ASUS Servers Outshine Competitors with 1.23X Superior Performance Efficiency¹

Since 1995, ASUS has leveraged deep technical expertise and experience to design and manufacture server systems. Over the years, we have become a leading provider of reliable and easily scalable solutions and services, making us a key partner for data-center development. With outstanding hardware design and low-level tuning capabilities, ASUS continues to push the limits of data center computing, taking AI performance to unprecedented heights.

In November of 2024, ASUS and Ubilink successfully completed Taiwan's largest supercomputing center (TOP500: 31st; Green500: 44th).  Built in just three months, this ambitious project is equipped with 128 NVIDIA HGX H100 servers and 1,024 GPUs, achieving an impressive 45.82 PFLOPS.

The ASUS-Ubilink project includes 128 AI-powered ESC N8-E11 servers that are equipped with NVIDIA HGX H100 SXM5 80GB GPUs and InfiniBand NDR400 connectivity. With this solution, ASUS achieved an Rmax of 45.82 PFLOPS, with an efficiency (Rmax / Rpeak) reaching 66.08%. These results were reported to Top 500.org. As seen in the table below, under the same GPU and high-speed network card conditions, the ASUS solution is more efficient, delivering 1.23X the overall performance of competitors¹.

COMPARISON RESULTS¹

ASUS AI/GPU SERVER SOLUTIONS

Leading Performance for Success

ASUS is not just a hardware provider: We focus on helping customers increase operational efficiency and reduce costs. In 2024, ASUS servers continued to break records in industry-leading performance benchmarks, setting 1,959+ world records⁵ and ranking at the top in multiple tests. Our goal is to maximize the operational efficiency of every rack through our software and hardware solutions, unlocking greater potential for long-term success.

Industry Leader in Data Center Performance Tuning

The expert ASUS solutions team specializes in performance calibration and system optimization. With extensive experience in MLPerf GPU tuning, we handle system and network configurations systematically to ensure smooth performance calibration and achieve optimal results.

Boosting Total Cost of Ownership (TCO) Efficiency

ASUS offers comprehensive infrastructure solutions that maximize server performance ― from firmware to software ― improving data center operational efficiency and reducing long-term IT expenditures.

1. Based on data from TOP500.org. All the data used in this analysis is sourced from the published information of the Top 500 list, November 2024. We assess the performance of each data center by comparing the computing efficiency, represented as the ratio of Rmax to Rpeak.
2. Brand A uses NVIDIA HGX H100 SXM5 80 GB, InfiniBand NDR400 with a total of 190 nodes
3. Brand B uses NVIDIA HGX H100 SXM5 80 GB, InfiniBand NDR400 with a total of 64 nodes
4. Efficiency is measured by Rmax / Rpeak, with a higher efficiency indicating better computing performance.
5. All results are available on the official SPEC.ORG website.
6. (Total cores - Accelerator/co-processor cores) ÷ Total cores per server = Number of nodes in the data center.

Learn more about the successful story

Learn more about ASUS infrastructure solution