Seminar

In this talk we will give a brief description of the organization of Microsoft Research and its role in Microsoft. In particular, we will discuss some of the on-going projects in MSR Silicon Valley lab, the newest lab of Microsoft Research which focuses on distributed system research. These projects range from anti-spam to specification and verification to distributed storage systems. MSR-SV is constantly looking for talented people and we hope this talk will give prospect candidates a picture of the on-going research activities in MSR-SV

Microsoft is building a next-generation compiler infrastructure that will be the basis for Microsoft's product compilers and other programming tools and a research platform as well. This project, code-named Phoenix, is a joint effort between the Developer Platforms Division and Microsoft Research. The general idea is to build a set of compiler building blocks that are unified using a common intermediate representation. The blocks will be composed to derive various compiler configurations and to support programming tools. The challenge is to span a range of input languages, target architectures, compiler configurations and to produce a useful research platform. We have assembled a team with experience in these different areas. In this talk, I'll give a high-level overview of the project, discussing goals, input languages, target architectures, compilation models, ideas for making it available as a research platform, and progress to date. The intermediate representation is still a work-in-progress, so I will only be giving a general overview of it. I will also discuss some research topics that my group plans to investigate using Phoenix as a research platform.

This talk will describe the goals, design, and current status of two projects named Boxwood and Koh-i-Noor that are in progress at Microsoft Research in Silicon Valley.

Boxwood is exploring the design and implementation of persistent, distributed, and fault-tolerant B-Trees. The goal of the project is to provide a high-performance and scalable "B-Tree Service" that is, in turn, used as the underlying substrate by clients such as distributed file systems and distributed data bases. The service is implemented as a cooperating set of CPUs with locally attached storage interconnected by a high speed network. Providing such a service is an interesting challenge in the presence of failures, load imbalances, and concurrency.

Koh-i-Noor is exploring the use of erasure codes to build reliable disk subsystems that tolerate several disk failures before data loss. Unlike mirroring or triplexing, which double or triple the number of disks required to store data, erasure codes can be very cost-effective in providing equivalent (or higher) reliability. Some of the key challenges we have addressed involve getting good read/write performance in the normal case as well as when data is being reconstructed after a failure.