Transferring Large Data Sets and Hands-on Tutorial with the Globus Transfer Tool, 3/28/17
Data is the cornerstone of any scientific research. As the scale of High Performance Computing (HPC) grows, datasets are also getting larger and more diverse. Finding enough storage space is one issue, but another big problem is moving these large datasets (raw or processed) around. It is important that researchers do not spend too much time and effort on transferring datasets from one place to another: (1) It should be easy to start a transfer, and (2) transfer speed should be fast.
First, the course will introduce the basics of dataset transfer. The course will identify possible bottlenecks that can deter fast dataset transfer and show how to overcome them. For example, we will learn why you can get less than 1 Gigabit per second (Gbps) transfer speed with a 10 Gbps network connection, and how to do better. The course will also introduce existing tools and infrastructure that Princeton University has (e.g., dedicated Data Transfer Nodes), which can greatly improve dataset transfer speed.
Then, the course will give a hands-on tutorial on using the Globus transfer tool (https://www.globus.org). Globus is a high-performance dataset transfer tool that is easy to use (clicks on a web browser), and many notable universities, national labs, federal agencies, and national computing facilities support it.
The mini-course assumes no prior experience or knowledge about networking or data transfer practices. Please bring a laptop to fully experience the hands-on tutorial.
Joon Kim is a Cyber Infrastructure Engineer (CIE) in the Princeton Institute for Computational Science & Engineering (PICSciE). His role is to improve the Princeton University's network infrastructure to better support scientific research. He has a Ph.D. in Computer Science from Georgia Tech, specializing in computer networks.
Registration is limited to 24 students. Please register online at the training website, www.princeton.edu/training or contact Andrea Rubinstein at firstname.lastname@example.org /258-1397.
Location: Room 347, Visualization Lab
Date/Time: 03/28/17 at 2:00 pm - 03/28/17 at 3:30 pm