Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “SOCKET”

Search indexed NASA NTRS and DOE OSTI research on propulsion, heat transfer, battery materials and energy systems. Follow report and document links to the original sources.

Quote a phrase for an exact phrase match. Source license links do not imply unrestricted reuse.

At least 91 records · Page 5

Optoelectronic system with a wedge-shaped adapter

Optoelectronic systems with an adapter and methods of manufacturing or assembling the same are provided. An example of an optoelectronic system according to the present disclosure includes a substrate, an interposer, an electronic component disposed on the interposer, and an optical component. The optoelectronic system includes a ferrule and an optical fiber coupled to the ferrule. The optoelectronic system also includes an optical socket configured to receive the ferrule therein. The optoelectronic system further includes an adapter positioned between the interposer and the optical socket. The adapter has a wedge-shaped configuration such that the ferrule is disposed at a non-zero angle relative to the interposer when the ferrule is received in the optical socket and the optical socket is coupled to the adapter.

Mathai, Sagi Varghese↗

NUMA-Aware Thread Scheduling for Big Data Transfers over Terabits Network Infrastructure

The evergrowing trend of big data has led scientists to share and transfer the simulation and analytical data across the geodistributed research and computing facilities. However, the existing data transfer frameworks used for data sharing lack the capability to adopt the attributes of the underlying parallel file systems (PFS). LADS (Layout-Aware Data Scheduling) is an end-to-end data transfer tool optimized for terabit network using a layout-aware data scheduling via PFS. However, it does not consider the NUMA (Nonuniform Memory Access) architecture. In this paper, we propose a NUMA-aware thread and resource scheduling for optimized data transfer in terabit network. First, we propose distributed RMA buffers to reduce memory controller contention in CPU sockets and then schedule the threads based on CPU socket and NUMA nodes inside CPU socket to reduce memory access latency. We design and implement the proposed resource and thread scheduling in the existing LADS framework. Experimental results showed from 21.7% to 44% improvement with memory-level optimizations in the LADS framework as compared to the baseline without any optimization.

97 MATHEMATICS AND COMPUTING↗

Exascale Hardware Architectures Working Group

The ASC Exascale Hardware Architecture working group is challenged to provide input on the following areas impacting the future use and usability of potential exascale computer systems: processor, memory, and interconnect architectures, as well as the power and resilience of these systems. Going forward, there are many challenging issues that will need to be addressed. First, power constraints in processor technologies will lead to steady increases in parallelism within a socket. Additionally, all cores may not be fully independent nor fully general purpose. Second, there is a clear trend toward less balanced machines, in terms of compute capability compared to memory and interconnect performance. In order to mitigate the memory issues, memory technologies will introduce 3D stacking, eventually moving on-socket and likely on-die, providing greatly increased bandwidth but unfortunately also likely providing smaller memory capacity per core. Off-socket memory, possibly in the form of non-volatile memory, will create a complex memory hierarchy. Third, communication energy will dominate the energy required to compute, such that interconnect power and bandwidth will have a significant impact. All of the above changes are driven by the need for greatly increased energy efficiency, as current technology will prove unsuitable for exascale, due to unsustainable power requirements of such a system. These changes will have the most significant impact on programming models and algorithms, but they will be felt across all layers of the machine. There is clear need to engage all ASC working groups in planning for how to deal with technological changes of this magnitude. The primary function of the Hardware Architecture Working Group is to facilitate codesign with hardware vendors to ensure future exascale platforms are capable of efficiently supporting the ASC applications, which in turn need to meet the mission needs of the NNSA Stockpile Stewardship Program. This issue is relatively immediate, as there is only a small window of opportunity to influence hardware design for 2018 machines. Given the short timeline a firm co-design methodology with vendors is of prime importance.

97 MATHEMATICS AND COMPUTING↗

Magnetic electrical connectors for biomedical percutaneous implants

A biomedical percutaneous connector is described which includes a socket having an enlarged disk shaped base portion for being implanted below the patient's skin and cylindrical portion which is integral with the base portion and extends outwardly of the skin. A conical recess in an upper end of the cylindrical portion has a magnet located in the base. Inclined conductive strips are carried on an upper end of the cylindrical portion to which electrical conductors are attached and extend into the patient's body. A complementary shaped plug which also has electrical contacts provided thereon is adapted to fit within the conical recess of the socket. The plug is held in the socket by magnetic force.

Owens, L. J.↗

Mechanical energy storage device for hip disarticulation

An artificial leg including a trunk socket, a thigh section hingedly coupled to the trunk socket, a leg section hingedly coupled to the thigh section and a foot section hingedly coupled to the leg section is outlined. A mechanical energy storage device is operatively associated with the artificial leg for storage and release of energy during the normal walking stride of the user. Energy is stored in the mechanical energy storage device during a weight-bearing phase of the walking stride when the user's weight is on the artificial leg. Energy is released during a phase of the normal walking stride, when the user's weight is removed from the artificial leg. The stored energy is released from the energy storage device to pivot the thigh section forwardly about the hinged coupling to the trunk socket.

Vallotton, W. C.↗

Ball-joint grounding ring

In ball and socket joint where electrical insulator such as polytetrafluoroethylene is used as line to minimize friction, good electrical contact across joint may be needed for lightning protection or to prevent static-charge build-up. Electrical contact is maintained by ring of spring-loaded fingers mounted in socket. It may be useful in industry for cranes, trailers, and other applications requiring ball and socket joint.

Aperlo, P. J. A.↗

Quartz ball valve

A ball valve particularly suited for use in the handling of highly corrosive fluids is described. It is characterized by a valve housing formed of communicating segments of quartz tubing, a pair of communicating sockets disposed in coaxial alignment with selected segments of tubing for establishing a pair of inlet ports communicating with a common outlet port, a ball formed of quartz material supported for displacement between the sockets and configured to be received alternately thereby, and a valve actuator including a rod attached to the ball for selectively displacing the ball relative to each of the sockets for controlling fluid flow through the inlet ports.

Goetz, C.↗

Padded Allen Wrench Grips Fastener

Addition of PTFE pressure pad increases utility of hexagonal socket wrench. Pad presses against opposing inside socket walls, causing frictional force that retains socket-head screw against gravitational and handling forces. Developed for spacecraft repairs, tool is useful wherever tight clearance prevents insertion or removal of screw by hand.

Salisbury, M. K.↗

Passive Capture Joint with Three Degrees of Freedom

A passive capture joint with three degrees of freedom is presented wherein two structural elements are joined together solely by moving the two elements into position, and which when joined together have rotation in all three axes. The inventive apparatus is comprised of two halves: (1) a joint ball mounted on a stem as in a common trailer hitch, and; (2) a socket. The socket consists of a base having an exterior wall and forming an interior chamber, the chamber having a top end and a bottom end, and an interior wall. The chamber is open at the top end, and forms a spherical cup at the bottom end. The socket base's interior chamber is sized to accept the joint ball. The base also forms at least one bore at an acute angle away from the interior chamber's open end. The bores have a first opening in the interior wall of the chamber, and a second opening in the exterior wall of the base. Retaining balls sized to fit within the bores, but to only partially pass through the first opening, are moveably housed within the bores. The retaining balls are moveably held in the first opening by a compression spring housed in the bore. As the joint ball is inserted in the chamber it forces the retaining balls back into the bore until the equator of the joint ball passes. Because the bore is at an acute angle to the chamber the joint ball cannot exit the chamber without the joint being unlocked. The joint is unlocked by rotating a locking ring which encircles the base and covers the second opening. The locking ring has a radial slot for each retaining ball, disposed angularly from the base, and sized to allow passage of the retaining ball in the radial direction when the locking ring is rotated to align the radial slot with the second opening.

Richard A Cloyd↗

Low-Friction, High-Stiffness Joint for Uniaxial Load Cell

A universal-joint assembly has been devised for transferring axial tension or compression to a load cell. To maximize measurement accuracy, the assembly is required to minimize any moments and non-axial forces on the load cell and to exhibit little or no hysteresis. The requirement to minimize hysteresis translates to a requirement to maximize axial stiffness (including minimizing backlash) and a simultaneous requirement to minimize friction. In practice, these are competing requirements, encountered repeatedly in efforts to design universal joints. Often, universal-joint designs represent compromises between these requirements. The improved universal-joint assembly contains two universal joints, each containing two adjustable pairs of angular-contact ball bearings. One might be tempted to ask why one could not use simple ball-and-socket joints rather than something as complex as universal joints containing adjustable pairs of angularcontact ball bearings. The answer is that ball-and-socket joints do not offer sufficient latitude to trade stiffness versus friction: the inevitable result of an attempt to make such a trade in a ball-and-socket joint is either too much backlash or too much friction. The universal joints are located at opposite ends of an axial subassembly that contains the load cell. The axial subassembly includes an axial shaft, an axial housing, and a fifth adjustable pair of angular-contact ball bearings that allows rotation of the axial housing relative to the shaft. The preload on each pair of angular-contact ball bearings can be adjusted to obtain the required stiffness with minimal friction, tailored for a specific application. The universal joint at each end affords two degrees of freedom, allowing only axial force to reach the load cell regardless of application of moments and non-axial forces. The rotational joint on the axial subassembly affords a fifth degree of freedom, preventing application of a torsion load to the load cell.

Lewis, James L.↗

Anti-Rotation Device Releasable by Insertion of a Tool

A drive mechanism enables a socket-type wrench to rotate a shaft and prevents accidental rotation of the shaft when the wrench is not coupled to the shaft. In the original intended application, the shaft would be part of an attachment mechanism on a spacecraft, and the purpose to be served by the drive is to prevent back-driving of the shaft by launch vibrations while enabling an astronaut equipped with the appropriate wrench to actuate the shaft while in orbit. The design could also be adapted to terrestrial applications in which it is necessary to prevent rotational back-driving. The mechanism includes a gear near the tip of the shaft, and a drive nut that constitutes the tip of the shaft. The gear and drive nut are positioned in a recess in a housing. The recess is sized to receive the wrench socket that mates with the drive nut. Also contained in the housing are four linkages that include pins that are spring-loaded into engagement with the gear to prevent rotation of the shaft. When the wrench socket is inserted in the recess, it pushes on the linkages in such a manner as to disengage the pins from the gear.

Warden, Harry K.↗

SiC-Based Miniature High-Temperature Cantilever Anemometer

The figure depicts a miniature cantilever-type anemometer that has been developed as a prototype of compact, relatively nonintrusive anemometers that can function at temperatures up to 600 C and that can be expected to be commercially mass-producible at low cost. The design of this anemometer, and especially the packaging aspect of the design, is intended to enable measurement of turbulence in the high-temperature, high-vibration environment of a turbine engine or in any similar environment. The main structural components of the anemometer include a single-crystal SiC cantilever and two polycrystalline SiC clamping plates, all made from chemical-vapor-deposited silicon carbide. Fabrication of these components from the same basic material eliminates thermal-expansion mismatch, which has introduced spurious thermomechanical stresses in cantilever-type anemometers of prior design. The clamping plates are heavily oxidized to improve electrical insulation at high temperature. A cavity that serves as a receptacle for the clamped end of the cantilever is etched into one end of one clamping plate. Trenches that collectively constitute a socket for a multipin electrical plug (for connection to external electronic circuitry) are etched into the opposite end of this clamping plate. Metal strips for electrical contact are deposited on one face of the other clamping plate. Piezoresistive single-crystal SiC thin-film strain gauges are etched in the n-type SiC epilayer in a Wheatstone-bridge configuration. Metal contact pads on the cantilever that extend into the clamping-receptacle area, are obtained by deposition and patterning using standard semiconductor photolithography and etching methods. The cantilever and the two clamping plates are assembled into a sandwich structure that is then clamped in a stainless-steel housing. The Wheatstone- bridge carrying SiC cantilever with the metal contact pads on the piezoresistors is slid into the receptacle in the bottom clamping plate. The top clamping plate is brought into contact with the bottom plate so that the narrow section of the metal strips on the top clamp plate aligns with the metal contact pads on the cantilever. When the parts are clamped together, the metal strips provide electrical connections between the Wheatstone-bridge contact points and the sides the trenches that constitute the socket for the multipin electrical plug. Hence, to connect the Wheatstone bridge to external circuitry for processing of the anemometer readout, one need only insert the plug in the socket. In operation, the cantilever end of the stainless-steel housing is mounted flush with an engine wall and the unclamped portion of the cantilever is exposed into the flow. The cantilever is deflected in direct proportion to the force induced by component of flow parallel to the engine wall and perpendicular to the broad exposed face of the cantilever. The maximum strain on the cantilever occurs at the clamped edge and is measured by the piezoresistors, which are located there. The corresponding changes in resistance manifest themselves in the output of the Wheatstone bridge.

Okojie, Robert S.↗

Multi-User Space Link Extension (SLE) System

The Multi-User Space (MUS) Link Extension system, a software and data system, provides Space Link Extension (SLE) users with three space data transfer services in timely, complete, and offline modes as applicable according to standards defined by the Consultative Committee for Space Data Systems (CCSDS). MUS radically reduces the schedule, cost, and risk of implementing a new SLE user system, minimizes operating costs with a lights-out approach to SLE, and is designed to require no sustaining engineering expense during its lifetime unless changes in the CCSDS SLE standards, combined with new provider implementations, force changes. No software modification to MUS needs to be made to support a new mission. Any systems engineer with Linux experience can begin testing SLE user service instances with MUS starting from a personal computer (PC) within five days. For flight operators, MUS provides a familiar-looking Web page for entering SLE configuration data received from SLE. Operators can also use the Web page to back up a space mission's entire set of up to approximately 500 SLE service instances in less than five seconds, or to restore or transfer from another system the same amount of data from a MUS backup file in about the same amount of time. Missions operate each MUS SLE service instance independently by sending it MUS directives, which are legible, plain ASCII strings. MUS directives are usually (but not necessarily) sent through a TCP-IP (Transmission Control Protocol Internet Protocol) socket from a MOC (Mission Operations Center) or POCC (Payload Operations Control Center) system, under scripted control, during "lights-out" spacecraft operation. MUS permits the flight operations team to configure independently each of its data interfaces; not only commands and telemetry, but also MUS status messages to the MOC. Interfaces can use single- or multiple-client TCP/IP server sockets, TCP/IP client sockets, temporary disk files, the system log, or standard in, standard out, or standard error as applicable. By defining MUS templates in ASCII, the flight operations team can include any MUS system variable in telemetry or command headers or footers, and/or in status messages. Data fields can be arranged within messages in different sequences, according to the mission s needs. The only constraints imposed are on the format of MUS directive strings, and some bare minimum logical requirements that must be met in order for MUS to read the mission control center's spacecraft command inputs. The MUS system imposes no limits or constraints on the numbers and combinations of missions and SLE service instances that it will support simultaneously. At any time, flight operators may add, change, delete, bind, connect, or disconnect.

Perkins, Toby↗

SeaStar Unchained: Multiplying the Performance of the Cray SeaStar Network

The Oak Ridge Leadership Computing Facility (OLCF) supports many different systems and many differ- ent interconnects. The only common programming interfaces across these systems are BSD Sockets and MPI. Due to the design assumptions such as implicit buffering leading to extra copies, Sockets performance is almost universally lower than the native interface. Even in the cases that Sockets provides similar bandwidth as the native interface, it suffers from excessive CPU usage. MPI is the de-facto interface for intra- job communication, but is difficult to use between jobs and provides no ability to communicate with service nodes or off- system nodes (e.g. for I/O forwarding). We have developed the Common Communication Interface (CCI), a programming interface that exposes the advances in interconnect hardware, notably Remote Direct Memory Access (RDMA) and operating system (OS) bypass, while imposing as little overhead as possible. This API directly supports inter-job as well as off- system communication. CCI is a lightweight abstraction layer that provides point-to-point messaging and remote memory access. The Cray SeaStar ASIC, with its programmable embedded processor, provides an excellent platform to investigate the properties of various network protocols and programming interfaces. This paper describes our native implementation CCI on the SeaStar platform, and details how we implemented full OS bypass for common operations. We demonstrate a 30% to 50% reduction in latency, more than a six-fold increase in message injection rate, and an almost 7x improvement in bandwidth for small message sizes when compared to the generic Cray Portals implementation.

Dillow, David A↗

Data Movement Dominates: Final Report

Over the past three years in this project, what we have observed is that the primary reason for data movement in large-scale systems is that the per-node capacity is not large enough—i.e., one of the solutions to the data-movement problem (certainly not the only solution that is required, but a significant one nonetheless) is to increase per-node capacity so that inter-node traffic is reduced. This unfortunately is not as simple as it sounds. Today’s main memory systems for datacenters, enterprise computing systems, and supercomputers, fail to provide high per-socket capacity [Dirik & Jacob 2009; Cooper-Balis et al. 2012], except at extremely high price points (factors of 10–100x the cost/bit of consumer main-memory systems) [Stokes 2008]. The reason is that our choice of technology for today’s main memory systems—i.e., DRAM, which we have used as a main-memory technology since the 1970s [Jacob et al. 2007]—can no longer keep up with our needs for density and price per bit. Main memory systems have always been built from the cheapest, densest, lowest-power memory technology available, and DRAM is no longer the cheapest, the densest, nor the lowest-power storage technology out there. It is now time for DRAM to go the way that SRAM went: move out of the way for a cheaper, slower, denser storage technology, and become a cache instead. This inflection point has happened before, in the context of SRAM yielding to DRAM. There was once a time that SRAM was the storage technology of choice for all main memories [Tomasulo 1967; Thornton 1970; Kidder 1981]. However, once DRAM hit volume production in the 1970s and 80s, it supplanted SRAM as a main memory technology because it was cheaper, and it was denser. It also happened to be lower power, but that was not the primary consideration of the day. At the time, it was recognized that DRAM was much slower than SRAM, but it was only at the supercomputer level (For instance the Cray X-MP in the 1980s and its follow-on, the Cray Y-MP, in the 1990s) that could one afford to build ever- larger main memories out of SRAM—the reasoning for moving to DRAM was that an appropriately designed memory hierarchy, built of DRAM as main memory and SRAM as a cache, would approach the performance of SRAM, at the price-per-bit of DRAM [Mashey 1999]. Today it is quite clear that, were one to build an entire multi-gigabyte main memory out of SRAM instead of DRAM, one could improve the performance of almost any computer system by up to an order of magnitude—but this option is not even considered, because to build that system would be prohibitively expensive. It is now time to revisit the same design choice in the context of modern technologies and modern systems. For reasons both technical and economic, we can no longer afford to build ever-larger main memory systems out of DRAM. Flash memory, on the other hand, is significantly cheaper and denser than DRAM and therefore should take its place. While it is true that flash is significantly slower than DRAM, one can afford to build much larger main memories out of flash than out of DRAM, and we show that an appropriately designed memory hierarchy, built of flash as main memory and DRAM as a cache, will approach the performance of DRAM, at the price-per-bit of flash. In our studies as part of this project, we have investigated Non-Volatile Main Memory (NVMM), a new main-memory architecture for large-scale computing systems, one that is specifically designed to address the weaknesses described previously. In particular, it provides the following features: non-volatility: The bulk of the storage is comprised of NAND flash, and in this organization DRAM is used only as a cache, not as main memory. Furthermore, the flash is journaled, which means that operations such as checkpoint/restore are already built into the system. 1+ terabytes of storage per socket: SSDs and DRAM DIMMs have roughly the same form factor (several square inches of PCB surface area), and terabyte SSDs are now commonplace. performance approaching that of DRAM: DRAM is used as a cache to the flash system. price-per-bit approaching that of NAND: Flash is currently well under $0.50 per gigabyte; DDR3 SDRAM is currently just over $10 per gigabyte [Newegg 2014]. Even today, one can build an easily affordable main memory system with a terabyte or more of NAND storage per CPU socket (which would be extremely expensive were one to use DRAM), and our cycle- accurate, full-system experiments show that this can be done at a performance point that lies within a factor of two of DRAM.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

Electric vehicle system for charging and supplying electrical power

A power system that provides power between an energy storage device, an external charging-source/load, an onboard electrical power generator, and a vehicle drive shaft. The power system has at least one energy storage device electrically connected across a dc bus, at least one filter capacitor leg having at least one filter capacitor electrically connected across the dc bus, at least one power inverter/converter electrically connected across the dc bus, and at least one multiphase motor/generator having stator windings electrically connected at one end to form a neutral point and electrically connected on the other end to one of the power inverter/converters. A charging-sourcing selection socket is electrically connected to the neutral points and the external charging-source/load. At least one electronics controller is electrically connected to the charging-sourcing selection socket and at least one power inverter/converter. The switch legs in each of the inverter/converters selected by the charging-source/load socket collectively function as a single switch leg. The motor/generators function as an inductor.

24 POWER TRANSMISSION AND DISTRIBUTION↗

Optoelectronic device with a support member

Optoelectronic devices with a support member and methods of manufacturing or assembling the same are provided. An example of an optoelectronic device according to the present disclosure includes a substrate and an optical component and an electronic component disposed thereon or therein. The optoelectronic device further includes a ferrule coupled to the optical fiber and an optical socket receiving the ferrule therein. The optoelectronic device includes a support member disposed between the substrate and the optical socket such that the optical socket is spaced from the substrate by the support member.

42 ENGINEERING↗

Streaming data analytics via message passing with application to graph algorithms

The need to process streaming data, which arrives continuously at high-volume in real-time, arises in a variety of contexts including data produced by experiments, collections of environmental or network sensors, and running simulations. Streaming data can also be formulated as queries or transactions which operate on a large dynamic data store, e.g. a distributed database. We describe a lightweight, portable framework named PHISH which enables a set of independent processes to compute on a stream of data in a distributed-memory parallel manner. Datums are routed between processes in patterns defined by the application. PHISH can run on top of either message-passing via MPI or sockets via ZMQ. The former means streaming computations can be run on any parallel machine which supports MPI; the latter allows them to run on a heterogeneous, geographically dispersed network of machines. We illustrate how PHISH can support streaming MapReduce operations, and describe streaming versions of three algorithms for large, sparse graph analytics: triangle enumeration, subgraph isomorphism matching, and connected component finding. Lastly, we also provide benchmark timings for MPI versus socket performance of several kernel operations useful in streaming algorithms.

97 MATHEMATICS AND COMPUTING↗