For this episode of the Data Center Frontier Show podcast, we sat down with liquid cooling data center partners Park Place Technologies and ZutaCore. During the podcast, DCF Editor in Chief Matt Vincent spoke with Chris Carreiro, Chief Technology Officer for Park Place Technologies, and Manfreid Chua, Vice President-Business Development, AI & Sustainability for ZutaCore, about how the companies' partnership is enhancing liquid cooling technology prospects for sustainable AI computing.

In September, Park Place announced the expansion of its portfolio of IT infrastructure services to include the two major liquid cooling formats for data centers, i.e immersion liquid cooling and direct-to-chip cooling. ZutaCore is a key developer and supplier of direct-to-chip, waterless liquid cooling technology which formally supports NVIDIA's GPUs.

Direct-to-chip advanced liquid cooling technologies apply coolant directly to the server components that generate the most heat, including CPUs and GPUs. And Park Place notes that immersion cooling empowers data center operators to do more with less: less space and less energy. Using liquid cooling methods, the company contends that businesses can increase their PUE by up to 18 times, and rack density by up to 10 times. Ultimately, this level of efficiency can lead to power savings of up to 50%, which in turn leads to lower operational costs.

Park Place also notes how, from an environmental perspective, liquid cooling is significantly more efficient than traditional air cooling. The company reckons that, at present, air cooling technology only captures 30% of the heat generated by the servers, compared to the 100% captured by immersion cooling, resulting in lower carbon emissions for businesses that opt for immersion cooling methods.

Park Place prides itself on providing a single-vendor outlet for the whole liquid cooling technology adoption process, from procuring the hardware, conversion of the servers for liquid cooling, to installation, maintenance, monitoring and management of the hardware and the cooling technology.

“Our turn-key liquid cooling offerings have the potential to have a significant impact on our customers’ costs and carbon emissions, two of the key issues they face today,” said Carreiro.  “Park Place Technologies is ideally positioned to help organizations cut their data center operations costs, giving them the opportunity to re-invest in driving innovation across their businesses."

In the course of our talk, Carreiro highlighted the challenges of data centers' AI sustainability conundrum, and the corresponding benefits of Park Place's warranties. For his part, ZutaCore's Manfreid Chua delved into the industry's shift from air to liquid cooling due to the demands of generative AI, and the advantages of his company partnering with Park Place for optimizing the energy efficiency footprint of data centers. 

Additionally, Chua shared insights regarding the economic value of NVIDIA's AI accelerators, and the finer points of the race to sustainability and net zero for large-scale AI data centers. Chua talked about talk about how resources like land, energy, and water all become possible limiting factors for AI factories at scale, and how liquid cooling can help alleviate such limitations.

Join us for this podcast as we explore the dynamic landscape of data centers and how Artificial Intelligence (AI) has reshaped them. We'll delve into the shift from a 'north-south' traffic system to the sophisticated 'east-west' system that revolutionized data processing.

Our guest, Dave Hessong from Corning, illustrates the crucial role of high-speed connections like 800G in meeting AI's demands. The discussion reveals how upgrading to this speed is not just beneficial, but essential in optimizing your data center.

Latency, a key factor in network performance, is also a core topic of our conversation. Understanding its significance and how reducing it can enhance performance provides an edge in today's competitive market.

The discussion further delves into the importance of state-of-the-art fiber optic cables, connectors, and cabling architecture in boosting a data center's performance. The complexities of AI deployment, its impact on fiber density, and the innovative solutions it necessitates are also explored.

As we unveil the future of data centers, the estimated rise in AI capacity and the associated challenges are discussed. These include the increased power requirements and the need for a more organized cable and fiber infrastructure.

While 800G might seem like just the beginning, the discussion elaborates on how this transition can future-proof your data centers for the next three to seven years. The extraordinary and transformative impact of AI, still in its infancy, on business and society is also a key highlight.

Looking to the future, the anticipated growth in bandwidth as AI continues to evolve, and the exciting prospect of technology reaching 1.6Tbps next year, are discussed.

We encourage you to tune in and engage with us as we navigate this rapidly evolving field. Regardless of your level of expertise, this conversation promises valuable insights into the future of data centers. Join us on this enlightening journey into the world of AI and data centers.

Prometheus Hyperscale is the new corporate entity formed this month which expands upon the footprint and the promise of the Wyoming Hyperscale White Box project, first reported on by DCF in 2022. 

For this episode of the Data Center Frontier Show podcast, we spoke with Trenton Thornock, founder of Wyoming Hyperscale, who has been appointed as Chief Executive Officer of Prometheus Hyperscale; Trevor Neilson, a seasoned climate-tech CEO and energy transition investor, who joins as the company's President; and John Gross, President of J.M. Gross Engineering, who is handling the project's liquid cooling infrastructure. 

The Wyoming Hyperscale White Box data center has been under construction since 2022 on 58 acres of land near Aspen Mountain Evanston, Wyoming, and represents a blueprint for creating super-efficient data centers with low impact on the environment and benefits for the local community. In the companies' transition, Wyoming Hyperscale has merged with Prometheus Hyperscale and been expanded from a 120 MW project to plans for a data center campus with 1 GW of IT capacity.

The data center is being built on land owned by Thornock's family, which has been involved in ranching for 6 generations. The location benefits from ready access to renewable energy from nearby wind and solar farms. Wyoming Hyperscale has a contract with Rocky Mountain Power for 120 megawatts of power and a 138 kV substation, which is fed by the same switchgear as the renewable energy generation sites. The site sits on a major east-west fiber highway that tracks the 41st parallel, along which data center hubs have emerged in places like Ohio, Iowa, Nebraska and Utah. The Union-Pacific Railroad line, which provides key right-of-ways for fiber deployment, runs through nearby Aspen Mountain.

The Evanston project underscores Prometheus Hyperscale’s commitment to sustainability and innovation. By integrating 100% renewable energy and advanced liquid cooling technology combined with heat reuse, the Evanston facility promises to be one of the most efficient and environmentally friendly data centers in the world.  Importantly, less than 10% of the project’s power development plan is grid dependent (120 MW of 1,220MW or 9.84%).

The first facilities yielded by Phase 1 of the Evanston project are expected to come online within the next 18 months. Prometheus Hyperscale has also revealed plans to construct four other data centers across Arizona and Colorado. And as previously reported by DCF, this May saw the announcement of a 20-year power purchase agreement (PPA) by fission-based nuclear small modular reactor (SMR) specialist Oklo to deliver 100 MW of power to Prometheus, using Oklo's Aurora Powerhouse reactors for power generation.

"Our partnership with Oklo not only provides us with a reliable, clean energy source but also positions us as a leader in sustainable data center operations," said Thornock. "Sam Altman’s and Jacob Dewitte’s vision for a sustainable future through advanced energy solutions aligns perfectly with our mission at Prometheus Hyperscale."

During the podcast, Thornock discussed the evolution of the Wyoming hyperscale project with Prometheus, highlighting its growth to a 1 GW prospect since the groundbreaking of the Evanston County project in 2022. For his part, Trevor Nielsen emphasized increasing demand for Prometheus driven by advancements in computing power and the importance of sustainability in the energy transition. 

Our conversation also covered the company's partnership with Oklo, focusing on the streamlined permitting process for small modular reactors in Wyoming and the strategic use of resources for data center energy generation.

Sustainability is a critical factor in data center design.  The topic encompasses a series of design trade offs including: reliability, site selection, water usage, operating parameters, construction materials and cooling efficiencies. Due to a couple of key paradigm shifts in the industry, today’s data center owners and operators are looking to meet their cooling demands with air cooled solutions. All this needs to be done in conjunction with optimizing energy efficiency leading to a significant change in HVAC system products and design.   

In this conversation, Jeffrey Jerwers can discuss the trends driving the need for water conservation and associated equipment impact. He will detail the types of economizers available for mechanical cooling systems, application by climate zone and their associated design tradeoffs.

Data centers are complex, high-stakes environments where downtime is not an option. The sheer volume of interconnected systems and components creates a daunting challenge for operators. This is where digital twins shine.

Because of this complexity, data centers require a new level of understanding. Digital twins—virtual models fed by real-time DCIM data—can offer a transformative solution. The key is that the Digital Twin is only as good as accurate real-time data.

This continuous flow of real-time information allows operators to see the bigger picture, from power usage to equipment health. Imagine a live, digital replica predicting bottlenecks, optimizing cooling, and enabling proactive maintenance.

A digital twin can allow your data center to analyze your infrastructure, highlight potential issues, and provide highly accurate details on the impact of proposed changes − viewable as your monitored values change. Watch an ATS or PDU view with the power load that reflects your changing values and plan. Watch a power load peak during a fail-over with your plan changes applied to the real-time data.

Power management with digital twin capabilities can simulate the failure of a device or load change and accurately model the effects of that failure, including triggering failure over to redundant partners and cascade failures.

With a DCIM solution with digital twin capabilities, you gain insightful reporting that identifies potential risk areas in your infrastructure. For instance, it can flag power distribution gear that represents a single point of failure, which could lead to equipment de-energization and impact customers and SLAs. This proactive approach to risk management is a vital advantage of a DCIM solution with digital twin capabilities.

Imagine a planning module based on a digital twin model and its capabilities. It doesn't just show simple details like additional loads but also simulates complex scenarios. For instance, it can predict when a device will fail, reroute load in the virtual model to redundant partners, and show the effect on those devices as well. This comprehensive approach to planning is where a true digital twin adds much more value than a simple load addition or removal.

By using real-time DCIM data, digital twins become intelligent partners, ensuring peak performance and a more resilient data center.

While the idea of a digital twin for your data center has existed for a long time, operators now can have that digital twin fed by millions of data points per minute inside a full-fledged powerful DCIM. The view of Modius is that anything less makes it just a "Digital Cousin".

The company believes its Modius ® OpenData® is the gateway to these next-gen capabilities and is using this podcast to kick off this effort.

As everyone on the Data Center Frontier and Endeavor Business Media (EBM) teams regroups from last week's sold-out DCF Trends Summit (Sept. 4-6) conference in Reston, Virginia, for today's episode of the DCF Show Podcast we bring you something a bit different.

Recorded earlier this year, EBM's Data Center, Communications, and Power Infrastructure Confluence Forum is a shared discussion among the lead editors of key brands and publications in EBM's Digital Infrastructure and Energy Groups.

The discussion frames and addresses the topic of rapidly expanding stakes and implications for the data center, information and communications technology (ICT), fiber broadband, and on-site power generation infrastructure sectors in the age of advanced computing and connectivity for AI/ML, IoT, 5G LTE, all flavors of Ethernet, and other pertinent technology applications.

EBM editors in order of their participation in this discussion include:

00:00 - 14:00 - Matt Vincent, Editor in Chief, Data Center Frontier

The discussion winds up with a bit of cross-questioning among the editors.

We at Data Center Frontier hope you'll enjoy this podcast, and will resume with our regular, data center industry-specific coverage later this month.

For this episode of the Data Center Frontier Show podcast, we welcome Mark Seymour, Distinguished Engineer with Cadence Design Systems, for a discussion of the big question on everyone’s mind right now in this industry: data center power demand and where it's going in the context of rapid digitalization and exponential growth of HPC and AI computing needs, and how that compares and contrasts, or even conflicts, with increasing environmental concerns and regulations. 

The conversation also highlights the importance of digital twins for managing data center efficiency and the advantages of liquid cooling technology, and particularly immersion cooling, as a sustainable alternative to traditional methods. In the course of our interview, Seymour also emphasizes the data center industry's responsiveness to societal demands for sustainability, citing initiatives such as ubiquitous tree planting by project developers, and the need to adapt to new technological challenges.

Here's a timeline of the podcast's key moments:

2:59 - Seymour explains that AI is essentially high-performance computing, which is now required in many data centers that previously did not need it.

12:05 - Addressing the challenges and potential of immersion cooling technology: Emphasizing its growing acceptance, but also the need for confidence in its operation.

17:52 - Talk turns to the importance of digital twins in ultimately managing data center efficiency, with Seymour highlighting the necessity for understanding the interrelated behaviors of IT infrastructure and cooling systems.

24:18 - Discussion circles back to immersion cooling as a sustainable option for data centers, with Seymour expounding on its advantages over traditional cooling methods.

27:44 - Seymour elaborates on the improvements in compute efficiency per watt in modern systems, arguing that the data center industry is responding and adapting to societal demands, rather than being inherently unsustainable.

30:42 - Seymour acknowledges the industry's focus on sustainability and environmental impact, citing examples such Cadence's tree planting initiatives and the ongoing challenge of meeting new technological demands.

Visit Data Center Frontier

Data Center Frontier opens our podcast interview catching up with CyrusOne CEO Eric Schwartz by discussing the company's recent $12 billion in announced financing, highlighted by a new $8 billion warehouse facility in the U.S. to support growth driven by demand from hyperscalers and AI technologies.

In the course of the discussion, Schwartz notes CyrusOne's strong growth trajectory, new leadership, and expansion plans in Europe and Japan, while emphasizing the organization's principles of earning customers' trust and a commitment to operational excellence. 

We also receive an update on the progress of the company's Intelliscale offering for build-to-suit AI data centers. Additionally, the talk covers CyrusOne's 2024 sustainability report, focusing on the company's carbon neutrality efforts, renewable energy investments, and the overall industry's commitment to reducing carbon footprints.

This May, Digital Realty (NYSE: DLR) announced a collaboration with Oracle to accelerate the growth and adoption of artificial intelligence (AI) computing among enterprises. For this episode of the Data Center Frontier Show podcast, we asked Digital Realty Chief Revenue Officer Colin McLean to expand on key points of his company's AI data center design and implementation efforts in light of the new partnership with Oracle.

In their announcement, Oracle and Digital Realty said their new strategic collaboration aims to develop hybrid integrated solutions that "address data gravity challenges, expedite time to market for enterprises deploying next-generation AI services, and unlock data and AI-based business outcomes."

We also asked McLean about how he's seen the trend lines for data center pricing, leasing and capacity changes over the past 5-6 years of the cloud industry, compared to roughly the past year of AI growth since he's been CRO for Digital Realty.

Here's a timeline of the interview's key moments:

 0:32 - Data Center Frontier asks McLean to elaborate on the salient points of AI data center design in light of Digital Realty's partnership with Oracle. McLean explains the significance of the partnership, emphasizing how it addresses the challenges of managing high-density workloads in AI and data-intensive applications.

6:19 - DCF continues asking about trends in AI data center design, particularly regarding pricing, leasing, and capacity changes over the past few years. Citing how enterprises and service providers are planning for increased capacity requirements due to AI growth, McLean highlights the need for forward-thinking capacity planning due to evolving requirements, increasing workload density, and the introduction of new programs to support higher density requirements.

8:14 - Trends in AI data center design since the industry's ChatGPT inflection point of 2023 are addressed. McLean emphasizes the importance of designing data centers to accommodate mixed densities globally, sustainability considerations, and the need to closely collaborate with clients and partners to meet evolving requirements.

12:20 - McLean discusses the global scale of Digital Realty, emphasizing capacity requirements, major metros, emerging markets like Frankfurt, and the growth of the platform across various regions.

15:06 - DCF Editor in Chief Matt Vincent directs the conversation toward the topic of power, highlighting its significance in the data center industry and asking McLean about aspects related to renewables, the grid, and onsite powering options.

15:45 - McLean elaborates on the importance of sustainability for Digital Realty, emphasizing the company's efforts to work with municipalities, support a greener world, and address power concerns globally, including plans for future expansion into markets such as India and Africa.

17:42 - DCF acknowledges the critical link between sustainability, AI, and power issues, prompting McLean to reiterate the company's commitment to supporting a sustainable world and navigating the balance between local needs and economic growth.

Welcome to the forefront of data center innovation! Today, we'll explore how Artificial Intelligence (AI) is driving a revolution in data center design. We'll delve into three key areas: high-performance networking, cutting-edge cooling solutions and advancements in fiber optic technology. These advancements are all essential for supporting the ever-growing demands of powerful AI systems.