availability. All other considerations such as labor availability, climate, proximity to disasters, etc. are becoming secondary. The reopening of Three Mile Island’s unit 1 is a clear example of what is becoming necessary to meet AI’s power needs. More stories like this are coming.