Coding techniques for handling failures in large disk arrays

A crucial issue in the design of very large disk arrays is the protection of data against catastrophic disk failures. Although today single disks are highly reliable, when a disk array consists of 100 or 1000 disks, the probability that at least one disk will fail within a day or a week is high. In this paper we address the problem of designing erasure-correcting binary linear codes that protect against the loss of data caused by disk failures in large disk arrays. We describe how such codes can be used to encode data in disk arrays, and give a simple method for data reconstruction. We discuss important reliability and performance constraints of these codes, and show how these constraints relate to properties of the parity check matrices of the codes. In so doing, we transform code design problems into combinatorial problems. Using this combinatorial framework, we present codes and prove they are optimal with respect to various reliability and performance constraints.This paper is a revised and expanded version of material that appeared at the Third International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS III), Boston, MA, March 1989. The work here was supported in part by the National Science Foundation under Grant Numbers MIP-8715235 and CCR-8411954, as well as an AT&T Bell Labs GRPW grant, a Siemens Corporation grant, and an IBM graduate fellowship.     

Nondisjunction of chromosome 15: origin and recombination

Thirty-two cases of uniparental disomy (UPD), ascertained from Prader-Willi syndrome patients (N = 27) and Angelman syndrome patients (N = 5), are used to investigate the pattern of recombination associated with nondisjunction of chromosome 15. In addition, the meiotic stage of nondisjunction is inferred by using markers mapping near the centromere. Two basic approaches to the analysis of recombination are utilized. Standard methods of centromere mapping are employed to determine the level of recombination in specific pairwise intervals along the chromosome. This method shows a significant reduction in recombination for two of five intervals examined. Second, the observed frequency of each recombinant class (i.e., zero, one, two, three, or more observable crossovers) is compared with expected values. This is useful for testing whether the reduction in recombination can be attributed solely to a proportion of cases with no recombination at all (because of asynapsis), with the remaining groups showing normal recombination (or even excess recombination), or whether recombination is uniformly reduced. Analysis of maternal UPD(15) data shows a slight reduction in the multiple-recombinant classes, with a corresponding increase in both the zero- and one-recombinant classes over expected values. The majority, more than 82%, of the extra chromosomes in maternal UPD(15) cases are due to meiotic I nondisjunction events. In contrast, most paternal UPD(15) cases so far examined appear to have a postzygotic origin of the extra paternal chromosome.     

Effects on channeling of radiation damage due to 28 GeV protons

For an irradiation of 2.9 × 10¹⁹ protons at a beam energy of 28 GeV, the channeling minimum yield in a silicon single crystal increased from 2.3% to 4.1%. The radiation damage occurred with a proton fluence of (4.1 ± 1.4) × 10²⁰/cm². The degradation was measured with MeV-range He ions using Rutherford backscattering. The relevance to bent crystal extraction of TeV beams is discussed.     

Uniformity of threshold voltage for MESFETs fabricated on VGF GaAs substrates

Improved uniformity of threshold voltage is shown for MESFETs fabricated on GaAs substrates grown by a novel vertical gradient freeze technique when compared to devices fabricated on LEC GaAs substrates. The improved uniformity is most likely related to the decreased dislocation density and reduced impurity clustering in the VGF material.