Making AI Multitenancy Measurable: A Practical Approach to Job Completion Time SLAs – Keysight Triple-T
In multitenant AI clusters, bandwidth is not the SLA—Job Completion Time (JCT) is. In this joint session, NextHop.ai and Keysight present a real-world study of how load balancing modes, ECN thresholds, and congestion control tuning directly impact JCT in shared AI Ethernet fabrics. Using production-class switching and traffic emulation, we demonstrate how overly conservative settings can double completion times and waste GPU cycles—and how systematic tuning makes JCT predictable. Attendees will gain a practical framework for designing and validating AI networks around measurable performance outcomes, not just throughput.
Sushil Srinivasan is a Director of Product and Business Development at Keysight
Technologies, where he focuses on AI data center and high-speed networking solutions.
He works closely with hyperscalers, enterprises, and partners to translate emerging AI
infrastructure challenges into practical validation and benchmarking strategies. With a
background spanning product management, cloud platforms, and go-to-market
execution, Sushil brings a strong blend of technical depth and business perspective. He
is passionate about helping customers and partners design, test, and scale next-
generation AI networks with confidence.
