There is definitely a movement afoot to jump on the data band wagon. This is nothing new. But it is becoming more common to hear how librarians are becoming data driven and how catalogers and metadata work with data. The number of workshops on how to handle the data deluge, how to help manage data, etc. continue to proliferate. The discussion that is conspicuously absent is about technical infrastructure. Working with data often means working with big data sets. What are big data sets? There’s a lot of variation. In the admissions office at the university where I work, the average data set is approximately 250,000 rows in a spreadsheet. As for myself, it ranges from 9000 to 12000 records at a time. The biggest problem isn’t the size of the data set in my case but the technical infrastructure that I have to work with to do everything I have to with these data. The cables in my old office were “cat 3” perhaps the original cables from the 70’s. My network speed was 10 mbps! At that time, I could only handle data sets of 2000-3000 records at a time. I moved to an office with better cables and network speeds hovering around 100 mbps. Now I can handle sets around 8000-10000 records at a time. The problem at the high-end of this scale are time outs or the circle of death where I have to restart my machine. In terms of my work computer, it is the base computer that most academic institutions acquire en masse. That means no extra memory, the base processing power and a short lifespan. Having the connection speed and the cables is only half the battle. Working with data and especially big data requires a machine equipped to process that amount of data and the memory to do what is needed to that data and create results that are big or just as big. My work computer is a good and solid machine. But its primary purpose is processing documents suited for your typically office job. As we move towards being data manipulators and data managers, we also need to be wary of the technical infrastructure that supports this work. If we don’t also invest in this technical infrastructure for our work, no amount of skills will be able to work around these hardware issues.