Logo

SiTime boosts GPU utilisation in AI data centres

SiTime Corporation has introduced the Elite 2 Super-TCXO® to increase GPU utilization and compute efficiency in AI data centers by delivering better time synchronization. The product targets a $1.5 billion cumulative market by 2030.

“Industry reports show GPU utilization in AI clusters can be as low as 20 to 40 percent—a large and largely hidden tax on AI infrastructure,” said Piyush Sevalia, chief business officer at SiTime. “AI workloads are distributed across GPUs in tightly orchestrated time slots. Even small timing errors force wait cycles to avoid data corruption, and in extreme cases can trigger GPU timeouts and system restarts. Poor synchronization directly caps GPU utilization.”

Sevalia continued, “To address this, the industry is driving towards a target of 10 nanoseconds time synchronization across an AI cluster, down from 1 microsecond today. We collaborated closely with leading AI system architects at hyperscalers and silicon providers and concluded that the right oscillator can significantly improve cluster wide synchronization. That’s why we developed the Elite 2 Super-TCXO. The device delivers sub-nanosecond synchronization, 10X better than target, which is enabled by its exceptional thermal and short-term stability. With these characteristics, Elite 2 minimizes time errors between GPUs, unlocking higher system utilization, greater throughput and better performance per watt. This is the result of SiTime leadership and systems thinking, applied to one of AI’s hardest problems.”

“AI networks must operate with extremely high efficiency to fully utilize expensive GPU resources,” said Sameh Boujelbene, vice president at Dell’Oro Group. “As AI back-end infrastructure refreshes at a much faster cadence than traditional non-accelerated infrastructure, time synchronization accuracy becomes increasingly important to sustaining performance across rapidly evolving data center architectures.”

Elite 2 Super-TCXO delivers better synchronization in the industry's smallest footprint

Key features: 

1 ns time synchronization accuracy—up to 100X better

±2 ppb/°C dF/dT (frequency temperature slope)—up to 25X better

6 × 10⁻¹² Allan Deviation (ADEV)—up to 8X lower

±50 ppb frequency stability over -40 to 105°C—up to 4X better

3.2 mm × 2.5 mm (8 mm²) footprint—up to 2X smaller 

Digital frequency tuning simplifies timing-aware network design

Eliminates activity dips and micro jumps inherent in quartz technology

Resistant to shock, vibration and board bending

ValorC3 Data Centers, provider of edge and digital infrastructure solutions in Tier 2 U.S. data...
Apx Data Centre Solutions is strengthening its ability to drive innovation and growth across...
AMD GPUs are now supported on Dell PowerEdge servers, extending on-premises AI compute capabilities.
12th September 2026 marks the second year of the awareness day, created to help recognise the role...
Integration of Edwin AI with IBM watsonx and Red Hat Ansible Automation Platform aims to unlock...
Rambus PCIe® 7.0 Switch IP with Time Division Multiplexing enables efficient, scalable PCIe...
Acquisition will expand Cognizant's AI builder technology stack with production-grade AI operations...
A new Hoare Lea report finds that data centre planning applications are being delayed by an average...