Back to News
quantum-computing

SDT Launches Korea’s First Commercial Quantum-AI Hybrid Data Center via NVIDIA NVQLink

Quantum Computing Report
Loading...
2 min read
0 likes
⚡ Quantum Brief
South Korea’s first commercial quantum-AI hybrid data center launched in Gangnam, Seoul, by SDT, transitioning quantum computing from research to enterprise applications. The facility integrates a 20-qubit superconducting quantum processor with NVIDIA’s DGX B200 GPUs. The center uses SDT’s QuREKA platform to orchestrate hybrid workloads, splitting tasks between classical GPUs for preprocessing and quantum processors for core computations in logistics, materials science, and financial modeling. SDT joined NVIDIA’s NVQLink ecosystem, enabling microsecond-latency GPU-QPU communication for real-time error correction and qubit calibration, aligning with global AI infrastructure standards. At NVIDIA GTC 2026, SDT debuted its Qubit Controller Unit (QCU), offering 8.5 GHz bandwidth for precise qubit control, ensuring stability within the NVQLink framework. SDT plans a 64-qubit system by late 2026 and photonic quantum computers by 2027, aiming to build a scalable, sovereign quantum-AI infrastructure for global markets.
SDT Launches Korea’s First Commercial Quantum-AI Hybrid Data Center via NVIDIA NVQLink

Summarize this article with:

SDT Launches Korea’s First Commercial Quantum-AI Hybrid Data Center via NVIDIA NVQLink SDT has inaugurated South Korea’s first privately-led Quantum-AI Hybrid Data Center in Gangnam, Seoul, marking a transition from laboratory research to commercial quantum services. The facility houses “Kreo,” a 20-qubit superconducting quantum computer fully integrated with NVIDIA DGX B200 GPU servers. This “full-stack” environment is managed by QuREKA, SDT’s hybrid quantum cloud platform, which orchestrates workloads between classical GPUs and quantum processors (QPUs). The center serves as a commercial hub where GPUs handle data-intensive preprocessing while the Kreo system executes core quantum kernels, targeting industrial applications in materials science, logistics, and financial risk optimization. In March 2026, SDT became the first South Korean company to join the NVIDIA NVQLink Ecosystem, an open architecture designed for high-bandwidth, ultra-low latency communication between GPUs and QPUs. This integration enables real-time feedback loops in microsecond (μs) increments, a critical requirement for advanced Quantum Error Correction (QEC) and the calibration of superconducting qubits. By adopting NVQLink, SDT has synchronized its QuREKA platform with global AI computing standards, allowing enterprise users to integrate quantum-AI workflows into existing infrastructures without the operational overhead of managing heterogeneous hardware. At NVIDIA GTC 2026, SDT unveiled its independently developed Qubit Controller Unit (QCU), a core hardware component providing high-precision multi-qubit control and readout with a signal bandwidth of up to 8.5 GHz. The QCU is engineered to maintain the stability of the Kreo system within the unified NVQLink environment, facilitating deterministic data exchange between the quantum control stack and classical accelerators. Following this launch, SDT’s roadmap includes the deployment of a 64-qubit superconducting system later in 2026 and the development of photonic integrated circuit (PIC) quantum computers by 2027, aimed at establishing a sovereign, scalable quantum-AI infrastructure for the global market. For full details, see the data center opening here, the NVQLink ecosystem announcement here, and the QCU hardware unveiling here. March 17, 2026 Mohamed Abdel-Kareem2026-03-17T18:53:12-07:00 Leave A Comment Cancel replyComment Type in the text displayed above Δ This site uses Akismet to reduce spam. Learn how your comment data is processed.

Read Original

Tags

superconducting-qubits
quantum-optimization
quantum-cloud
quantum-commercialization
quantum-computing
quantum-hardware
quantum-error-correction

Source Information

Source: Quantum Computing Report