Today's storage systems can accommodate unheard-of-quantities of raw data, and all the indications are that tomorrow's will hold even more. We have become accustomed to phenomenal growth in aggregate storage capacity, performance, and reliability, and our dependence on digital storage is assured - and probably simply assumed by many people. Does this mean we are "done"? If not, what's missing, and what can we do about it? What's next -- and where might we be able to go in the future?
Science and industry rely on large-scale batch computing to solve important strategic problems with brute force. However, batch computing has never sufficiently supported workloads with significant data needs. A conventional but inadequate solution is to couple a stock batch system to a stock distributed filesystem. This approach is limited in scalability and reliability and has forced users to either limit their data needs or distribute data manually.
To solve this problem, we introduce a new system structure called the Batch-Aware Distributed Filesystem (BAD-FS). Unlike traditional systems, the components of BAD-FS expose their state and policies to an external scheduler. Using this control, the scheduler allocates and orchestrates a personal, data-intensive computing system on the fly. By moving control from the core to the edge, BAD-FS is able to solve several perennial problems in distributed computing, such as space allocation, consistency management, and the safety-performance tradeoff. We demonstrate the utility of BAD-FS with a suite of five scientific workloads in a controlled local cluster and in a wide-area multi-cluster environment.
A difficult but common scenario in computer security is being required to place confidence in trusted software running in an untrusted environment. Remote query of a hardware-based authentication identifier is just such a scenario. Increasingly, anti-tamper and software obfuscation techniques are being employed in an attempt to achieve the level of confidence desired. Recently, the Systems Assessment and Research Center of Sandia National Laboratories evaluated one such system. This talk presents the process of the evaluation, the software's protection mechanisms and what we did to evade them. The talk concludes with some observations about anti-tamper systems, obfuscation, and attackers.
Dr. Ghormley graduated in 1991 with a B.S. with honors in Computer Engineering from Carnegie Mellon University. In 1998, he obtained his Ph.D. in Computer Science from the University of California at Berkeley, specializing in distributed operating systems. He is currently a Principle Member of the Technical Staff at the Systems Assessment and Research Center of Sandia National Laboratories where he is a computer security analyst.