ADMI 2011 Cloudy View of Computing Workshop

Workshop is hosted by Elizabeth City State University and lectures are offered by Indiana University

Introducation
Jerome Mitchell who was an undergraduate at Elizabeth City State University (ECSU) and is now in Indiana University PhD program. Jerome has enrolled in Professor Judy Qiu's CSCI-B649 Cloud Computing and CSCI-B534 Distributed System classes. As part of his B534 class project, Jerome has been developing educational modules, such as virtual appliances, to advance education and training in cloud computing. This project is in collaboration with Professor Renato J. Figueiredo of University of Florida who provides technical guidance in configuration and optimization for virtual appliances at LAN and WAN. Initial appliances are Hadoop and Twister for science applications. The workshop produced a simple prototype where materials are provided and authorized by Indiana University but electronic cloud curriculum resource will be made available to ADMI , A4RC and other universities who wish to include this cloud computing content into their curriculum.

Workshop Development
The participants of the ADMI faculty workshop have diverse backgrounds; so, to better serve them in the area of cloud computing for data-intensive applications, a preliminary discussion was hosted at the ADMI conference during April 14-16, 2011 before an in depth one-week session in June; this provided participants with an understanding of several practical applications, and it scoped the participant’s relevance/expertise domain. The desired competencies for faulty members to acquire and/or refine in cloud computing were:

• Understand and articulate the challenges associated with distributed solutions to large-scale problems, e.g., scheduling, load balancing, fault tolerance, memory and bandwidth limitations, etc.
• Understand and explain the concepts behind MapReduce
• Understand and express well-known algorithms in the MapReduce framework.
• Understand and reason about engineering tradeoffs in alternative approaches to processing large datasets.
• Understand how current solutions to the particular research problem can be cast into the MapReduce framework.
• Explain the advantages in using a MapReduce framework over existing approaches.
• Articulate how adopting the MapReduce framework can potentially lead to advances in the state of the art by enabling processing not possible before.

Hands-on Workshop
The hands-on workshop was June 6-10, 2011. Participants were immersed in a “MapReduce boot camp”, where ADMI faulty members sought introduction to the MapReduce programming framework. The following were themes for five boot camp sessions:

• Introduction to parallel and distributed processing
• From functional programming to MapReduce and the Google File System (GFS)
• “Hello World” MapReduce Lab
• Graph Algorithms with MapReduce
• Information Retrieval with MapReduce

An overview of parallel and distributed processing provided a transition into the abstractions of functional programming, which introduces the context of MapReduce along with its distributed file system. Lectures focused on specific case studies of MapReduce, such as graph analysis and information retrieval. The workshop concluded with a programming exercise (PageRank or All-Pairs problem) to ensure faculty members have a substantial knowledge of MapReduce concepts and the Twister/Hadoop API.

Source of Computing Resources
FutureGrid is an experimental testbed, which uses virtualization and provisioning of Infrastructure-as-a-Service to provide unique capabilities in deploying customized environments for experiments in grid and cloud computing. It has been leveraged to create a self-contained, flexible, plug-and-play educational "virtual appliance". Academic institutions with restrictive computational resources can use virtual appliances enabling students the opportunity to easily experiment with cloud technology. The virtual appliances support multiple virtualization technologies allowing them to run on a variety of resources, including user workstations and desktop grids. The “Cloudy View on Computing” workshop will highlight two specific educational virtual appliances detailing different middleware stacks used actively in clouds: Hadoop and Twister, each with varying data-intensive applications.

DeShea Simon
Hampton University

Timothy Holston
Mississippi Valley State University

Mohammad Hasan
Elizabeth City State University

Yenhung Hu
Hampton University