Compal Challenges AI Server Status Quo with 'Storage-as-Memory' Push

ST. LOUIS, MO – November 17, 2025 – Compal Electronics, a leading Taiwanese manufacturer, is making a significant push into the high-growth AI and High-Performance Computing (HPC) server market. Unveiled at Supercomputing 2025 (SC25), the company’s next-generation servers, including the SGX30-2 and SX420-2A, represent a challenge to established players like Dell, HPE, and Supermicro, with a unique emphasis on ‘Storage-as-Memory’ architectures and advanced cooling technologies.

A New Architecture for Demanding Workloads

Compal’s strategy centers around a shift from traditional CPU-centric data processing to systems where memory plays a more central role. The SGX30-2, built on NVIDIA's HGX B300 platform, supports up to eight NVIDIA Blackwell Ultra GPUs, while the SX420-2A, leveraging NVIDIA RTX PRO 6000 Blackwell Server Edition GPUs, is designed for flexibility and high-density deployments. This emphasis on memory-centric computing isn’t just about boosting performance; it’s about fundamentally altering how data centers operate.

“The industry is recognizing that simply adding more GPUs isn’t enough,” explained an industry analyst attending SC25. “Data movement is becoming the biggest bottleneck. Compal is positioning itself as a company that can address that challenge with their focus on advanced memory interconnects and technologies like Storage Class Memory.”

The SX420-2A’s capabilities go beyond raw processing power. It utilizes NVIDIA’s BlueField-3 DPUs and ConnectX-8 SuperNICs, enhancing networking performance and data transfer speeds. This combination of powerful GPUs and high-speed networking aims to deliver the throughput and low latency required for demanding AI workloads, such as large language model (LLM) training and inference.

Competing in a Crowded Market

The AI server market is experiencing explosive growth, projected to reach over $1.6 trillion by 2033. While Dell and HPE currently dominate the landscape, Supermicro’s agility and early adoption of new technologies have allowed it to gain significant market share. Compal’s entry into this competitive arena isn't without its challenges.

“The biggest hurdle for Compal is establishing brand recognition in a market dominated by established players,” commented a server procurement manager attending SC25. “They need to demonstrate that their servers can deliver on their promises and offer a compelling value proposition to attract customers.”

Compal’s strategy appears to be focused on specialization and customization. The company emphasizes its ability to work closely with clients to tailor solutions to their specific needs, offering flexibility in terms of platform, blade, and multi-node server configurations. This approach differentiates Compal from larger vendors who often prioritize standardized offerings.

Addressing the Heat Challenge

As AI workloads become increasingly demanding, the issue of heat dissipation has become a critical concern. High-density GPU deployments generate significant heat, which can impact performance and reliability. Compal is tackling this challenge with a tiered thermal management system that incorporates air cooling, liquid cooling, and even immersion cooling technologies.

“Liquid cooling is becoming increasingly important for high-performance servers,” noted a thermal management specialist at SC25. “It allows you to pack more processing power into a smaller space and maintain stable temperatures.”

Compal’s tiered approach allows customers to choose the cooling solution that best fits their needs and budget. This flexibility is particularly appealing to data center operators who are looking to optimize energy efficiency and reduce operating costs.

A Vision for the Future

Compal’s push into the AI server market isn’t just about offering new hardware; it’s about shaping the future of data center architecture. By embracing ‘Storage-as-Memory’ technologies and prioritizing energy efficiency, the company is positioning itself as a leader in the next generation of computing infrastructure.

“The industry is at a turning point,” said an industry analyst attending SC25. “We’re moving beyond simply adding more hardware. The focus is now on optimizing the entire system, from memory to cooling. Compal seems to understand this and is building its strategy around that vision.”

With significant investments in US operations, including a $300 million commitment to establish new subsidiaries for its server business, Compal is signaling its long-term commitment to the AI server market. Whether this strategic move will be enough to carve out a significant market share remains to be seen. But, the company’s innovative approach and focus on addressing the key challenges facing data center operators suggest that Compal is a force to be reckoned with in the evolving landscape of AI and HPC.