Abstract: A deterministic, non-deadlocking technique to achieving distributed consensus in a multithreaded multiprocessing computing environment is provided. A communicator is established across multiple processes in the multithreaded computer environment notwithstanding that multiple groups of threads may be simultaneously trying to establish communicators. The technique includes communicating across the multiple processes to establish a candidate identifier for the communicator for a group of participating threads of the multiple processes; and communicating across the multiple processes to check at each participating thread of the multiple processes whether the candidate identifier can be claimed at its process, and if so, claiming the candidate identifier as the new identifier thereby establishing the communicator. As one example, the technique can be implemented via a subroutine call within a message passing interface (MPI) library.

Abstract: A facility is provided for managing a service connection between a client process having a single threaded library and a server process. The facility includes monitoring activity of a client code of the client process by the client library, and commensurate therewith monitoring activity of the client library across a service connection to the server process, wherein activity of the client library on the service connection is indicative of activity of the client code. The service connection between the client process and the server process is terminated if a message is not received at the server process from the single threaded client library within a defined tolerance interval. Upon termination of the service connection, any locked client resource is recovered at the server process.

Abstract: The modular exponentiation function used in public key encryption and decryption systems is implemented in a standalone engine having at its core modular multiplication circuits which operate in two phases which share overlapping hardware structures. The partitioning of large arrays in the hardware structure, for multiplication and addition, into smaller structures results in a multiplier design comprising a series of nearly identical processing elements linked together in a chained fashion. As a result of the two-phase operation and the chaining together of partitioned processing elements, the overall structure is operable in a pipelined fashion to improve throughput and speed. The chained processing elements are constructed so as to provide a partitionable chain with separate parts for processing factors of the modulus. In this mode, the system is particularly useful for exploiting characteristics of the Chinese Remainder Theorem to perform rapid exponentiation operations.

Abstract: An updating capability updates a component of a computing environment, while maintaining the availability of the computing environment. In particular, a component of the computing environment which is associated with at least a portion of a unit of work is updated from one version to another version. Subsequently, the updated component emulates the original version, or operates in a compatibility mode, while other components of the computing environment, which are similarly associated with the same unit of work, remain at the original version.

Abstract: Electronic system components are cooled using a field configurable pump comprising a selectable plurality of identical or nearly identical modular pump units. The pump units are disposed in a stacked arrangement and are fluidly connected to adjacent modular pump units via a removable coupling, preferably one of the quick-disconnect variety. Each modular pump unit comprises a housing, an impeller and a sealing mechanism. When disposed in a stack the combined pump structure is driven by a single motor. In this fashion, pump structures may be employed which are adaptable to heat flux demands imposed by the electronic components.

Abstract: A non-integer fractional divider divides a reference clock signal having period P by a non-integer ratio K. The divider includes multiplexers to receive a plurality N of clock signals wherein each clock signal is equally phase shifted by a P/N delay. Incrementers coupled to the multiplexers select first and second clock signals between the N clock signals. Such that the phase shift delay between the two selected clock signals is representative of the non-integer value of K. The selected clock signals are combined to output a divided clock signal. The enabling time of each selected clock signal is respectively representative of the duration of the low level and the high level of the divided clock signal.

Abstract: A system, method and computer readable medium for deferring copy-on-write of a snapshot is disclosed. The method includes the generation of snapshot of a source file. Upon modification of a first data block referenced by the source file, the first data block is referenced by the snapshot and a second data block is allocated for the source file. Then, a first variable associated with the source file is set to a value indicating an incomplete source file data block and a second variable associated with the source file is set to a value indicating the valid portion of the second data block. Any portion of the second data block that is overwritten is considered valid. The second data block is then modified and the second variable is changed to reflect the modification. Upon reception of a read request, the corresponding portion of the second data block is retrieved.

Abstract: In a transmission protocol in which a user running an application in an address space in one data processing system wishes to transmit a data packet to another address space in another data processing system by means of direct memory access directly from a sending buffer to a receiving buffer with no copy, a mechanism is provided for minimizing the need for retransmission and for insuring proper entry into the target data processing system address space. In particular, when the first system does not receive an acknowledgment from the receiver, a special data packet with a retransmit flag bit set is sent to the second system. When this system receives the data packet with the retransmit flag bit set the second system responds either by sending a new acknowledgment or by sending a request for retransmission.

Abstract: A method and apparatus for managing communications in a distributed computing system wherein a plurality of computers are interconnected with a communications network. At least a portion of the network is established as a subnetwork with at least one node attached to the subnetwork. The node in the subnetwork multicast a verification message over the subnetwork in which it is attached. A subnetwork leader for each subnetwork is identified and gathers the verification message sent from each node in the subnetwork. The subnetwork leader assembles each verification received into a subnetwork list. The subnetwork list is sent a node identified as the network leader. The network leader assembles each subnetwork list received into a master list and sends this master list out to the subnetwork leaders. In turn, the subnetwork leaders for each subnetwork multicast the master list out to each node. In one embodiment, the communications between each subnetwork leader and the network leader is over multicast.

Abstract: A method for maintaining the reachability of IP multicast communications across a communications network with one or more subnetworks wherein each subnetwork comprises at least one node coupled thereto and a node identified as a subnetwork leader and the communications network comprises one node identified as a network leader. The method comprising the steps of: receiving a host address list containing a list of all the network addresses of the, nodes in a network including at least two subnetworks, and a node in each subnetwork identified as a subnetwork leader; using the host address list for determining which subnetworks the subnetwork leaders are associated therewith. The reachability of at least one subnetwork leader in the network is periodically monitored.

Abstract: Various components are provided to manage a clustered environment. These components include a System Registry that provides a global data storage; a Configuration manager that stores data locally on nodes of the clustered environment and globally within the System Registry; a Liveness component to provide status of communications paths of the cluster; a Group Services component that provides services to one or more other components of the clustered environment; and a Resource Management component that communicates with one or more resource controllers of the clustered environment. Each of the components of the clustered environment has one or more data and/or functional dependencies on one or more other components of the environment. However, relationships between the components are created such that the data and functional dependencies form an acyclic graph (i.e., a cycle of dependency relationships is avoided).

Abstract: At least a portion of a table stored in a global data repository is selected to be accessed and modified. The selected portion is stored in a local table in local memory of a node of a client-server environment. The local table is modified, and then at least a portion of the modified local table is merged with the global table. The merger is performed without requiring any user locking of the global table.

Abstract: According to the present invention, an image processing apparatus includes an original image data input unit 11 for receiving the original low resolution image data; a connection detector 13 for detecting a connection of a target pixel in the original image data and neighboring pixels; a rule application unit 17 for employing a relationship between a coordinate position of a pixel in the original image data and a coordinate position of a resultant pixel in the expansion image data to define a plurality of types into which the relationship can be sorted, and for applying a specific rule for the plurality of types; and a results output unit 19 for determining the value of the resultant pixel in accordance with the rule, and for outputting the expansion image data, wherein, when the connection detector 13 ascertains that the target pixel constitutes a one-pixel width line, the rule application unit 17 applies a rule that will maintain the one-pixel width even in the expansion image data.

Abstract: A printed circuit board assembly which is capable of being inserted and removed from the system to which it is electrically connected, even when fully populated with connected printed circuit cards, comprises a multilayer structure. A middle layer printed board is sandwiched between a nonconductive base on the bottom and a stiffener disposed on the top. The stiffener is preferably electrically conductive to provide an EMI (electromagnetic interference) shielding function. The base support and stiffener together support a mechanism for insertion and removal of the entire board via an edge connector disposed along a back edge of the board.

Abstract: A method is provided for limiting access to disks and files in the event of a system partitioning occurring as the result of a node failure in a data distributed processing system in which files are shared amongst a plurality of independently running nodes. The nodes in the system are provided with limited duration leases which permit data access and which have to be periodically renewed. Each partition is also associated with a quorum which is also a necessary element in providing access at proper times. The lease and quorum concepts work together to insure that recovery operations do not occur until recovery is possible. The lease concept thus is also seen as further operating in close cooperation with the quorum concept to prevent long delays in the recovery process that could otherwise occur.

Abstract: Method, system and program storage device are provided for monitoring and ameliorating congestion in a tightly coupled network. Commensurate with sending a packet into the network, a first time stamp is recorded. Upon receipt of an acknowledgment back across the network responsive to sending of the packet, a second time stamp is recorded. The round trip time of the packet is determined and an amount of congestion is estimated using the determined round trip time and a statically predetermined round trip representative of at least one of no network congestion or a known degree of network congestion. The number of flow control tokens for the destination node can be dynamically varied in response to the estimate of the amount of network congestion. If desired, monitoring and estimating of network congestion can be initiated only after identifying the existence of network congestion, for example, represented by a lack of flow control tokens at a sender node for a destination node.

Abstract: External job scheduling capabilities are provided to a local job control system lacking or having insufficient job scheduling capabilities. This is accomplished by encapsulating running of a user job by the local job control system (LJCS) within running of a marker job at a node management system, which is responsive to an external resource scheduler. The technique includes starting a marker job by the resource scheduler external to the local job control system, wherein the marker job corresponds to a user job to be run by the LJCS; responsive to starting of the marker job, starting the user job by the LJCS; and upon termination of the user job, ending the marker job started by the resource scheduler, wherein starting of the user job by the LJCS is responsive to the resource scheduler's starting of the marker job and wherein ending of the marker job is responsive to termination of the user job.

Abstract: One or more tables stored in a global data repository are concurrently accessible and modifiable. A client requests the locking of one or more tables within a global server data repository on behalf of a local tree. This locking creates a lock block, which enables at least a portion of the locked tables to be selected, modified and then merged back with the tables residing in the server. One or more other clients may also request to use the same lock block and thus, the same local tree, if shared concurrency is allowed. These other clients would also be allowed to select, modify and merge at least a portion of the tables. The merging of the tables is serialized at the global data repository, if necessary.

Abstract: An electromagnetic shield system for cartridge-based printed circuit cards inserted into a printed circuit mother board comprises a card level system and a board level system which are electrically connected upon card insertion to provide more effective EMI shielding system. Prongs on a card level tab provide continuous EMI shielding even as card insertion operations are taking place. The principal linkage between the card level and board level systems is provided by a flexible shield strip which mates with apertures in a conductive board stiffener and simultaneously contacts not only a conductive cartridge bezel, but also contacts an EMI shield plate disposed on the printed circuit card.

Abstract: An apparatus and method is provided for correcting data words resulting from a package fail within a memory array in which coded data is divided into a plurality of multi-bit packages of b bits each. The coded data comprises n-bit words with r error correcting code bits and n-r data bits. The invention is capable of correcting one package which has suffered at least one hard failure. The invention correcting exploits single error correcting (SEC)-and double error detecting (DED) codes, requiring no additional check bits, which give a syndrome when the data word has suffered an error coming from at least one error in a package.