The whole-room approach to cooling is out.
Precision monitoring and predictive cooling at the equipment level is in.
See how DDC is revolutionizing scalable data center design at SC24.
[SPONSORED GUEST ARTICLE] The theme of the 2024 SC conference is “HPC Creates.” Speakers will discuss how high-performance computing (HPC) and artificial intelligence (AI) are driving advances in astrophysics, biomedical simulations, and sustainable engineering while also revolutionizing creative industries such as film and music.
In addition to inspirational material about the future potential of HPC and AI, attendees will also spend time learning about how advancements in data center infrastructure and cooling techniques are making all of this possible behind the scenes. HPC and AI are pushing data center boundaries to its limits. Traditional cooling solutions struggle to keep up with the intense heat generated by compute-intensive workloads.
Luckily, new infrastructure advancements are revolutionizing data center cooling. DDC Solutions will be at SC24 November 17-22, 2024 in booth #1413 to share how data center operators can leverage the highest-density air-cooled cabinets with direct-to-chip liquid cooling capabilities to enable efficiency and performance for the next generation of innovation.
What Got Us Here Won’t Get Us There
In the past, when hardware rack power requirements rarely exceeded 20 kilowatts (kW), data centers could rely on traditional in-room air-based cooling and monitoring systems to maintain safe operating temperatures. However, HPC and AI deployments are now running up to 50-70kW in a single rack on the low side. And with so many new applications for supercomputing, a megawatt is on the horizon. Conventional air-cooling, management and monitoring methods can’t keep up.
Traditional vast-space cooling approaches did not account for the individual needs of each piece of equipment in a data center – some of which could be running hotter, higher density workloads than other equipment nearby. This left stranded assets throughout the data center, stressing expensive equipment, increasing the risk of downtime, and reducing overall operational efficiency.
While the industry is exploring several cooling solutions, operators must find proven solutions that work for every customer and do not require massive retooling of the data center and the computing hardware being deployed.
Standard Workloads to Supercomputing – A Scalable Approach
Self-contained, surgically monitored and controlled cabinet enclosures, which offer high density air cooling and liquid to chip within the same cabinet, are an innovative option for data center operators today. As workloads increase and decrease, the intelligent system predicts equipment needs based on real-time monitoring inside the cabinet itself. Dynamically adjusting air and water flow before the equipment becomes overheated maintains a constant temperature throughout the cabinet, creating a more sustainable and cost-effective operation.
Cabinet enclosures act as self-contained building blocks so operators can add cabinets and better manage capital investments. A data center should be able to support 6kW equipment in one cabinet right next to 85kW cabinets and 400kW cabinets—all without sacrificing efficiency or stranding valuable resources. In addition, data centers must be able to scale from small multi-rack deployments to massive multi-megawatt installations without impacting existing operations.
What Does the Data Center of the Future Look Like?
- A flexible building blocks solution that can scale as a data center adds additional equipment and customers. Data center operators will no longer need to rip and replace as technology evolves.
- Cabinets that are equipped to handle every sized workload – from small to huge. Each self-contained cabinet can efficiently cool small workloads with air cooling technology but can also handle HPC and AI workloads with liquid to chip capabilities.
- Predictive, dynamic management and real-time control of airflow and water cooling – As workloads increase and decrease, DCIM software predicts equipment needs based on real-time monitoring inside the cabinet itself and dynamically adjusts airflow and water temperatures before the equipment becomes overheated – maintaining a constant temperature throughout the cabinet. This saves operators thousands in cooling costs.
- Real-time equipment monitoring – taking place at the equipment level instead of the “vast-space approach”. Reduce risk and improve sustainable operations with real-time condition monitoring for every piece of equipment, plus the ability to zoom out for a panoptic view of your entire data center from a single screen.
See What’s Next in Data Center Infrastructure Technology at SC24
See these innovative cabinets firsthand at SC24 where DDC is unveiling its groundbreaking new cabinet solution. This new scalable solution is poised to support the next generation of supercomputing equipment efficiently and effectively.
- Explore the S-Series:Experience DDC’s award-winning cabinets, currently delivering up to 85kW per cabinet and unmatched efficiency. DDC is unveiling a game-changing advancement in high-density cooling at SC24 that far surpasses their current specifications.
- See DCIM in action:Witness the power of dynamic management and real-time monitoring software for optimizing data center performance.
- Discuss your HPC challenges:DDC experts are ready to have one-on-one conversations about your specific needs.
We would love to show you what’s possible. Register for an in-person meeting with our team at SC24.
