A blizzard is raging outside your window at the Colorado State University campus on January 1. Calculate the amount of snow falling at exactly 09:32:00. Find the temperature and relative humidity at that same moment. Find the number of 911 calls placed in the last 15 minutes. Analyze the data. Now do it all again a millisecond later.
CSU computer science Ph.D. student Matt Malensek understands that challenge and is researching innovative ways to evaluate queries over huge, multidimensional datasets – in real time. His dissertation, titled “Low-Latency, Query-Driven Analytics over Voluminous Multidimensional, Spatiotemporal Datasets,” has received the 2018 IEEE TCSC Outstanding Ph.D. Dissertation Award.
Finding snowflakes in a blizzard of data
Querying large datasets can take a long time. Several things slow down the process. The larger the dataset, the longer it takes to parse. Dataset information may be split up and stored on separate machines, so data takes longer to find and retrieve. Processing subsystems (like Input/Output) are orders of magnitude slower than the CPU, and concurrent queries from other users are a drag on the system.
To overcome these challenges, Malensek is improving the query evaluations, which are done using relational algebra. His novel approach can scale up to accommodate more data volume, machines, and concurrent queries. It can also handle high-throughput, real-time queries, and avoid irritating processing slowdowns. He is finding the optimal way to retrieve data, delivering users the precise data they request, not just faster, but in real time.
The annual IEEE Technical Committee on Scalable Computing (TCSC) Outstanding Ph.D. Dissertation Award recognizes a Ph.D. student who has written an outstanding dissertation in the field of the scalable computing. The award encourages doctoral research that combines theory and practice or makes in-depth technical contributions. Malensek and five other recipients were honored at the IEEE International Conference on Scalable Computing and Communications (ScalCom) 2018 held in October in Guangzhou, China.
Discovering a passion for research
Advised by Sangmi Pallickara and Shrideep Pallickara, Malensek received his Ph.D. from CSU in Spring 2017 and is now an assistant professor in the Department of Computer Science at the University of San Francisco. His research interests are centered around big data, distributed systems, and cloud computing.
“Currently I am working on Internet of Things (IoT) systems that involve communications between small, low-powered devices to speed up data retrieval and analysis. So instead of contacting a website hosted in a data center somewhere to find out if your bus is on time, your smart phone or mobile device would just communicate with the bus directly to find out where it is. The same goes for weather, mapping, and other sorts of geospatial applications where real-time information is useful.”
Reflecting on his CSU computer science graduate experience, Malensek is grateful for the education and opportunities it provided. “My Ph.D. experience at CSU really solidified what I wanted to do with my career by helping me find my passion for research,” he said. “I was incredibly lucky to get to work with Shrideep and Sangmi Pallickara as well as the other talented students and faculty in the department. I miss CSU every day!”