Random-access Devices Research Articles

Phases in HfO2-Based Ferroelectric Thin Films and Their Integration in Low-Power Devices

HfO2-based ferroelectrics are applied in a large spectrum of electronic devices ranging from ultralow-power logic to nonvolatile memory. The efficacy of these ferroelectrics is that these offer complementary metal oxide semiconductor compatibility in conjunction with large coercive fields and ferroelectricity at sub-10 nm thicknesses. Because of these advantages compared with conventional lead-based thick perovskite films (>50 nm), the present spotlight article focuses on their use to surpass the physical limit of the subthreshold swing (60 mV/dec at room temperature) of field-effect transistors (FETs) via the stabilization of the negative capacitance. In addition, the discussion on HfO2-ferroelectric-based memories is focused on two-terminal random-access devices, tunnel junctions, three-terminal ferroelectric FETs and their respective 3D stacked architectures.

English

Modern magnetic disks are, as is well known, dramatically slower at random reads than sequential reads. Technological progress has exacerbated the problem; disk throughput has increased by a factor of 60 to 85 over the past twenty-five years, but seek times have decreased by a factor of only 15. Disks are less and less like random access devices in terms of performance. Although flash memory reduces the cost differential of random accesses, disks continue to offer vast amounts of inexpensive storage. For the foreseeable future, it will remain important to optimize the performance of applications that access disk like devices that is, devices with much faster sequential than random access. The best solution to this performance problem is to avoid critical path disk access altogether and reading unneeded data as sequential access and discarding it later. .

Compressed Bit Fail Maps for Memory Fail Pattern Classification

Process optimization in a manufacturing environment for dynamic random access devices (DRAMS) can be improved using fail bit maps in production. Since fail bit maps for 64 Mbit, 256 Mbit and 1 Gbit are quite huge, it is difficult to generate, store and analyze these maps in a manufacturing environment. This paper presents a new scheme for generating compressed bit fail maps during test or from full bit fail maps with minimum loss of fail pattern information. Construction of the special compression scheme for a typical memory array with typical fail patterns will be shown. The described method has been successfully implemented for a 64 Mbit DRAM in a manufacturing environment compressing the bit fail maps to 2 kBit, allowing classification of 13 fail types.

3] Equipment for mass storage and processing of data

The chapter discusses equipment for mass storage and processing of data. The chapter discusses some of the available solutions to the problems related to mass storage and processing of data. Computers become more powerful daily, and tasks that require supercomputers one year are simple on personal computers the next. The chapter tries to provide enough background information to enable users to assess what the future has to offer in an informed and rational way. Confocal images demand substantial storage capacity. Random access devices are disks of one type or another and can move their reading heads to any part of the disk surface to read a particular file without reference to whatever else is stored on the disk. As confocal images are always large, the speed of the device can be an important factor, particularly its sustained transfer rate. Substantial increases in capacity were obtained from floptical technology. Magneto-optical (MO) drives are rewritable media that are based around a hybrid of Winchester and CD technology. Morphometry, segmentation, and measurement of two-dimensional features in images, is a routine part of many imaging laboratories and many two-dimensional measurement features are included among the standard software of most confocal microscopes.

Single-word multiple-bit upsets in static random access devices

Presents the results of an investigation of the SMU (single-word multiple-bit upset) vulnerability of a number of high density SRAM (static random-access memory) device types. The primary objectives of this study were to examine the extent of SMUs in SRAMs, determine design characteristics that predispose devices to this type of upset, and investigate SMU mitigation techniques applicable to space-based electronic systems. The results reported suggest that a nonnegligible SMU rate can be expected for most high-density SRAM types when exposed to the space radiation environment. However, the range of SMU rates is also very large, suggesting that careful selection of device types should be emphasized during the design phase. Furthermore, susceptibility to SEU is not necessarily a reliable indicator of SMU vulnerability. A more important determinant, in many cases, is the architecture of the device, especially the physical separation between logically adjacent cells. It is therefore inappropriate to consider SEU (single event upset) studies to be a proxy for SMU investigations. >

A low-cost, high-speed direct access device

This paper will deal with the novel approach taken to produce a low-cost, high-capacity, high-speed direct access tape drive. The drive utilizes a removable one-half inch magnetic tape cartridge to compliment industry trends. The device will initially store 80 Mbytes per cartridge with an initial data rate of 500 Kbytes per second. The cartridge format is made up of 615 blocks of 131 Kbytes per second. Within each block the tape is motionless, and the drive operates as a random access device with two active heads and sixteen cylinders. A stepping motor, in conjunction with a peripheral drive, is used to access the tape from block-to-block.

Applications of Partitioned Data Sets in Nuclear Physics

The storage of spectral data from nuclear physics experiments and the processing of such data through analysis programs is found to be more convenient and flexible if the Partitioned Data Set (PDS) concept is used. A PDS structure is best imposed on a Word Addressable Data Set (WADS) where data are transferred to or from a random-access device by supplying a word address and word count to I/O routines. A package of programs is described that may be used to process such data. For theoretical applications, a PDS may be used to store amplitudes from reaction calculations or data generated by shell-model programs.

A finite element solution program for large structures

AbstractA straightforward and general computer program for assembling and solving (using Gauss elimination technique) widely sparsed finite element matrix equations with very large bandwidth and capable of handling different degrees‐of‐freedom and variable bandwidth at different nodes, is described herein. The program assembles any type of finite elements having arbitrary number of nodes and each node may have differnt degrees‐of‐freedom. It requires only a small core memory in the computer, although a fast random access device is also needed. The two very important features of this program are (i) it does not store any zero submatrices within the band and (ii) during the solution of equations all operations dealing with zero submatrices within the band are automatically skipped and thus the savings of a considerable amount of disc storage space and computer time can be effected in many cases. Another feature is that many right hand sides can be handled simultaneously. Hence the program is very economical for structures having widely sparsed matrix equations. A listing of the computer program written in FORTRAN IV for CDC 6400 computer is readily available from the authors, but unfortunately could not be given here because of lack of space. The program is so general that it can be used to solve a wide class of finite element problems without actually having to understand fully the techniques behind it.

Least Squares Solution by Diagonal Partitioning

A method has been developed in the Geodetic Survey Squadron to solve very large sets of weighted, linear observation equations on a computer with minimal core storage. The upper triangular portion of a symmetric matrix is partitioned and inverted to obtain the variance-covariance elements and the solution vector. Observation equations are stored externally with the zero elements eliminated. The normal equation partitions are formed while reading one observation equation at a time from an external storage device. The inversion technique requires storage of only two partitions in core with the remaining partitions residing on a random access device. A set of 25,000 observation equations and 2,003 unknowns was solved in 35 hours on an IBM 7094/MOD 1 using 20,000 words of core and external disk storage. Partition size was 91×91. An increase in partition size significantly reduces execution time. The solution can be interrupted and restarted to avoid long continuous computer runs.

Design and Trial of a Computer-Assisted Lesson in Rhythm

For this study, a computer-assisted lesson for teaching selected behaviors that are related to the area of rhythm perception was designed and programed for students majoring in elementary education. The lesson's effects on students were then determined by testing the lesson with subjects from the University of Illinois at Urbana-Champaign. The investigator used the computer-based Education Research Laboratory at the University of Illinois to design and program the lesson, set it into operational form, and test it. The TUTOR language and PLATO III system were used, consisting of a keyset, cathode-ray tube, and random-access slide selector. Also used was a random-access audio device designed for the PLATO IV. Two program objectives were stated for the CAI lesson, called RHYTHM: (1) The student can demonstrate a knowledge of the function of basic rhythmic notation, and (2) the student can demonstrate a knowledge of the relation of rhythmic notation to aural rhythmic patterns. During the two-week experimental program, data were collected by the computer and printed out, then analyzed and reported. The data included: (1) total time spent and amount of program covered, (2) amount of time on each main routine, (3) number of tries and OK or NO responses from the computer, and (4) special keys pressed within an exercise problem. At the end of each lesson session, the investigator interviewed individual students to find out their opinions of the enjoyable and frustrating aspects of the CAI lesson. Participating students generally felt that the lesson was enjoyable and valuable. They especially valued PLATO'S ability to feedback and to define a task, demand input, and quickly judge input. As a whole, students felt they had learned the basic behaviors related to notation encountered in the lesson. Most students felt that certain messages of the audio output were transmitted poorly.

Cost-performance evaluation of memory hierarchies

The stack processing method is used to compare the cost-performance ($/access) between the following: 1) two-level and three-level hierarchies; and 2) hierarchies using random access and serial access memories as a backing store. It was found that busswidth between levels in the hierarchy strongly affects the system cost-performance, and the access time ratio between the backing store and buffer somewhat less. With the technology cost assumptions used in this study, the three-level hierarchy does not have as good a figure of merit ($/access) as the two-level hierarchy unless the access time ratio exceeds about fifty. Also, random access devices are advantageous over serial access devices for backing store applications only when the memory capacity is less than 1 Mbyte. For capacities of 4 Mbyte and 16 Mbyte serial access stores with shift register lengths of 256 bit and 1024 bit, respectively, look favorable.

Hardware, Software and Psyche

EVEN the most retarded observer is aware that computers “have arrived” in advertising. It is a rare conversation currently that is not peppered with acronyms representing one or another computerized scheme for getting the advertising job done. Such a state of the art is encouraging, but perplexing to many people. Unless one's training happens to include notions of “32K storage”, “random access devices”, “matrix eigenvector programs”, and the like, it is relatively easy to become intimidated (or nauseated) by the pace of technology. In perhaps a shy, but nevertheless quite sincere spirit, one feels compelled to wonder just what computers have to do with making advertising decisions. Put another way, one wishes to know the nature and extent of the man‐machine relationship as regards advertising decision making. Does the machine supplant or complement man? To what extent and in what ways? What does the computer leave to the human's judgment? Is the time‐honoured notion of creativity in advertising losing currency?

The USE OF The COMPUTER IN INFORMATION WORK

The type of search programme we can use depends on the way in which the information is stored within the computer. There are two principle techniques for arranging the files of information about your documents—the item file, in which the record is the document, its identity, its title, etc., and all its descriptors; and the inverted file, in which the record is the descriptor followed by the identities of all the documents to which it applies. Generally speaking, an item file is usually held serially, most commonly on magnetic tape, whereas an inverted file is usually held in a random access device. The reasons for this will become clear later when we discuss the programs.

Average Motion Times of Positioners in Random Access Devices

An analysis is made of the average motion times of mechanical positioners based on trapezoidal velocity vs time curves. The results are plotted in terms of dimensionless motion parameters. It is shown that average motion time may be optimized by balancing acceleration and velocity in a proper way. The selection of optimum transmission ratios between motor and load is discussed and demonstrated in an example.

Digital Applications of Thermoplastic Recording

Thermoplastic films' unique combination of thermal, electrical, and optical properties results in a versatile information storage medium which finds both digital and analog applications. As a digital storage medium, thermoplastic is best suited as a file memory where large capacity and rapid read access is a necessity. Reading and writing rates are completely independent. The fabrication of either sequential or random access devices using tapes or plates is possible. These couple high density capability with electron beam agility to provide large capacity storage and outstanding performance with regards to data transfer rates and read access. The recording medium consists of a substrate for mechanical carriage, a conductive film for heat developing, and the thermoplastic film. The recording process consists of writing line charges with a data-modulated electron beam and then developing with heat. The surface grooves which are formed can be read optically with a flying spot scanner. The position feedback Information inherent in the recorded data allows tracking along a single groove regardless of density.