Arun Buduri (standing), UM computer and information science alumnus, conducted a Big-Data Hands-On Workshop at the department recently.
Managing huge amounts of data can be a challenge for even the most savvy computer scientist. So when University of Mississippi students, faculty and staff got an opportunity to learn more skills from a talented alumnus who is enjoying a successful professional career, they took advantage of it.
Arun Burduri, a 2000 UM alumnus who works as a venture accelerator, conducted the Big Data Hadoop MapReduce Workshop in mid-November. Some 50 undergraduate and graduate students from the Department of Computer and Information Science joined faculty and UM Information Technology staff for the daylong event.
“The workshop’s purpose was to learn the fundamentals of distributed computing hands-on by getting into the internals of one of most popular, open-source ‘Big Data’ tools, Apache Hadoop,” said Byunghyun Jang, assistant professor of computer and information science and co-coordinator of the workshop. “Topics included fundamentals of HDFS-distributed file system, fundamentals of Hadoop, internals of MapReduce, how MapReduce works and the components of the system.”
The training also covered writing and running a basic MapReduce job (in Java), creating a cluster of laptops to run a job in true distributed mode, processing 1 billion rows of data using just laptop(s), tweaking the cluster config parameters to understand its effect on the performance and Google Cloud, said Dawn Wilkins, professor of computer and information science and workshop co-coordinator.
Buduri, who has worked for Nortel Networks, Microsoft, Ingersoll Rand and other companies, helps accelerate early-stage startups in taking their product or business to market. The workshop was received very well, he said.
“I have been conducting these Big Data Hadoop hands-on workshops in the U.S. and India and plan to conduct in Singapore and other countries early 2015,” Buduri said. “The workshop ends with an introduction to Hadoop on Google Cloud platform so the attendees can learn and build bigger solutions on the cloud.”
Buduri said he would definitely enjoy returning to his alma mater to conduct more workshops in the future.
“Depending on the students’ availability and free time for a similar full-day workshop, I’d love to bring them up to speed on some of the latest cloud techniques,” he said. “Google is working with me on my workshops by sponsoring a $500 Google Cloud credit to anyone who attends.”
While such hands-on workshops typically cost hundreds or even thousands of dollars per person elsewhere, Buduri conducts these via a Meetup group free of charge.
“My main goal is to share my Big Data experience and knowledge with the academia and the general public whenever I have spare time,” Buduri said.
For more information, visit http://www.meetup.com/FREE-Big-Data-Hands-On-Workshops/.