UW Operating Systems
  Systems Seminar

The talk will discuss the basic architectural elements of a disc drive. It will touch on some of the more important design decisions that have to be made with each new model and try to give some insight into the challenges that will have to be faced in the near future. It will conclude with short summary of the important interfaces used today and the important differences between desktop (ATA) and enterprise (SCSI) drives.

Dave Anderson is Director of Strategic Planning for Seagate and has over 20 years' experience in the computer field. His responsibilities include overall interface strategy for all disc interfaces.

Dave has been involved in the architecture and planning of Fibre Channel since its first proposal as a disc interface. He was also one of the principal architects of the disc XOR commands that are now a part of the standard SCSI interface specification and was the author and original editor of the Object based Storage Device proposal being developed by SNIA for submission to the SCSI standards committee. Dave was one of the original nine elected members of the SNIA (Storage Networking Industry Association) Technical Council. He was also one of the founding members of the Serial Attached SCSI working group, which defined this new interface.

Dave is a member of ACM and the IEEE Computer Society.

Many interesting large-scale systems are distributed systems of multiple communicating components. Such systems can be very hard to debug, especially when they exhibit poor performance. The problem becomes much harder when systems are composed of "black-box" components: software from many different (perhaps competing) vendors, usually without source code available. Typical solutions-provider employees are not always skilled or experienced enough to debug these systems efficiently. Our goal is to design tools that enable modestly-skilled programmers (and experts, too) to isolate performance bottlenecks in distributed systems composed of black-box nodes.

We approach this problem by obtaining message-level traces of system activity, as passively as possible and without any knowledge of node internals or message semantics. We have developed two very different algorithms for inferring the dominant causal paths through a distributed system from these traces. One uses timing information from RPC messages to infer inter-call causality; the other uses signal-processing techniques. Our algorithms can ascribe delay to specific nodes on specific causal paths. Unlike previous approaches to similar problems, our approach requires no modifications to applications, middleware, or messages.

Rising Internet connectivity and emerging standards are driving a new model of federated computing, where software systems will span multiple collaborative components distributed across multiple organizations. The emergence of this computing model is evident at every level of collaborative interactions, including peer-to-peer information sharing, scientific computing grids, and business-to-business ecommerce. However, while federated computing holds great potential for collaboration and sharing across the Internet, its realization faces a number of challenges. In particular, it is difficult to build these systems because they typically span multiple administrative domains, and so their management and control are fundamentally decentralized. Further, behaviors of federated components may vary widely and may change dynamically according to local decisions.

In this talk, I will describe PlanetP, an infrastructure designed to ease the task of building federated software systems. At its lowest level, PlanetP implements an epidemic communication layer that is highly robust to failures (which may be caused by local disconnect decisions). Then, using this information diffusion layer, we have explored several fundamental building blocks for federated systems, including content ranked addressing, autonomous replication for data availability, and a novel federated filesystem. Several principles pervade our design, including autonomous decisions coordinated with only small amounts of loosely synchronized state information and extensive use of randomness to tolerate stale data and unsynchronized concurrent actions.