Transparent checkpointing middleware for distributed computing
This technology provides a software layer that automatically saves the state of a running computation across many networked computers, without requiring the operating system or the applications themselves to be aware it is happening. It uses a dual-buffer swapping technique so that checkpointing (saving progress) can happen in the background while the computation continues uninterrupted. If a compute node fails or needs to be taken offline, the running processes can be moved to a different machine by remapping their hardware addresses, again transparently. The result is fault-tolerant, migratable distributed computing without rewriting applications or modifying the underlying OS.
What you could build
A middleware library or cloud infrastructure component that provides transparent checkpointing and live migration for high-performance computing (HPC) clusters and large-scale cloud workloads; buyers would be HPC cloud providers, national labs, and enterprises running long-duration batch or simulation workloads.
Who in Virginia should care
Northern Virginia hyperscale cloud operators (AWS, Microsoft, Google) and defense contractors running large HPC simulations (Leidos, SAIC, Booz Allen) would have direct use for transparent fault-tolerant distributed compute infrastructure.
Readiness: Lab validated
Concept — described but not yet demonstrated. Lab validated — supported by experimental results in the patent. Prototype likely — the text describes a built, working embodiment.
Readiness is inferred from the patent text, not from a lab visit.
The record
- Inventors
- Srinidhi Varadarajan, Joseph Ruscio
- Granted
- September 1, 2015
- Status
- Granted patent
- Patent number
- 9122714
Ready to talk?
Virginia Tech Intellectual Properties handles licensing for this technology.
Prosim summaries are generated from public patent text and are not legal advice.