Overview
CURIS (curis.stanford.edu) is a CS undergraduate summer research program. Below are some RAMCloud proposals we may submit for 2012.
Proposals are apparently around 2 paragraphs. We are permitted to submit multiple separate proposals, each of which can be taken up by one or more students. Different proposals will feature a generic RAMCloud paragraph followed by a more specific description of the project(s) in each proposal.
Generic Introduction
The RAMCloud project is building a RAM-based storage systems for future datacenters. The end goal is to make exciting new applications possible by pushing the boundaries of scale and latency in datacenter storage systems. Our goal is to unify the memories of thousands of servers and enable client applications to access and store data in these servers 100 to 1000 times more quickly than is currently possible. This is a large, open-source project headed by Professors John Ousterhout and Mendel Rosenblum, and there are four full-time graduate students currently working on various aspects of the system. We are also all committed to making RAMCloud a robust, production-ready system, rather than just a research prototype. We currently build and test RAMCloud on an 80-node, 320-core Linux cluster with an aggregate of nearly 2 TB of main memory and 20 TB of flash storage, all connected by a high-performance Infiniband network.
Proposal 1: Web Dashboard for Cluster Monitoring & Management
Since RAMCloud is a large and complicated distributed system, being able to monitor and visualize what is going on in real time is critically important. We are looking for students interested in web development and visualization to design and build a dashboard for RAMCloud to help give us better insight into how the system is behaving as a whole. The dashboard will monitor machines in the cluster, aggregate and report server statistics, and allow us to visualize the system in real time. The same dashboard would also be used to manage the system; to add and remove machines from the cluster, migrate data between individual servers, etc. AJAX and javascript skills will be essential. Choosing an appropriate web framework to build on will be up to the CURIS students.
Proposal 2: RAMCloud Core Systems Development
RAMCloud currently consists of about 75,000 lines of C++, docs, and unit tests, but is far from complete. There is a lot of research, design, and development work left to be done. Students will be able to work in whatever part of the system interests them. Potential projects include making more pieces of the system multithreaded to increase server throughput, working on a highly-available and durable store for critical cluster state, extending the simple read/write operations with additional operators, and many more. Strong C++ and systems skills are important, though students skilled in C and familiar with similar object-oriented languages like Java should be able to get up to speed quickly.
Generic Footer
Should probably have a reference to the wiki, and/or lighter RAMCloud articles like the position paper for background. We could also link to the SOSP paper, for the students who look at research publications.