Master Advanced Computing and Systems (MSc) postgraduate personal statement example

Personal statement example

Reimaging a trolley of thirty laptops used to take me most of a Friday afternoon. In my job as an IT support technician at a secondary school, I noticed that the bottleneck was not the machines but the network switch in the storeroom, which all thirty were pulling the same image through at once. After I staggered the start times and moved the image to a local cache on a spare desktop, the job finished before the end of lunch. It was a small fix, but it is the kind of problem I most enjoy: understanding where time actually goes inside a system, rather than where people assume it goes. I am applying for postgraduate study in advanced computing and systems because I want to work on that question properly, at the level of processors, memory and parallel software.

My undergraduate degree in Computer Science gave me a reasonable grounding. I did best in the modules on computer architecture, operating systems and concurrent programming, and less well in the more theoretical modules on formal languages, which I now think I underrated. For my final-year project I built a small cluster from four Raspberry Pi boards and studied how a naive matrix multiplication in C behaved as I changed the loop order and introduced blocking. Using the Linux perf tool to read hardware counters, I could see cache misses fall as the blocks were sized to fit the cache, and the speed-up was larger than I had expected. Spreading the work across the four boards with MPI was more humbling: for smaller matrices, communication over the cheap Ethernet switch cost more than the computation saved. Writing up why the scaling curve flattened taught me more than the parts that worked, and made me want to understand interconnects and memory hierarchies in much greater depth than an undergraduate project allows.

Since graduating I have continued reading around the subject. I worked through much of Hennessy and Patterson's Computer Architecture: A Quantitative Approach, which helped me put my project results in context, particularly the discussion of memory hierarchy design. I have also been teaching myself basic CUDA programming on an old graphics card at home, rewriting the same matrix kernel to compare a GPU's behaviour with what I saw on the Pi cluster. My code is not sophisticated, but I am comfortable profiling it and I know what questions to ask when results look wrong.

The school job has developed skills that matter beyond the technical. I support about two hundred staff, many of whom are under pressure in the minutes before a lesson, so I have learned to diagnose quickly, explain clearly and write documentation that a supply teacher can follow. I also manage my own queue of tickets and small projects, such as migrating the staff shared drive, which required planning around term dates and testing permissions carefully before anyone lost access to their files.

Outside work I play bass in a covers band that rehearses weekly, which mostly involves arguing pleasantly about tempo, and once a month I help a neighbour run a repair café in the church hall, where I usually end up replacing laptop batteries or freeing up space on someone's ageing phone. Neither has much to do with high-performance computing, but both keep me practical and patient with people.

At postgraduate level I would like to study parallel and distributed systems, computer architecture and performance engineering, ideally with a research project that involves measuring real hardware. Longer term, I hope to work on systems software or scientific computing infrastructure, where the gap between theoretical and achieved performance still matters. I am ready to move from fixing slow systems by intuition to understanding them rigorously.