Engineering Papers⌕ Search

Engineering topics

Chen, Dong

Publications and source records attributed to Chen, Dong.

At least 91 records · Page 5

Tracking the Performance Evolution of Blue Gene Systems

— IBM’s Blue Gene supercomputer has evolved through three generations from the original Blue Gene/L to P to Q. A higher level of integration has enabled greater single-core performance, and a larger concurrency per compute node. Although these changes have brought with them a higher overall system peak-performance, no study has examined in detail the evolution of perfor-mance across system generations. In this work we make two significant contri-butions – that of providing a comparative performance analysis across Blue Gene generations using a consistent set of tests, and also in providing a validat-ed performance model of the NEK-Bone proxy application. The combination of empirical analysis and the predictive performance model enable us to not only directly compare measured performance but also allow for a comparison of sys-tem configurations that cannot currently be measured. We provide insights into how the changing characteristics of Blue Gene have impacted on the application performance, as well as what future systems may be able to achieve.

Kerbyson, Darren J.↗

Extended write combining using a write continuation hint flag

A computing apparatus for reducing the amount of processing in a network computing system which includes a network system device of a receiving node for receiving electronic messages comprising data. The electronic messages are transmitted from a sending node. The network system device determines when more data of a specific electronic message is being transmitted. A memory device stores the electronic message data and communicating with the network system device. A memory subsystem communicates with the memory device. The memory subsystem stores a portion of the electronic message when more data of the specific message will be received, and the buffer combines the portion with later received data and moves the data to the memory device for accessible storage.

97 MATHEMATICS AND COMPUTING↗

Distributed parallel messaging for multiprocessor systems

A method and apparatus for distributed parallel messaging in a parallel computing system. The apparatus includes, at each node of a multiprocessor network, multiple injection messaging engine units and reception messaging engine units, each implementing a DMA engine and each supporting both multiple packet injection into and multiple reception from a network, in parallel. The reception side of the messaging unit (MU) includes a switch interface enabling writing of data of a packet received from the network to the memory system. The transmission side of the messaging unit, includes switch interface for reading from the memory system when injecting packets into the network.

97 MATHEMATICS AND COMPUTING↗

Pausing and activating thread state upon pin assertion by external logic monitoring polling loop exit time condition

A system and method for enhancing performance of a computer which includes a computer system including a data storage device. The computer system includes a program stored in the data storage device and steps of the program are executed by a processer. The processor processes instructions from the program. A wait state in the processor waits for receiving specified data. A thread in the processor has a pause state wherein the processor waits for specified data. A pin in the processor initiates a return to an active state from the pause state for the thread. A logic circuit is external to the processor, and the logic circuit is configured to detect a specified condition. The pin initiates a return to the active state of the thread when the specified condition is detected using the logic circuit.

97 MATHEMATICS AND COMPUTING↗

Optimizing TLB entries for mixed page size storage in contiguous memory

A system and method for accessing memory are provided. The system comprises a lookup buffer for storing one or more page table entries, wherein each of the one or more page table entries comprises at least a virtual page number and a physical page number; a logic circuit for receiving a virtual address from said processor, said logic circuit for matching the virtual address to the virtual page number in one of the page table entries to select the physical page number in the same page table entry, said page table entry having one or more bits set to exclude a memory range from a page.

97 MATHEMATICS AND COMPUTING↗

Global synchronization of parallel processors using clock pulse width modulation

A circuit generates a global clock signal with a pulse width modification to synchronize processors in a parallel computing system. The circuit may include a hardware module and a clock splitter. The hardware module may generate a clock signal and performs a pulse width modification on the clock signal. The pulse width modification changes a pulse width within a clock period in the clock signal. The clock splitter may distribute the pulse width modified clock signal to a plurality of processors in the parallel computing system.

97 MATHEMATICS AND COMPUTING↗

Deadlock-free class routes for collective communications embedded in a multi-dimensional torus network

A computer implemented method and a system for routing data packets in a multi-dimensional computer network. The method comprises routing a data packet among nodes along one dimension towards a root node, each node having input and output communication links, said root node not having any outgoing uplinks, and determining at each node if the data packet has reached a predefined coordinate for the dimension or an edge of the subrectangle for the dimension, and if the data packet has reached the predefined coordinate for the dimension or the edge of the subrectangle for the dimension, determining if the data packet has reached the root node, and if the data packet has not reached the root node, routing the data packet among nodes along another dimension towards the root node.

97 MATHEMATICS AND COMPUTING↗

Zone routing in a torus network

A system for routing data in a network comprising a network logic device at a sending node for determining a path between the sending node and a receiving node, wherein the network logic device sets one or more selection bits and one or more hint bits within the data packet, a control register for storing one or more masks, wherein the network logic device uses the one or more selection bits to select a mask from the control register and the network logic device applies the selected mask to the hint bits to restrict routing of the data packet to one or more routing directions for the data packet within the network and selects one of the restricted routing directions from the one or more routing directions and sends the data packet along a link in the selected routing direction toward the receiving node.

97 MATHEMATICS AND COMPUTING↗

Network support for system initiated checkpoints

A system, method and computer program product for supporting system initiated checkpoints in parallel computing systems. The system and method generates selective control signals to perform checkpointing of system related data in presence of messaging activity associated with a user application running at the node. The checkpointing is initiated by the system such that checkpoint data of a plurality of network nodes may be obtained even in the presence of user applications running on highly parallel computers that include ongoing user messaging activity.

97 MATHEMATICS AND COMPUTING↗

Migration of point defects and a defect pair in zinc oxide using the dimer method

The migration mechanism and the minimum energy path of vacancies, interstitials and an interstitial-vacancy pair in zinc oxide have been studied by the dimer method. The in-plane and out-of-plane migrations of zinc and oxygen vacancies are found to be anisotropic. The kick-out mechanism is energetically preferred to zinc and oxygen interstitials that can easily migrate through the ZnO crystal lattice. In addition, the migration process of an interstitial-vacancy pair as a complex of an octahedral oxygen interstitial and a zinc vacancy is dominated by an oxygen interstitial/zinc vacancy successive migration. The energy barriers indicate that the existence of oxygen interstitial in the defect pair can promote the mobility of zinc vacancy, whereas the migration of oxygen interstitial is slowed down due to the presence of zinc vacancy. In the end, we show a possible migration path of the interstitial-vacancy pair that can be dissociated through a set of displacement movements.

Computer simulations↗

Simplifying and speeding the management of intra-node cache coherence

A method and apparatus for managing coherence between two processors of a two processor node of a multi-processor computer system. Generally the present invention relates to a software algorithm that simplifies and significantly speeds the management of cache coherence in a message passing parallel computer, and to hardware apparatus that assists this cache coherence algorithm. The software algorithm uses the opening and closing of put/get windows to coordinate the activated required to achieve cache coherence. The hardware apparatus may be an extension to the hardware address decode, that creates, in the physical memory address space of the node, an area of virtual memory that (a) does not actually exist, and (b) is therefore able to respond instantly to read and write requests from the processing elements.

97 MATHEMATICS AND COMPUTING↗

Managing coherence via put/get windows

A method and apparatus for managing coherence between two processors of a two processor node of a multi-processor computer system. Generally the present invention relates to a software algorithm that simplifies and significantly speeds the management of cache coherence in a message passing parallel computer, and to hardware apparatus that assists this cache coherence algorithm. The software algorithm uses the opening and closing of put/get windows to coordinate the activated required to achieve cache coherence. The hardware apparatus may be an extension to the hardware address decode, that creates, in the physical memory address space of the node, an area of virtual memory that (a) does not actually exist, and (b) is therefore able to respond instantly to read and write requests from the processing elements.

97 MATHEMATICS AND COMPUTING↗

Increasing available FIFO space to prevent messaging queue deadlocks in a DMA environment

Embodiments of the invention may be used to manage message queues in a parallel computing environment to prevent message queue deadlock. A direct memory access controller of a compute node may determine when a messaging queue is full. In response, the DMA may generate an interrupt. An interrupt handler may stop the DMA and swap all descriptors from the full messaging queue into a larger queue (or enlarge the original queue). The interrupt handler then restarts the DMA. Alternatively, the interrupt handler stops the DMA, allocates a memory block to hold queue data, and then moves descriptors from the full messaging queue into the allocated memory block. The interrupt handler then restarts the DMA. During a normal messaging advance cycle, a messaging manager attempts to inject the descriptors in the memory block into other messaging queues until the descriptors have all been processed.

97 MATHEMATICS AND COMPUTING↗

Local rollback for fault-tolerance in parallel computing systems

A control logic device performs a local rollback in a parallel super computing system. The super computing system includes at least one cache memory device. The control logic device determines a local rollback interval. The control logic device runs at least one instruction in the local rollback interval. The control logic device evaluates whether an unrecoverable condition occurs while running the at least one instruction during the local rollback interval. The control logic device checks whether an error occurs during the local rollback. The control logic device restarts the local rollback interval if the error occurs and the unrecoverable condition does not occur during the local rollback interval.

97 MATHEMATICS AND COMPUTING↗

Message passing with a limited number of DMA byte counters

A method for passing messages in a parallel computer system constructed as a plurality of compute nodes interconnected as a network where each compute node includes a DMA engine but includes only a limited number of byte counters for tracking a number of bytes that are sent or received by the DMA engine, where the byte counters may be used in shared counter or exclusive counter modes of operation. The method includes using rendezvous protocol, a source compute node deterministically sending a request to send (RTS) message with a single RTS descriptor using an exclusive injection counter to track both the RTS message and message data to be sent in association with the RTS message, to a destination compute node such that the RTS descriptor indicates to the destination compute node that the message data will be adaptively routed to the destination node. Using one DMA FIFO at the source compute node, the RTS descriptors are maintained for rendezvous messages destined for the destination compute node to ensure proper message data ordering thereat. Using a reception counter at a DMA engine, the destination compute node tracks reception of the RTS and associated message data and sends a clear to send (CTS) message to the source node in a rendezvous protocol form of a remote get to accept the RTS message and message data and processing the remote get (CTS) by the source compute node DMA engine to provide the message data to be sent.

97 MATHEMATICS AND COMPUTING↗

Error correcting code with chip kill capability and power saving enhancement

A method and system are disclosed for detecting memory chip failure in a computer memory system. The method comprises the steps of accessing user data from a set of user data chips, and testing the user data for errors using data from a set of system data chips. This testing is done by generating a sequence of check symbols from the user data, grouping the user data into a sequence of data symbols, and computing a specified sequence of syndromes. If all the syndromes are zero, the user data has no errors. If one of the syndromes is non-zero, then a set of discriminator expressions are computed, and used to determine whether a single or double symbol error has occurred. In the preferred embodiment, less than two full system data chips are used for testing and correcting the user data.

97 MATHEMATICS AND COMPUTING↗

Collective network for computer structures

A system and method for enabling high-speed, low-latency global collective communications among interconnected processing nodes. The global collective network optimally enables collective reduction operations to be performed during parallel algorithm operations executing in a computer structure having a plurality of the interconnected processing nodes. Router devices ate included that interconnect the nodes of the network via links to facilitate performance of low-latency global processing operations at nodes of the virtual network and class structures. The global collective network may be configured to provide global barrier and interrupt functionality in asynchronous or synchronized manner. When implemented in a massively-parallel supercomputing structure, the global collective network is physically and logically partitionable according to needs of a processing algorithm.

97 MATHEMATICS AND COMPUTING↗

Tungsten Cluster Migration on Nanoparticles: Minimum Energy Pathway and Migration Mechanism

Transition state searches have been employed to investigate the migration mechanisms of W clusters on W nanoparticles, and to determine the corresponding migration energies for the possible migration paths of these clusters. The tungsten clusters containing up to four adatoms are found to prefer 2D-compact structures with relatively low binding energies. The effect of interface and vertex regions on the migration behavior of the clusters is significantly strong, as compared to that of nanoparticle size. The migration mechanisms are quite different when the clusters are located at the center of the nanoparticle and near the interface or vertex areas. Near the interfaces and vertex areas, the substrate atoms tend to participate in the migration processes of the clusters, and can join the adatoms to form a larger cluster or lead to the dissociation of a cluster via the exchange mechanism, which results in the adatom crossing the facets. The lowest energy paths are used to be determined the energy barriers for W cluster migrations (from 1- to 4-atoms) on the facets, edges and vertex regions. The calculated energy barriers for the trimers suggest that the concerted migration is more probable than the successive jumping of a single adatom in the clusters. In addition, it of interest to note that the dimer shearing is a dominant migration mechanism for the tetramer, but needs to overcome a relatively higher migration energy than other clusters.

71 CLASSICAL AND QUANTUM MECHANICS, GENERAL PHYSIC↗