News

GetWorks demonstrates GPU visualization with IBM Instana to optimize AI data center operations

Sunday, March 22, 2026 at 10:40 PM

GetWorks has successfully demonstrated GPU visualization using IBM Instana, aimed at enhancing data center operations for AI workloads. The pilot shows improvements in monitoring hardware utilization within AI infrastructure environments.

Context

Japanese IT solutions provider GetWorks has successfully demonstrated a new GPU visualization capability using IBM Instana to enhance the management of AI-focused data centers. The demonstration centers on the integration of NVIDIA Data Center GPU Manager (DCGM) metrics into the Instana observability platform, providing real-time visibility into power consumption, thermal performance, and compute utilization. This advancement allows operators to optimize resource allocation and identify performance bottlenecks within high-density AI infrastructure, which often utilizes Supermicro liquid-cooled systems. Industry interest in this integration is driven by the growing complexity of hybrid and multi-cloud AI workloads. By utilizing OpenTelemetry (OTel) collectors and the NVIDIA GPU Operator, IBM enables automated monitoring of the entire stack, from hardware drivers to high-level application performance. While the original report indicates that the demonstration provided "a solid response to the sophisticated operations required in the AI era," specific quotes from individual executives at the event could not be independently verified via existing public documentation. The development marks a significant step in IBM's strategy to position Instana as a critical tool for GPU resource optimization.

Related Companies

IBM
IBM
IBM