The focus of masters research: timing protocol performance. Specifically, how the authentication mechanisms introduced by Network Time Security (NTS) effected the performance of networked time transfer when compared to NTPv4.

My familiarity with the mechanisms of IEEE 1588 and 802.1AS, thanks to my time at the University of New Hampshire Interoperability Laboratory, enabled me to quickly understand NTP and NTS and begin research almost immediately.

An open source implementation of NTS had been previously developed by Cloudflare and was readily available. By augmenting this implementation, known as CFNTS, with Rust's built-in benchmarking framework, Cargo Bench, we were able to measure discrete time intervals on all machines in the secure time domain and compare their performance to unauthenticated NTP.

This study determines both the performance and scalability of NTS, so two separate experiments were developed to complete each of these objectives individually. The first experiment utilized one client connected to one server in an unloaded environment to measure performance of NTS. The second suite implemented a many-to-one pattern where clients transmitted an increasing number of requests per second to a single server in order to simulate a large network and measure scalability of NTS.

After four months, my academic advisor, Doctor Radim Bartos, and I had completed our research work, the "Experimental Analysis of the Performance and Scalability of Network Time Security for the Network Time Protocol".

I am pleased to announce the acceptance of this work into the International IEEE Symposium on Precision Clock Synchronization for Measurement, Control, and Communication (ISPCS), where I presented the findings in Vienna, Austria in October 2022.

The final draft of this research work is available here.
The slides I utilized when I presented this work at ISPCS 2022 are available here.
The augmented version of CFNTS is available here.

This project's goal was to design and construct a low cost, high performance computing (HPC) cluster for use by University of New Hampshire researchers looking to complete experiments involving intense computational components.

The project cluster utilized six rack mounted servers, or nodes, reclaimed by the project's sponsor. These nodes were provisioned with a modern Linux operating system and networked together to form the project's computational cluster.

The project demonstrates it is possible to construct high performance computer clusters using reclaimed technology, while achieving respectable performance at minimal cost.
The complete explanation of the project methodology, performance benchmarking, and conclusions can be found here.

This project was submitted as part of the PEARC 2021 research conference and presented to the Northeast Cyberteam.

My HTML knowledge was limited to a single introductory college course. While I learned basic HTML design principles, my experience needed to be expanded.

Consequently, I designed this personal website to showcase my experience, projects and publications that cannot be included on a resume of reasonable size.

In October 2022, I conducted a website redesign using CSS Grid. GriffinLeclerc.github.io now supports mobile device browsing as well as 4k displays.

The source code can be found here and my resume can be accessed via the "Resume" tab.

When I was a Life Scout in Boy Scout Troop 292, Hooksett NH, a few of my colleagues were underway with their Eagle Scout service projects, which included construction of an ice skating hut on the shore of a local pond, and refurbishment of a popular walking path in the town forest.

I wanted to complete a less traditional project that would immediately and directly impact students belonging to low income families whose access to technology was limiting access to learning opportunities.

After some research, and raising $1500 in grant funding and other efforts, I contacted the executive director of the Computer Technology Assistance Corporation, a non-profit with a similar mission. They allowed me to use their space over three working days to store, assess, and refurbish 147 computers I had collected from members of the public who donated their older systems.

Of the 147 collected, 82 were made fully functional and ready for distribution, 37 were disassembled for functional components, and 28 were disassembled and recycled. All 82 were distributed at no cost to low income families to support their student's educational experience.