Abstract: The present invention provides a signal integrity measurement method and apparatus which allows for signal characteristics to be measured by obtaining samples taken at the midpoint of the data stream. The invention provides a measurement device that is suitable for use in the field to provide a measurement of signal characteristics within transmitted data streams. The invention is particularly suitable for field measurement of signal characteristics of data streams or continuous in-line monitoring of signal characteristics within transmitted data streams. The signal characteristics include, but are not limited to eye opening jitter, noise, slope efficiency, average power and peak-to-peak amplitude.

Abstract: A flexible/foldable fastener carrier, comprising one or more captive fasteners inserted at predetermined positions on the carrier. The positions correspond to fastening points on a separate article being secured by the fastener, permitting the each fastener to align with the fastening point on the article when the carrier is placed on or around the article. Because the fasteners remain captive in the fastener carrier, they cannot be lost or inadvertently dropped into the device to which the article is being secured. After the article has been secured, the fastener carrier is retained in place for use if the article is removed or replaced, or the carrier is removed from the article without disturbing either the secure article or the fasteners.

Abstract: A method for a user process to specify a policy for allocating pages of physical memory on the nodes of a multinode multiprocessor computer system. Through means such as a system call, an application program can specify to the operating system that physical pages of memory for an application-specified portion of virtual address space are to be physically allocated upon a specified set of nodes, subject to the additional selection criteria that the pages are to be allocated at first reference upon: 1) the node upon which the reference first occurs; 2) the node which has the most free memory, or 3) that the pages should be evenly distributed across the indicated set of nodes. In effect, the operating system remembers the specified allocation policy and node set from which the physical pages can be subsequently allocated as established by a system call. Subsequent use of the virtual address space for which the allocation policy is defined results in the memory being allocated accordingly.

Abstract: A method of invalidating shared cache lines such as on a sharing list by issuing an invalidate acknowledgement before actually invalidating a cache line. The method is useful in multiprocessor systems such as a distributed shared memory (DSM) or non-uniform memory access (NUMA) machines that include a number of interconnected processor nodes each having local memory and caches that store copies of the same data. In such a multiprocessor system using the Scalable Content Interface (SCI) protocol, an invalidate request is sent from the head node on the sharing list to a succeeding node on the list. In response to the invalidate request, the succeeding node issues an invalidate acknowledgement before the cache line is actually invalidated. After issuing the invalidate acknowledgement, the succeeding node initiates invalidation of the cache line. The invalidate acknowledgement can take the form of a response to the head node or a forwarding of the invalidate request to the next succeeding node on the list.

Abstract: A multi processor computer system includes processors 102 in processor receptacles 104, and voltage modules 106 and voltage module receptacle 108. The system accepts both direct voltage processors powered directly from the system power plane, and regulated voltage processors which require a voltage regulator 300. The regulated voltage processors require the voltage regulator while the direct voltage processors cannot operate with a voltage regulator. Voltage module control circuit 110 tests and compares processor type signals from the processor receptacles and voltage module type signals from the voltage module receptacles, and produces a voltage validation signal which indicates the presence of a mismatch between any of the processors and its respective voltage module. The voltage validation signal modifies the conventional processor presence signal from the processor receptacles if a mismatch exists, preventing the system from being powered on.

Abstract: A disk storage device includes two data buses 100, 200 for data transfer. Hard disks (HDDs) 0 and 1 are permanently connected to the data buses 100, 200, and hard disks 2 and 3 are selectively connectable to either of the data buses. When a microcontroller (MCU) 32 outputs a specified signal to a data path controller (DPC) 10 in response to a command from a host, multiplexers 20, 22 connect the selectively connectable hard disks 2 and 3 to one of the data buses. When data is written to or read from the hard disks, the write or read can be performed quickly with fewer data buses and a simple device configuration. A parity data generator (PGEN) 24 is connected to both of the buses, receiving new data via one bus and writing new parity data via the other bus.

Abstract: A method for synchronous broadcasting of multiple bytes over a VME bus broadcasts multiple bytes of data across the VME bus using hardware which interfaces between the bus and attached devices. An VME address modifier code is used to identify the type of broadcast and is sent by a master device, without requiring any response from the slave devices. In a first type of broadcast an address location is transmitted over the address bus and a data message is transmitted over the data bus. In a second type of broadcast data messages are transmitted over both the data and the address buses. Multiple broadcast cycles are used to transmit the desired amount of data. An address strobe qualifies the address and data buses for a message broadcast cycle and is used by the receiving slave to clock in the message.

Abstract: An aliasing method and apparatus carried out on a digital computer are provided for generating an object code from a source program. An aliasing data structure is defined and includes an initial empty check-list for each variable in the source program. A pseudo variable is constructed to represent each predefined large alias class. The pseudo variable is attached to the check-list of each variable included in the predefined alias class. Inverse alias checking upon use of a variable is provided using the check-lists for the variable.

Abstract: A data storage medium having the ability to recover from media errors includes a directory located at any desired region of a data storage area with redundant directory pointers at reserved locations. A unique file token is included in the directory entry for each data file and is also included in each extent list where the locations of parts of a fragmented data file are stored. Multiple extent lists are chained in two directions to facilitate reconstruction of the chain if part of the pointer information is lost.

Abstract: Method and apparatus are provided for transmitting a stream of multimedia digital data over a distribution communications network. A multimedia stream server segments the multimedia digital data stream into data blocks on a first boundary and a second boundary. The first boundary is a set number of transport system data packets and the second boundary is a transport system data packet including a timestamp. A scheduler schedules the segmented data blocks for transmission. The multimedia stream server decodes the segmented data blocks to locate the timestamps and matches the transmission of the located timestamp data block with a time value indicated by the timestamp. The set number of transport system data packets can be determined at connection setup and is not a predetermined value for all sessions. At the receiver, batch processing of received multimedia data can be provided.

Abstract: A magnetic data storage disk includes a surface with a coarsely textured head landing zone and a finely textured data storage area. The coarse texture is applied over the fine texture in the landing zone without forming a step in elevation at the boundary. A method for texturing the magnetic disk includes first applying a fine texture to both the data storage area and the landing zone, and then applying a coarse texture to the landing zone only. Texturing apparatus includes peripheral disk clamps permitting axial compliant movement when the opposed disk surfaces are engaged by balanced, pneumatically biased abrading tools. A free abrasive slurry is introduced to pads carried by the tools.

Abstract: Apparatus and method of data recovery are provided for data detection in a partial-response (PR) data channel. The PR data channel includes an analog-to-digital converter (ADC) for providing digital samples of a readback data signal and a voltage controlled oscillator (VCO) timing control coupled to the ADC. A data recovery procedure (DRP) is established responsive to a detected readback error. Using amplitude offset circuitry, amplitude of the readback data signal is selectively adjusted responsive to the detected readback error; and using gain control, a correction current applied to the VCO timing control is changed responsive to the detected readback error.

Abstract: In a data storage method for checksum DASD arrays, files are classified by length and/or other characteristic. Each relatively shorter file is written to an address or contiguous addresses typically on a single DASD of the array. The checksum stored on a checksum DASD is updated by reading the existing checksum, XORing that with the new data and writing the new checksum. Each relatively longer file is subdivided into portions all of the same size, and the number of equal portions is equal to the number of data DASDs where the file is to be written. The portions are interleaved in a stripe of addresses on the data DASDs, and the checksum of the portions is written to the checksum DASD. The characteristic transition length is dynamically varied in order that each interleaved file is provided with contiguous addresses matched to the file size.

Abstract: Apparatus and method for asynchronous gain adjustment are provided for data detection in a partial-response maximum-likelihood (PRML) data channel. The PRML data channel includes an analog to digital converter (ADC) having a normal operating range and a filter, gain and timing control coupled to the ADC. A plurality of samples are detected from the ADC. Each of the detected samples are sequentially compared with predetermined threshold values. The predetermined threshold values include a zero value, and a minimum value and a maximum value of the normal operating range of the ADC. An absolute value of each of the detected samples are sequentially compared with a forth predetermined threshold value. A gain adjustment correction value is determined utilizing the sequentially compared values.

Abstract: A method of removing uncommitted changes made to stored data by a database management system, wherein the system stores a duplicate copy of data prior to effecting changes to the original copy, and upon failure of the system or of a transaction or transactions effecting the changes, replaces the copy containing uncommitted changes by the duplicate copy. If the system fails to replace the original copy (as in the case of a backout failure), the system prevents subsequent attempts by the system to further change the data until the uncommitted changes have been removed by a process of backout failure processing. Backout failure log records are created to track whether the system completed its backout failure processing (END), which transactions had failed backout (SUBSEQUENT), and which transaction was the first to have a backout failure (FIRST). A batch backout utility program uses the set of backout failure log records to bracket the duration of the backout failure for a particular file.

Abstract: Disclosed is a new torus switch with low latency performance improves torus network connection time by trying multipaths in one single high speed operation. This multipath approach can be directed at establishing a connection between two specific nodes over various alternate routes simultaneously. If only one route is available, the multipath approach will find that path instantaneously and establish the desired connection with minimal latency. If several links are available, the multipath method establishes the desired connection over only one of the available links and leaves the other options free to be used by other connections. In addition, routing at intermediate torus network stages improves over the wormhole approach.

Abstract: In a cooperative processing system, a client sends bundles of padded messages to a server which receives the messages, processes them and deallocates the storage for each message piecemeal. A storage manager of the server handles integer multiples of some minimum granularity (n) of storage units, and allows storage to be deallocated in amounts smaller than the amount originally allocated. The client packages and bundles each message into an amount of storage equal to an integer multiple of the server storage granularity, before sending the bundle to the server via a communications transport. The server places the bundle into a single buffer, with each message occupying a corresponding portion of the buffer. Each message is then unbundled for processing in place by the server process to which it was sent, after which its portion of storage is immediately deallocated.

Abstract: Apparatus and method of even harmonic distortion compensation are provided for digital data detection in a disk file. The disk file includes a magneto-resistive (MR) head or other magnetic transducer head and an analog-to-digital converter (ADC) for providing a sampled readback signal. An even harmonic filter is coupled to the ADC for providing a filtered readback signal having only even harmonic components. An adder subtracts the filtered readback signal from the sampled readback signal for providing a compensated output signal. The compensated output signal is applied to a digital channel of the disk file.

Abstract: A method and apparatus are provided for determining an adjustable precompensation value in a partial-response maximum-likelihood (PRML) data channel. The PRML data channel includes an adjustable precompensation function for modulating the write data waveform. A predetermined test pattern is written using write circuitry in the disk file. The predetermined test pattern is insensitive to channel misequalization and provides a reference for gain and timing control. The written predetermined test pattern is read back using read circuitry in the disk file. An error in sample values in the read back predetermined test pattern is identified using logic provided in the PRML data channel while performing a standard read operation of a data sector written with the test pattern. The adjustable write precompensation value is adjusted responsive to identifying an error.

Abstract: A method for accessing a stored data base comprising a number of blocks, each of said blocks containing a plurality of data files and a directory with access keys of the files stored in each of the respective blocks, and comprising means for effecting a sorted ordering of the data files such that the access keys in each block have a predetermined relation. Before the data base is accessed the directories of all blocks are read once in a predetermined order to determine the range of access keys in each directory. Thereafter, during each access operation first of all those directories are sorted out (identified) in which, based on the determined ranges of access keys, the search key of the requested data file might be present. Only blocks corresponding to the sorted out directories are then searched for the requested data file.