Megaport, a Brisbane-based, ASX-listed connectivity provider, has selected VAST Data’s AI Operating System as the data layer for its global platform, as it pushes beyond networking into compute and GPU services.
The deal underpins a significant strategic shift for Megaport, which is extending from automated connectivity into integrated compute and GPU following its acquisition of bare-metal provider Latitude.sh.
The company plans to combine its programmable, private connectivity across more than 1,100 data centres, Latitude.sh’s automated bare-metal compute and GPU services, and VAST’s enterprise data services into what it describes as a single global services fabric for deploying distributed AI workloads.
A central piece of the partnership is VAST DataSpace, which gives Megaport’s platform a global namespace so customers can access and manage data consistently across on-premises systems, public clouds, GPU-focused 'neocloud' providers and edge locations without duplicating it into new silos.
“Enterprises are no longer thinking about networking, compute and data as separate decisions. They want infrastructure that is automated, global and flexible enough to support what AI requires next,” said Michael van Rooyen, executive vice-president of global innovation at Megaport.
“VAST gives us one software layer to support enterprise data services across both Megaport and Latitude.sh, allowing us to expand beyond connectivity into the services layer customers need to build and scale modern AI workloads, with fewer operational silos and a faster path from distributed infrastructure to production AI.”
Phil Manez, vice-president of strategic initiatives at VAST Data, said AI does not scale on infrastructure that is powerful in pieces but fragmented in practice.
“The next era will be built on global fabrics that connect data, compute and services wherever customers need them," he said.
"Megaport has built one of the most important connectivity platforms in the world, and with the VAST AI Operating System, that platform can become a data-aware foundation for production AI, helping infrastructure, compute and governed data operate together.”




