Advanced Distributed Systems and Networks
600.667

Instructor

• Yair Amir

The course is managed as a few discussion groups, each is focused around a selected research topic. Each group investigates far-reaching ideas, and designs and implements a useful semester-long project related to the topic.

Generally, we have a meeting every Tuesday and Thursday 3:00-4:15pm. These tend to go longer based on interest.
For more information about the course please contact the instructor.

Here are the projects conducted during the Spring 2017 semester:

• Evaluating BFT Protocols for Spire (SCADA for the Power Grid)

• Genie: Distributed Systems Synthesis and Verification

• Distributed Learning in Virtual Realistic Environments

• Emergency Vehicle Early Warning Detection System

Here are the projects conducted during the Spring 2015 semester:

• Toward Open-Source Intrusion-Tolerant SCADA

• Toward Timely, Reliable and Cost-Effective Transport Service Through Dissemination Graphs

• Adding Fault Tolerance Capabilities to K3, a Programming Language for Data Intensive Systems

• Small-Form Computing

Here are the projects conducted during the Spring 2013 semester:

• Water Molecule Simulation on Multi-Core Cluster

• Software Defined Networking and OpenFlow

• Increasing Resiliency in Tor

• K-Path Overlay Routing for Resilient Clouds

• Toward Wireless Mesh with WiFi-Direct Apps

• Movie-Distribution Service on the Internet

• Big Data Consistency

Here are the projects conducted during the Spring 2011 semester:

• Spiny Android - Overlay Networking on Android Phones

• Remote TeleSurgery

• Resilient Dissemination through Flooding

Here are the projects conducted during the Fall 2008 semester:

• Handover Protocol Between Wi-Fi and 3G

• Bluefi - Bringing Bluetooth "Into the Cloud"

Here are the projects conducted during the Spring 2007 semester:

• Fighting E-mail Spam
  Spam is one of the major problems on the Internet, making e-mail a lot less effective tool. Spam filters are quite effective, blocking about 90% of spam messages. However, due to the large volume of spam, 10% is still quite significant compared with the number of non-spam messages. This project had two goals: first, to increase the effectiveness of the filters to 95% (reducing the ratio of unfiltered spam to filtered spam by a factor of 2). Second, to try to exact some cost on computers that send spam at the time they actually send a spam message.

• Privacy for Web-based Applications
  Web based applications such as Google Docs and Spreadsheets represent a new take on the old thin-client / the-network-is-the-computer computing paradigm. Current providers such as Google and Zoho store user files and provide applications that manipulate these files within the framework of an Internet browser. The content of these files is accessible to the provider, presenting a serious privacy problem (or feature, in the eye of the provider). This project investigates the possibility of providing user privacy using the existing applications and application providers, without undermining usefulness.

• A Dynamic Web Service.

• Wireless Mesh Network with Push-to-Talk Support.