The newest release of Entuity Software™ from Park Place Technologies offers users unprecedented oversight of GPU and cooling infrastructure – the only infrastructure monitoring and management solution capable of doing so across three major vendors.
Entuity V23.0 P03 now monitors GPUs from AMD®, Intel®, and NVIDIA®, alongside the servers, interconnect fabrics and liquid cooling systems that keep AI computing environments running.
According to Gartner research on top AI and Gen AI Infrastructure Barriers, a rack full of AI servers can draw 30kW to 120kW or more, compared with 5kW to 15kW for a standard server rack. The resulting heat load is why power and cooling rank among the top barriers to deploying AI infrastructure on premises. Entuity is designed to help IT teams manage this challenge.
Newer AI chips increasingly rely on direct-to-chip cooling, and when cooling conditions degrade, GPUs can throttle or go offline. Monitoring the cooling systems alongside the GPUs gives teams the context to act early, which is why this release extends GPU monitoring to cooling infrastructure, making Entuity the only platform to do so.
Previous versions offered OS monitoring, but Entuity V23.0 P03 allows customers to specify an IP address within an OS and connect to GPUs for more detailed infrastructure insights.
Entuity also delivers switch-level visibility into the interconnect fabrics linking GPUs together, helping enterprises get ahead of the link failures that are a leading cause of GPU cluster downtime. Entuity can flag early signs of link degradation, so IT teams can act before a failure occurs.
Entuity V23.0 P03 helps IT teams identify issues that have been difficult to detect previously, including:
- Expose “gray failures” that traditional monitoring can miss, including hardware conditions that may not trigger availability alarms but can gradually reduce AI infrastructure performance over time.
- Detect GPU hardware degradation before it causes workload failures by monitoring memory health indicators, ECC errors, and other accelerator telemetry that can reveal emerging problems before a GPU goes offline.
- See thermal conditions before they become production issues by correlating GPU temperatures, power consumption, and cooling-system telemetry to help teams detect conditions that can lead to throttling, instability, or downtime.
- Reveal underutilized AI infrastructure through visibility into GPU partitions, accelerator instances, memory usage, and utilization metrics that can uncover stranded accelerator capacity.
Entuity customers with active maintenance contracts can upgrade to V23.0 P03 at no additional cost, while new users can be onboarded directly to this latest version of the software.
Entuity V23.0 P03 Highlights:
AI Accelerator/GPU Monitoring
- Observability for three GPU vendors on one platform; AMD®, Intel®, and NVIDIA®
- Five object types beneath the server: GPU Device; GPU Instance (MIG, AMD partition, SR-IOV); Compute Instance; GPU Link (NVLink, xGMI, Xe Link, Gaudi); and GPU Fabric Switch
- In-band plus out-of-band Redfish through iDRAC and iLO (In-band takes priority, so no GPU is counted twice)
- Monitor ECC corrected and uncorrected; link CRC; replay and recovery; fabric fatal errors; enforced power limit; and PCIe throughput
- 14 thresholds enabled by default
Liquid Cooling Monitoring
- Out-of-the-box support for top vendors including ACCELSIUS™, GRC®, nVent®, and ZutaCore®
- Monitor pump flow, speed, and feedback; supply and return coolant temperature and differential; system and manifold pressure; reservoir level; valve position; and leak detection
- Roughly 415 telemetry streams across accelerators and cooling
To learn more about Entuity Software and how it can enhance enterprise infrastructure observability, visit www.entuity.com.
.
Related
This entry was posted on October 22, 2026 at 8:53 am and is filed under Commentary with tags Park Place Technologies. You can follow any responses to this entry through the RSS 2.0 feed.
You can leave a response, or trackback from your own site.
Entuity Software from Park Place Technologies Launches AI Accelerator and Cooling Monitoring Across AMD, Intel and NVIDIA
The newest release of Entuity Software™ from Park Place Technologies offers users unprecedented oversight of GPU and cooling infrastructure – the only infrastructure monitoring and management solution capable of doing so across three major vendors.
Entuity V23.0 P03 now monitors GPUs from AMD®, Intel®, and NVIDIA®, alongside the servers, interconnect fabrics and liquid cooling systems that keep AI computing environments running.
According to Gartner research on top AI and Gen AI Infrastructure Barriers, a rack full of AI servers can draw 30kW to 120kW or more, compared with 5kW to 15kW for a standard server rack. The resulting heat load is why power and cooling rank among the top barriers to deploying AI infrastructure on premises. Entuity is designed to help IT teams manage this challenge.
Newer AI chips increasingly rely on direct-to-chip cooling, and when cooling conditions degrade, GPUs can throttle or go offline. Monitoring the cooling systems alongside the GPUs gives teams the context to act early, which is why this release extends GPU monitoring to cooling infrastructure, making Entuity the only platform to do so.
Previous versions offered OS monitoring, but Entuity V23.0 P03 allows customers to specify an IP address within an OS and connect to GPUs for more detailed infrastructure insights.
Entuity also delivers switch-level visibility into the interconnect fabrics linking GPUs together, helping enterprises get ahead of the link failures that are a leading cause of GPU cluster downtime. Entuity can flag early signs of link degradation, so IT teams can act before a failure occurs.
Entuity V23.0 P03 helps IT teams identify issues that have been difficult to detect previously, including:
Entuity customers with active maintenance contracts can upgrade to V23.0 P03 at no additional cost, while new users can be onboarded directly to this latest version of the software.
Entuity V23.0 P03 Highlights:
AI Accelerator/GPU Monitoring
Liquid Cooling Monitoring
To learn more about Entuity Software and how it can enhance enterprise infrastructure observability, visit www.entuity.com.
.
Share this:
Like this:
Related
This entry was posted on October 22, 2026 at 8:53 am and is filed under Commentary with tags Park Place Technologies. You can follow any responses to this entry through the RSS 2.0 feed. You can leave a response, or trackback from your own site.