Engineering Papers⌕ Search

SEARCH · Engineering Papers

Results for “Asynchronous data management”

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 73 records · Page 4

Proactive Data Containers for Scientific Storage (Final Report)

Emerging HPC systems are expected to be deployed with an unprecedented level of complexity, due to a deep system memory/storage hierarchy and heterogeneity of the storage hardware. This hierarchy is expected to range from CPU cache through several levels of volatile memory to nonvolatile memory, traditional hard disks, and tape. Simple and efficient methods of data management and movement through this hierarchy is critical for scientific applications using exascale systems. Existing storage system and I/O (SSIO) technologies face severe challenges in dealing with these requirements. POSIX and MPI I/O standards that are the basis for existing I/O libraries and parallel file systems present fundamental challenges in the areas of scalable metadata operations, semantics-based data movement performance tuning, asynchronous operation, and support for scalable consistency of distributed operations. Moving toward new paradigms for SSIO in the extreme-scale era, we have proposed to investigate novel object-based data abstractions and storage mechanisms that take advantage of the deep storage hierarchy and enable proactive automated performance tuning. In order to achieve these overarching goals, we initiated an effort to develop a fundamental new data abstraction, called Proactive Data Containers (PDC). A PDC is a container within a locus of storage (memory, NVRAM, disk, etc.) that stores science data in an object-centric manner. Managing data as objects enables powerful optimization opportunities for data movement and transformations. The R&D focus of this project are: 1) formulation of object-oriented PDCs and their mapping in different levels of the exascale storage hierarchy; 2) efficient strategies for moving data in deep storage hierarchies using PDCs; 3) techniques for transforming and reorganizing data based on application requirements; and 4) novel analysis paradigms for enabling data transformations and user-defined analysis on data in PDCs. Toward achieving these overarching goals, we designed an object-centric application programing interface (API) for HPC, scalable metadata management for object-centric storage systems, and data movement optimizations such as Data Elevator for moving data between two levels of storage devices and TAPIOCA for efficient aggregation of data on compute nodes. We then implemented several components of the PDC system. They include metadata management, data placement services, remote procedure calls, data aggregation, etc. We have put them together into the overall PDC framework.

97 MATHEMATICS AND COMPUTING↗

Event Driven Messaging with Role-Based Subscriptions

Event Driven Messaging with Role-Based Subscriptions (EDM-RBS) is a framework integrated into the Service Management Database (SMDB) to allow for role-based and subscription-based delivery of synchronous and asynchronous messages over JMS (Java Messaging Service), SMTP (Simple Mail Transfer Protocol), or SMS (Short Messaging Service). This allows for 24/7 operation with users in all parts of the world. The software classifies messages by triggering data type, application source, owner of data triggering event (mission), classification, sub-classification and various other secondary classifying tags. Messages are routed to applications or users based on subscription rules using a combination of the above message attributes. This program provides a framework for identifying connected users and their applications for targeted delivery of messages over JMS to the client applications the user is logged into. EDMRBS provides the ability to send notifications over e-mail or pager rather than having to rely on a live human to do it. It is implemented as an Oracle application that uses Oracle relational database management system intrinsic functions. It is configurable to use Oracle AQ JMS API or an external JMS provider for messaging. It fully integrates into the event-logging framework of SMDB (Subnet Management Database).

Bui, Tung↗

AsyncStageOut: Distributed User Data Management for CMS Analysis

AsyncStageOut (ASO) is a new component of the distributed data analysis system of CMS, CRAB, designed for managing users' data. It addresses a major weakness of the previous model, namely that mass storage of output data was part of the job execution resulting in inefficient use of job slots and an unacceptable failure rate at the end of the jobs. ASO foresees the management of up to 400k files per day of various sizes, spread worldwide across more than 60 sites. It must handle up to 1000 individual users per month, and work with minimal delay. This creates challenging requirements for system scalability, performance and monitoring. ASO uses FTS to schedule and execute the transfers between the storage elements of the source and destination sites. It has evolved from a limited prototype to a highly adaptable service, which manages and monitors the user file placement and bookkeeping. To ensure system scalability and data monitoring, it employs new technologies such as a NoSQL database and re-uses existing components of PhEDEx and the FTS Dashboard. We present the asynchronous stage-out strategy and the architecture of the solution we implemented to deal with those issues and challenges. The deployment model for the high availability and scalability of the service is discussed. The performance of the system during the commissioning and the first phase of production are also shown, along with results from simulations designed to explore the limits of scalability.

Riahi, H.↗

Design Issues for Traffic Management for the ATM UBR + Service for TCP Over Satellite Networks

This project was a comprehensive research program for developing techniques for improving the performance of Internet protocols over Asynchronous Transfer Mode (ATM) based satellite networks. Among the service categories provided by ATM networks, the most commonly used category for data traffic is the unspecified bit rate (UBR) service. UBR allows sources to send data into the network without any feedback control. The project resulted in the numerous ATM Forum contributions and papers.

Jain, Raj↗

Method and system for redundancy management of distributed and recoverable digital control system

A method and system for redundancy management is provided for a distributed and recoverable digital control system. The method uses unique redundancy management techniques to achieve recovery and restoration of redundant elements to full operation in an asynchronous environment. The system includes a first computing unit comprising a pair of redundant computational lanes for generating redundant control commands. One or more internal monitors detect data errors in the control commands, and provide a recovery trigger to the first computing unit. A second redundant computing unit provides the same features as the first computing unit. A first actuator control unit is configured to provide blending and monitoring of the control commands from the first and second computing units, and to provide a recovery trigger to each of the first and second computing units. A second actuator control unit provides the same features as the first actuator control unit.

Stange, Kent↗

Adaptive parallel logic networks

Adaptive, self-organizing concurrent systems (ASOCS) that combine self-organization with massive parallelism for such applications as adaptive logic devices, robotics, process control, and system malfunction management, are presently discussed. In ASOCS, an adaptive network composed of many simple computing elements operating in combinational and asynchronous fashion is used and problems are specified by presenting if-then rules to the system in the form of Boolean conjunctions. During data processing, which is a different operational phase from adaptation, the network acts as a parallel hardware circuit.

Martinez, Tony R.↗

DYnamic and Asynchronous Data Streamliner

DYAD aims to help sharing data files between producer and consumer job elements, especially between co-scheduled jobs or within an ensemble. DYAD provides the service by two components: a FLUX module and a I/O wraper set. DYAD transparently synchronizes file I/O between producer and consumer, and transfers data from the producer location to the consumer location managed by the service. Users only need to use the file path that is under the directory managed by the service.

Ahn, DongH↗

Service Management Database for DSN Equipment

This data- and event-driven persistent storage system leverages the use of commercial software provided by Oracle for portability, ease of maintenance, scalability, and ease of integration with embedded, client-server, and multi-tiered applications. In this role, the Service Management Database (SMDB) is a key component of the overall end-to-end process involved in the scheduling, preparation, and configuration of the Deep Space Network (DSN) equipment needed to perform the various telecommunication services the DSN provides to its customers worldwide. SMDB makes efficient use of triggers, stored procedures, queuing functions, e-mail capabilities, data management, and Java integration features provided by the Oracle relational database management system. SMDB uses a third normal form schema design that allows for simple data maintenance procedures and thin layers of integration with client applications. The software provides an integrated event logging system with ability to publish events to a JMS messaging system for synchronous and asynchronous delivery to subscribed applications. It provides a structured classification of events and application-level messages stored in database tables that are accessible by monitoring applications for real-time monitoring or for troubleshooting and analysis over historical archives.

Zendejas, Silvino↗

On the Suitability of MPI as a PGAS Runtime

Partitioned Global Address Space (PGAS) models are emerging as a popular alternative to MPI models for designing scalable applications. At the same time, MPI remains a ubiquitous communication subsystem due to its standardization, high performance, and availability on leading platforms. In this paper, we explore the suitability of using MPI as a scalable PGAS communication subsystem. We focus on the Remote Memory Access (RMA) communication in PGAS models which typically includes {\em get, put,} and {\em atomic memory operations}. We perform an in-depth exploration of design alternatives based on MPI. These alternatives include using a semantically-matching interface such as MPI-RMA, as well as not-so-intuitive interfaces such as MPI two-sided with a combination of multi-threading and dynamic process management. With an in-depth exploration of these alternatives and their shortcomings, we propose a novel design which is facilitated by the data-centric view in PGAS models. This design leverages a combination of highly tuned MPI two-sided semantics and an automatic, user-transparent split of MPI communicators to provide asynchronous progress. We implement the asynchronous progress ranks approach and other approaches within the Communication Runtime for Exascale which is a communication subsystem for Global Arrays. Our performance evaluation spans pure communication benchmarks, graph community detection and sparse matrix-vector multiplication kernels, and a computational chemistry application. The utility of our proposed PR-based approach is demonstrated by a 2.17x speed-up on 1008 processors over the other MPI-based designs.

Computational modeling↗

Dispatch Manager for NEML2 Constitutive Model Calculations Embedded in MOOSE

This report describes the extended capabilities of the NEML2 constitutive modeling library, including a flexible and efficient work dispatching system designed to leverage both CPU and GPU resources. This enhancement addresses one of the primary computational challenges in large-scale simulations: the ability to distribute and execute batches of material model evaluations across heterogeneous computing devices. The new dispatch system introduces a modular set of dispatcher and scheduler classes that coordinate the flow of data and execution between devices. The dispatcher is responsible for efficiently packaging work, managing device-specific memory operations, and synchronizing results. This modularity allows for extensibility, making it straightforward to integrate additional computing backends in the future. From an implementation standpoint, the dispatcher system interfaces seamlessly with NEML2's existing models. They handle device-aware tensor operations, optimize memory transfers, and support asynchronous execution when applicable. This design ensures that batches of material points can be evaluated concurrently, substantially improving throughput compared to previous single-device or serial implementations. These improvements not only enhance the raw performance of NEML2 but also improve its usability in multiscale and high-fidelity simulations, where the simultaneous evaluation of large material point batches is critical. Benchmarks included in the report demonstrate the system’s scalability, highlighting its effectiveness when leveraging modern GPU architectures.

22 GENERAL STUDIES OF NUCLEAR REACTORS↗

UPC++ v1.0 Programmer’s Guide, Revision 2021.9.0

UPC++ is a C++ library that provides Partitioned Global Address Space (PGAS) programming. It is designed for writing parallel programs that run efficiently and scale well on distributed-memory parallel computers. The PGAS model is single program, multiple-data (SPMD), with each separate constituent process having access to local memory as it would in C++. PGAS additionally provides one-sided Remote Memory Access (RMA) to a global address space, which is allocated in shared segments that are distributed over the processes. UPC++ also features Remote Procedure Call (RPC) communication, making it easy to move computation to operate on data that resides on remote processes. In UPC++, all communication operations are explicit, which encourages programmers to be aware of the cost of communication and data movement. Moreover, all communication operations are asynchronous by default, to enable programmers to write code that scales well even on hundreds of thousands of cores.

96 KNOWLEDGE MANAGEMENT AND PRESERVATION↗

High-Speed Ring Bus

The high-speed ring bus at the Jet Propulsion Laboratory (JPL) allows for future growth trends in spacecraft seen with future scientific missions. This innovation constitutes an enhancement of the 1393 bus as documented in the Institute of Electrical and Electronics Engineers (IEEE) 1393-1999 standard for a spaceborne fiber-optic data bus. It allows for high-bandwidth and time synchronization of all nodes on the ring. The JPL ring bus allows for interconnection of active units with autonomous operation and increased fault handling at high bandwidths. It minimizes the flight software interface with an intelligent physical layer design that has few states to manage as well as simplified testability. The design will soon be documented in the AS-1393 standard (Serial Hi-Rel Ring Network for Aerospace Applications). The framework is designed for "Class A" spacecraft operation and provides redundant data paths. It is based on "fault containment regions" and "redundant functional regions (RFR)" and has a method for allocating cables that completely supports the redundancy in spacecraft design, allowing for a complete RFR to fail. This design reduces the mass of the bus by incorporating both the Control Unit and the Data Unit in the same hardware. The standard uses ATM (asynchronous transfer mode) packets, standardized by ITU-T, ANSI, ETSI, and the ATM Forum. The IEEE-1393 standard uses the UNI form of the packet and provides no protection for the data portion of the cell. The JPL design adds optional formatting to this data portion. This design extends fault protection beyond that of the interconnect. This includes adding protection to the data portion that is contained within the Bus Interface Units (BIUs) and by adding to the signal interface between the Data Host and the JPL 1393 Ring Bus. Data transfer on the ring bus does not involve a master or initiator. Following bus protocol, any BIU may transmit data on the ring whenever it has data received from its host. There is no centralized arbitration or bus granting. The JPL design provides for autonomous synchronization of the nodes on the ring bus. An address-synchronous latency adjust buffer (LAB) has been designed that cannot get out of synchronization and needs no external input. Also, a priority-driven cable selection behavior has been programmed into each unit on the ring bus. This makes the bus able to connect itself up, according to a maximum redundancy priority system, without the need for computer intervention at startup. Switching around a failed or switched-off unit is also autonomous. The JPL bus provides a map of all the active units for the host computer to read and use for fault management. With regard to timing, this enhanced bus recognizes coordinated timing on a spacecraft as critical and addresses this with a single source of absolute and relative time, which is broadcast to all units on the bus with synchronization maintained to the tens of nanoseconds. Each BIU consists of up to five programmable triggers, which may be programmed for synchronization of events within the spacecraft of instrument. All JPL-formatted data transmitted on the ring bus are automatically time-stamped.

Wysocky, Terry↗

Using the Sirocco File System for high-bandwidth checkpoints.

The Sirocco File System, a file system for exascale under active development, is designed to allow the storage software to maximize quality of service through increased flexibility and local decision-making. By allowing the storage system to manage a range of storage targets that have varying speeds and capacities, the system can increase the speed and surety of storage to the application. We instrument CTH to use a group of RAM-based Sirocco storage servers allocated within the job as a high-performance storage tier to accept checkpoints, allowing computation to potentially continue asynchronously of checkpoint migration to slower, more permanent storage. The result is a 10-60x speedup in constructing and moving checkpoint data from the compute nodes. This demonstration of early Sirocco functionality shows a significant benefit for a real I/O workload, checkpointing, in a real application, CTH. By running Sirocco storage servers within a job as RAM-only stores, CTH was able to store checkpoints 10-60x faster than storing to PanFS, allowing the job to continue computing sooner. While this prototype did not include automatic data migration, the checkpoint was available to be pushed or pulled to disk-based storage as needed after the compute nodes continued computing. Future developments include the ability to dynamically spawn Sirocco nodes to absorb checkpoints, expanding this mechanism to other fast tiers of storage like flash memory, and sharing of dynamic Sirocco nodes between multiple jobs as needed.

99 GENERAL AND MISCELLANEOUS↗

The uintah framework: a unified heterogeneous task scheduling and runtime system

The development of a new unified, multi-threaded runtime system for the execution of asynchronous tasks on heterogeneous systems is described in this work. These asynchronous tasks arise from the Uintah framework, which was developed to provide an environment for solving a broad class of fluid-structure interaction problems on structured adaptive grids. Uintah has a clear separation between its MPI-free user-coded tasks and its runtime system that ensures these tasks execute efficiently. This separation also allows for complete isolation of the application developer from the complexities involved with the parallelism Uintah provides. While we have designed scalable runtime systems for large CPU core counts, the emergence of heterogeneous systems, with additional on-node accelerators and co-processors presents additional design challenges in terms of effectively utilizing all computational resources on-node and managing multiple levels of parallelism. Our work addresses these challenges for Uintah by the development of new hybrid runtime system and Unified multi-threaded MPI task scheduler, enabling Uintah to fully exploit current and emerging architectures with support for asynchronous, out-of-order scheduling of both CPU and GPU computational tasks. This design coupled with an approach that uses MPI to communicate between nodes, a shared memory model on-node and the use of novel lock-free data structures, has made it possible for Uintah to achieve excellent scalability for challenging fluid-structure problems using adaptive mesh refinement on as many as 256K cores on the DoE Jaguar XK6 system. This design has also demonstrated an ability to run capability jobs on the heterogeneous systems, Keeneland and TitanDev. In this work, the evolution of Uintah and its runtime system is examined in the context of our new Unified multi-threaded scheduler design. The performance of the Unified scheduler is also tested against previous Uintah scheduler and runtime designs over a range of processor core and GPU counts.

Source record↗

Asynchrony among local communities stabilises ecosystem function of metacommunities

Temporal stability of ecosystem functioning increases the predictability and reliability of ecosystem services, and understanding the drivers of stability across spatial scales is important for land management and policy decisions. We used species-level abundance data from 62 plant communities across five continents to assess mechanisms of temporal stability across spatial scales. We assessed how asynchrony (i.e. different units responding dissimilarly through time) of species and local communities stabilised metacommunity ecosystem function. Asynchrony of species increased stability of local communities, and asynchrony among local communities enhanced metacommunity stability by a wide range of magnitudes (1–315%); this range was positively correlated with the size of the metacommunity. Additionally, asynchronous responses among local communities were linked with species’ populations fluctuating asynchronously across space, perhaps stemming from physical and/or competitive differences among local communities. Accordingly, we suggest spatial heterogeneity should be a major focus for maintaining the stability of ecosystem services at larger spatial scales.

59 BASIC BIOLOGICAL SCIENCES↗

Collaborative Resource Allocation

Collaborative Resource Allocation Networking Environment (CRANE) Version 0.5 is a prototype created to prove the newest concept of using a distributed environment to schedule Deep Space Network (DSN) antenna times in a collaborative fashion. This program is for all space-flight and terrestrial science project users and DSN schedulers to perform scheduling activities and conflict resolution, both synchronously and asynchronously. Project schedulers can, for the first time, participate directly in scheduling their tracking times into the official DSN schedule, and negotiate directly with other projects in an integrated scheduling system. A master schedule covers long-range, mid-range, near-real-time, and real-time scheduling time frames all in one, rather than the current method of separate functions that are supported by different processes and tools. CRANE also provides private workspaces (both dynamic and static), data sharing, scenario management, user control, rapid messaging (based on Java Message Service), data/time synchronization, workflow management, notification (including emails), conflict checking, and a linkage to a schedule generation engine. The data structure with corresponding database design combines object trees with multiple associated mortal instances and relational database to provide unprecedented traceability and simplify the existing DSN XML schedule representation. These technologies are used to provide traceability, schedule negotiation, conflict resolution, and load forecasting from real-time operations to long-range loading analysis up to 20 years in the future. CRANE includes a database, a stored procedure layer, an agent-based middle tier, a Web service wrapper, a Windows Integrated Analysis Environment (IAE), a Java application, and a Web page interface.

Wang, Yeou-Fang↗

Django Remote Submission

The Django Remote Submission (DRS) is a Django (Django, n.d.) application to manage long running job submission, including starting the job, saving logs, and storing results. It is an independent project available as a standalone pypi package (PyPi, n.d.). It can be easily integrated in any Django project. The source code is freely available as a GitHub repository (django-remote-submission, n.d.). To run the jobs in background, DRS takes advantage of Celery (Celery, n.d.), a powerful asynchronous job queue used for running tasks in the background, and the Redis Server (Redis, n.d.), an in-memory data structure store. Celery uses brokers to pass messages between a Django Project and the Celery workers. Redis is the message broker of DRS. In addition DRS provides real time monitoring of the progress of Jobs and associated logs. Through the Django Channels project (Channels, n.d.), and the usage of Web Sockets, it is possible to asynchronously display the Job Status and the live Job output (standard output and standard error) on a web page.

97 MATHEMATICS AND COMPUTING↗

UMap: An application-oriented user level memory mapping library

Exploiting the prominent role of complex memories in exascale node architecture, the UMap page fault handler offers new capabilities to access large memory-mapped data sets directly. UMap provides flexible configuration options to customize page handling to each application, including analysis of massive observational and simulation data sets. The high-performance design features I/O decoupling, dynamic load balancing, and application-level controls. Page faults triggered by application threads and processes accessing data mapped to a UMapp’ed region are handled via the Linux userfaultfd protocol, an asynchronous message-oriented kernel-user communication mechanism that avoids the context switch penalty of traditional signal fault handlers. UMap is fully open source. In this paper, we give an overview of the UMap library architecture, its extensible plugin architecture, and the use/performance of UMap in emerging heterogeneous memory hierarchies such as near-node Non-volatile Memory (NVM) and network attached memories. We highlight new capabilities in two pagefault management plugins, the NetworkStore and SparseStore. We demonstrate the integration between UMap and multiple ECP products including Caliper, Metall, ZFP, Mochi, and Ripples.

97 MATHEMATICS AND COMPUTING↗