Case Study: YTL AI Cloud achieves sovereign AI scalability and secure multi-tenancy with WEKA NeuralMesh

YTL Corporation powers Malaysia’s sovereign AI with Weka and 1 GB/s per GPU throughput

YTL Corporation, a major Malaysian conglomerate, faced the challenge of building Malaysia's premier AI cloud from the ground up. Their goal was to deploy one of the largest NVIDIA GB200 clusters in Southeast Asia, which required storage infrastructure capable of delivering sustained high throughput per GPU and secure multi-tenancy. They partnered with the vendor WEKA to find a solution.

WEKA implemented its NeuralMesh storage platform to power the AI infrastructure. The solution delivered the required performance of 1 GB/s read throughput per GPU with linear scalability and sub-0.5ms latency. This enabled YTL AI Cloud to successfully launch a sovereign AI platform, which now powers Malaysia's first locally developed large language model and supports AI innovation across key sectors like education and healthcare.

YTL Corporation