Search NASA⌕ Search

SEARCH · Search NASA

Results for “Memory systems”

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 433 records · Page 24

Hierarchical Roofline analysis for GPUs: Accelerating performance optimization for the NERSC‐9 Perlmutter system

Summary The Roofline performance model provides an intuitive and insightful approach to identifying performance bottlenecks and guiding performance optimization. In preparation for the next‐generation supercomputer Perlmutter at NERSC, this paper presents a methodology to construct a hierarchical Roofline on NVIDIA GPUs and extends it to support reduced precision and Tensor Cores. The hierarchical Roofline incorporates L1, L2, device memory, and system memory bandwidths into one single figure, and it offers more profound insights into performance analysis than the traditional DRAM‐only Roofline. We use our Roofline methodology to analyze three proxy applications: GPP from BerkeleyGW, HPGMG from AMReX, and conv2d from TensorFlow. In doing so, we demonstrate the ability of our methodology to readily understand various aspects of performance and performance bottlenecks on NVIDIA GPUs and motivate code optimizations.

Yang, Charlene↗

Instructions for performing multi-line memory accesses

A system is described that performs memory access operations. The system includes a processor in a first node, a memory in a second node, a communication interconnect coupled to the processor and the memory, and an interconnect controller in the first node coupled between the processor and the communication interconnect. Upon executing a multi-line memory access instruction, the processor prepares a memory access operation for accessing, in the memory, a block of data including at least some of each of at least two lines of data. The processor then causes the interconnect controller to use a single remote direct memory access memory transfer to perform the memory access operation for the block of data via the communication interconnect.

Roberts, David A.↗

Analyzing the Impact of System Reliability Events on Applications in the Titan Supercomputer

Extreme-scale computing systems employ Reliability, Availability and Serviceability (RAS) mechanisms and infrastructure to log events from multiple system components. In this paper, we analyze RAS logs in conjunction with the application placement and scheduling database, in order to understand the impact of common RAS events on application performance. This study conducted on the records of about 2 million applications executed on Titan supercomputer provides important insights for system users, operators and computer science researchers. Specifically, we investigate the impact of RAS events on application performance and its variability by comparing cases where events are recorded with corresponding cases where no events are recorded. Such a statistical investigation is possible since we observed that system users tend to execute their applications multiple times. Our analysis reveals that most RAS events do impact application performance, although not always. We also find that different system components affect application performance differently. In particular, our investigation includes the following components: parallel file system processor, memory, graphics processing units, system and user software issues. Our work establishes the importance of providing feedback to system users for increasing operational efficiency of extreme-scale systems.

Ashraf, Rizwan↗

Comb

Comb is a communication performance benchmarking tool. It is used to determine performance tradeoffs in implementing communication patterns on high performance computing (HPC) platforms. At its core comb runs combinations of communication patterns with execution patterns, and memory spaces. The current set of capabilities Comb provides includes: o Configurable structured mesh halo exchange communication. o A variety of communication patterns based on grouping messages. o A variety of execution patterns including serial, openmp threading, cuda, cuda batched kernels, and cuda persistent kernels. o A variety of memory spaces including default system allocated memory, pinned host memory, cuda device memory, and cuda managed memory with different cuda memory advice.

Burmark, Jason↗

Performance evaluation of the Engineering Analysis and Data Systems (EADS) 2

The Engineering Analysis and Data System (EADS)II (1) was installed in March 1993 to provide high performance computing for science and engineering at Marshall Space Flight Center (MSFC). EADS II increased the computing capabilities over the existing EADS facility in the areas of throughput and mass storage. EADS II includes a Vector Processor Compute System (VPCS), a Virtual Memory Compute System (CFS), a Common Output System (COS), as well as Image Processing Station, Mini Super Computers, and Intelligent Workstations. These facilities are interconnected by a sophisticated network system. This work considers only the performance of the VPCS and the CFS. The VPCS is a Cray YMP. The CFS is implemented on an RS 6000 using the UniTree Mass Storage System. To better meet the science and engineering computing requirements, EADS II must be monitored, its performance analyzed, and appropriate modifications for performance improvement made. Implementing this approach requires tool(s) to assist in performance monitoring and analysis. In Spring 1994, PerfStat 2.0 was purchased to meet these needs for the VPCS and the CFS. PerfStat(2) is a set of tools that can be used to analyze both historical and real-time performance data. Its flexible design allows significant user customization. The user identifies what data is collected, how it is classified, and how it is displayed for evaluation. Both graphical and tabular displays are supported. The capability of the PerfStat tool was evaluated, appropriate modifications to EADS II to optimize throughput and enhance productivity were suggested and implemented, and the effects of these modifications on the systems performance were observed. In this paper, the PerfStat tool is described, then its use with EADS II is outlined briefly. Next, the evaluation of the VPCS, as well as the modifications made to the system are described. Finally, conclusions are drawn and recommendations for future worked are outlined.

Debrunner, Linda S.↗

Memory Errors in Modern Systems: The Good, The Bad, and The Ugly

Several recent publications have shown that hardware faults in the memory subsystem are commonplace. These faults are predicted to become more frequent in future systems that contain orders of magnitude more DRAM and SRAM than found in current memory subsystems. These memory subsystems will need to provide resilience techniques to tolerate these faults when deployed in high-performance computing systems and data centers containing tens of thousands of nodes. Therefore, it is critical to understand the efficacy of current hardware resilience techniques to determine whether they will be suitable for future systems. In this paper, we present a study of DRAM and SRAM faults and errors from the field. We use data from two leadership-class high-performance computer systems to analyze the reliability impact of hardware resilience schemes that are deployed in current systems. Our study has several key findings about the efficacy of many currently deployed reliability techniques such as DRAM ECC, DDR address/command parity, and SRAM ECC and parity. We also perform a methodological study, and find that counting errors instead of faults, a common practice among researchers and data center operators, can lead to incorrect conclusions about system reliability. Lastly, we use our data to project the needs of future large-scale systems. We find that SRAM faults are unlikely to pose a significantly larger reliability threat in the future, while DRAM faults will be a major concern and stronger DRAM resilience schemes will be needed to maintain acceptable failure rates similar to those found on today's systems.

97 MATHEMATICS AND COMPUTING↗

Memory object tagged memory monitoring method and system

Described are a method and processing apparatus to tag and track objects related to memory allocation calls. An application or software adds a tag to a memory allocation call to enable object level tracking. An entry is made into an object tracking table, which stores the tag and a variety of statistics related to the object and associated memory devices. The object statistics may be queried by the application to tune power/performance characteristics either by the application making runtime placement decisions, or by off-line code tuning based on a previous run. The application may add a tag to a memory allocation call to specify the type of memory characteristics requested based on the object statistics.

Roberts, David A.↗

Momentum transfer resolved memory in a magnetic system with perpendicular anisotropy

We have used resonant, coherent soft x-ray scattering to measure wave vector re- solved magnetic domain memory in Co/Pd multilayers. The technique uses angular cross correlation functions and can be applied to any system with circular annuli of constant values of scattering wave vector q. In our Co/Pd film, the memory exhibits a maximum at q = 0.0384 nm-1 near initial reversal that decreases in magnitude as the magnetization is further reversed. The peak is attributed to bubble domains that nucleate reproducibly near initial reversal and which grow into a labyrinth domain structure that is not reproduced from one magnetization cycle to the next.

75 CONDENSED MATTER PHYSICS, SUPERCONDUCTIVITY AND↗

A class Hierarchical, object-oriented approach to virtual memory management

The Choices family of operating systems exploits class hierarchies and object-oriented programming to facilitate the construction of customized operating systems for shared memory and networked multiprocessors. The software is being used in the Tapestry laboratory to study the performance of algorithms, mechanisms, and policies for parallel systems. Described here are the architectural design and class hierarchy of the Choices virtual memory management system. The software and hardware mechanisms and policies of a virtual memory system implement a memory hierarchy that exploits the trade-off between response times and storage capacities. In Choices, the notion of a memory hierarchy is captured by abstract classes. Concrete subclasses of those abstractions implement a virtual address space, segmentation, paging, physical memory management, secondary storage, and remote (that is, networked) storage. Captured in the notion of a memory hierarchy are classes that represent memory objects. These classes provide a storage mechanism that contains encapsulated data and have methods to read or write the memory object. Each of these classes provides specializations to represent the memory hierarchy.

Russo, Vincent F.↗

Optical Addressing And Clocking Of RAM's

Proposed random-access-memory (RAM) addressing system, in which memory linked optically to read/write logic circuits, greatly increases computer operating speed. System - comprises addressing circuits including numerous lasers as signal sources, numerous optical gates including optical detectors associated with memory cells, and holographic element to direct light signals to desired memory-cell locations - applied to high-capacity digital systems, supercomputers, and complex microcircuits.

Johnston, Alan R.↗

Memory access response merging in a memory hierarchy

A system and method for efficiently processing memory requests are described. A computing system includes multiple compute units, multiple caches of a memory hierarchy and a communication fabric. A compute unit generates a memory access request that misses in a higher level cache, which sends a miss request to a lower level shared cache. During servicing of the miss request, the lower level cache merges identification information of multiple memory access requests targeting a same cache line from multiple compute units into a merged memory access response. The lower level shared cache continues to insert information into the merged memory access response until the lower level shared cache is ready to issue the merged memory access response. An intermediate router in the communication fabric broadcasts the merged memory access response into multiple memory access responses to send to corresponding compute units.

97 MATHEMATICS AND COMPUTING↗

A study of the viability of exploiting memory content similarity to improve resilience to memory errors

Building the next-generation of extreme-scale distributed systems will require overcoming several challenges related to system resilience. As the number of processors in these systems grow, the failure rate increases proportionally. One of the most common sources of failure in large-scale systems is memory. In this paper, we propose a novel runtime for transparently exploiting memory content similarity to improve system resilience by reducing the rate at which memory errors lead to node failure. We evaluate the viability of this approach by examining memory snapshots collected from eight high-performance computing (HPC) applications and two important HPC operating systems. Based on the characteristics of the similarity uncovered, we conclude that our proposed approach shows promise for addressing system resilience in large-scale systems.

97 MATHEMATICS AND COMPUTING↗

On-board landmark navigation and attitude reference parallel processor system

An approach to autonomous navigation and attitude reference for earth observing spacecraft is described along with the landmark identification technique based on a sequential similarity detection algorithm (SSDA). Laboratory experiments undertaken to determine if better than one pixel accuracy in registration can be achieved consistent with onboard processor timing and capacity constraints are included. The SSDA is implemented using a multi-microprocessor system including synchronization logic and chip library. The data is processed in parallel stages, effectively reducing the time to match the small known image within a larger image as seen by the onboard image system. Shared memory is incorporated in the system to help communicate intermediate results among microprocessors. The functions include finding mean values and summation of absolute differences over the image search area. The hardware is a low power, compact unit suitable to onboard application with the flexibility to provide for different parameters depending upon the environment.

Gilbert, L. E.↗

Multi-port, optically addressed RAM

A random access memory addressing system utilizing optical links between memory and the read/write logic circuits comprises addressing circuits including a plurality of light signal sources, a plurality of optical gates including optical detectors associated with the memory cells, and a holographic optical element adapted to reflect and direct the light signals to the desired memory cell locations. More particularly, it is a multi-port, binary computer memory for interfacing with a plurality of computers. There are a plurality of storage cells for containing bits of binary information, the storage cells being disposed at the intersections of a plurality of row conductors and a plurality of column conductors. There is interfacing logic for receiving information from the computers directing access to ones of the storage cells. There are first light sources associated with the interfacing logic for transmitting a first light beam with the access information modulated thereon. First light detectors are associated with the storage cells for receiving the first light beam, for generating an electrical signal containing the access information, and for conducting the electrical signal to the one of the storage cells to which it is directed. There are holographic optical elements for reflecting the first light beam from the first light sources to the first light detectors.

Johnston, Alan R.↗

Mapping entry invalidation

A memory access system may include a first memory address translator, a second memory address translator and a mapping entry invalidator. The first memory address translator translates a first virtual address in a first protocol of a memory access request to a second virtual address in a second protocol and tracks memory access request completions. The second memory address translator is to translate the second virtual address to a physical address of a memory. The mapping entry invalidator requests invalidation of a first mapping entry of the first mapping address translator requests invalidation of a second mapping entry of the second memory address translator corresponding to the first mapping entry following invalidation of the first mapping entry and based upon the tracked memory access request completions.

Walker, Shawn K.↗

SharP Data Constructs: Data Constructs to Enable Data-Centric Computing

Extreme-scale applications (i.e., Big-Compute) are becoming increasingly data-intensive, i.e., producing and consuming increasingly large amounts of data. The HPC systems traditionally used for these applications are now used for Big-Data applications such as data analytics, social network analysis, machine learning, and genomics. As a consequence of these trends, the system architecture should be flexible and data-centric. This can already be witnessed in the pre-exascale systems with TBs of on-node hierarchical and heterogeneous memories, PBs of system memory, low-latency, high-throughput networks, and many threaded cores. As such, the pre-exascale systems suit the needs of both Big-Compute and Big-Data applications. Though the system architecture is flexible enough to support both Big-Compute and Big-Data, we argue there is a software gap. Particularly, we need data-centric abstractions to leverage the full potential of the system, i.e., there is a need for native support for data resilience, the ability to express data locality and affinity, mechanisms to reduce data movement, the ability to share data, and abstractions to express User's data usage and data access patterns. In this paper, we (i) show the need for taking a holistic approach towards data-centric abstractions, (ii) show how these approaches were realized in the SHARed data-structure centric Programming abstraction (SharP) library, a data-structure centric programming abstraction, and (iii) apply these approaches to a variety of applications that demonstrate its usefulness. Particularly, we apply these approaches to QMCPack and the Graph500 benchmark and demonstrate the advantages of this approach on extreme-scale systems.

Aderholdt, William Ferrol↗