Search NASA⌕ Search

SEARCH · Search NASA

Results for “MULTIPROCESSORS”

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.

626 records · Page 35

Hardware Architectures for Data-Intensive Computing Problems: A Case Study for String Matching

DNA analysis is an emerging application of high performance bioinformatic. Modern sequencing machinery are able to provide, in few hours, large input streams of data, which needs to be matched against exponentially growing databases of known fragments. The ability to recognize these patterns effectively and fastly may allow extending the scale and the reach of the investigations performed by biology scientists. Aho-Corasick is an exact, multiple pattern matching algorithm often at the base of this application. High performance systems are a promising platform to accelerate this algorithm, which is computationally intensive but also inherently parallel. Nowadays, high performance systems also include heterogeneous processing elements, such as Graphic Processing Units (GPUs), to further accelerate parallel algorithms. Unfortunately, the Aho-Corasick algorithm exhibits large performance variability, depending on the size of the input streams, on the number of patterns to search and on the number of matches, and poses significant challenges on current high performance software and hardware implementations. An adequate mapping of the algorithm on the target architecture, coping with the limit of the underlining hardware, is required to reach the desired high throughputs. In this paper, we discuss the implementation of the Aho-Corasick algorithm for GPU-accelerated high performance systems. We present an optimized implementation of Aho-Corasick for GPUs and discuss its tradeoffs on the Tesla T10 and he new Tesla T20 (codename Fermi) GPUs. We then integrate the optimized GPU code, respectively, in a MPI-based and in a pthreads-based load balancer to enable execution of the algorithm on clusters and large sharedmemory multiprocessors (SMPs) accelerated with multiple GPUs.

Tumeo, Antonino↗

Cache directory look-up re-use as conflict check mechanism for speculative memory requests

In a cache memory, energy and other efficiencies can be realized by saving a result of a cache directory lookup for sequential accesses to a same memory address. Where the cache is a point of coherence for speculative execution in a multiprocessor system, with directory lookups serving as the point of conflict detection, such saving becomes particularly advantageous.

97 MATHEMATICS AND COMPUTING↗

Method and apparatus for efficiently tracking queue entries relative to a timestamp

An apparatus and method for tracking coherence event signals transmitted in a multiprocessor system. The apparatus comprises a coherence logic unit, each unit having a plurality of queue structures with each queue structure associated with a respective sender of event signals transmitted in the system. A timing circuit associated with a queue structure controls enqueuing and dequeuing of received coherence event signals, and, a counter tracks a number of coherence event signals remaining enqueued in the queue structure and dequeued since receipt of a timestamp signal. A counter mechanism generates an output signal indicating that all of the coherence event signals present in the queue structure at the time of receipt of the timestamp signal have been dequeued. In one embodiment, the timestamp signal is asserted at the start of a memory synchronization operation and, the output signal indicates that all coherence events present when the timestamp signal was asserted have completed. This signal can then be used as part of the completion condition for the memory synchronization operation.

97 MATHEMATICS AND COMPUTING↗

Mapping virtual addresses to different physical addresses for value disambiguation for thread memory access requests

A multiprocessor system includes nodes. Each node includes a data path that includes a core, a TLB, and a first level cache implementing disambiguation. The system also includes at least one second level cache and a main memory. For thread memory access requests, the core uses an address associated with an instruction format of the core. The first level cache uses an address format related to the size of the main memory plus an offset corresponding to hardware thread meta data. The second level cache uses a physical main memory address plus software thread meta data to store the memory access request. The second level cache accesses the main memory using the physical address with neither the offset nor the thread meta data after resolving speculation. In short, this system includes mapping of a virtual address to a different physical addresses for value disambiguation for different threads.

97 MATHEMATICS AND COMPUTING↗

Cache directory lookup reader set encoding for partial cache line speculation support

In a multiprocessor system, with conflict checking implemented in a directory lookup of a shared cache memory, a reader set encoding permits dynamic recordation of read accesses. The reader set encoding includes an indication of a portion of a line read, for instance by indicating boundaries of read accesses. Different encodings may apply to different types of speculative execution.

97 MATHEMATICS AND COMPUTING↗

A Framework for Batched and GPU-Resident Factorization Algorithms Applied to Block Householder Transformations

As modern hardware keeps evolving, an increasingly effective approach to developing energy efficient and high-performance solvers is to design them to work on many small size and independent problems. Many applications already need this functionality, especially for GPUs, which are currently known to be about four to five times more energy efficient than multicore CPUs. We describe the development of one-sided factorizations that work for a set of small dense matrices in parallel, and we illustrate our techniques on the QR factorization based on Householder transformations. We refer to this mode of operation as a batched factorization. Our approach is based on representing the algorithms as a sequence of batched BLAS routines for GPU-only execution. This is in contrast to the hybrid CPU-GPU algorithms that rely heavily on using the multicore CPU for specific parts of the workload. But for a system to benefit fully from the GPU's significantly higher energy efficiency, avoiding the use of the multicore CPU must be a primary design goal, so the system can rely more heavily on the more efficient GPU. Additionally, this will result in the removal of the costly CPU-to-GPU communication. Furthermore, we do not use a single symmetric multiprocessor(on the GPU) to factorize a single problem at a time. We illustrate how our performance analysis, and the use of profiling and tracing tools, guided the development and optimization of our batched factorization to achieve up to a 2-fold speedup and a 3-fold energy efficiency improvement compared to our highly optimized batched CPU implementations based on the MKL library(when using two sockets of Intel Sandy Bridge CPUs). Compared to a batched QR factorization featured in the CUBLAS library for GPUs, we achieved up to 5x speedup on the K40 GPU.

Dong, Tingzing Tim↗

A Latency Tolerant Runtime System for Mass Market Computer Systems

Grappa is a latency tolerant runtime system for mass market computer systems comprised of multiple nodes having x86 multicore processors and infiniband interconnect. Grappa is designed to enable in-memory irregular applications written in the C++ programming language to scale up on multiprocessor systems. Grappa addresses the load balance, communication, and synchronization challenges posed by running applications that do not partition evenly across large systems. Two mechanisms are essential to its success: fast task switching and message aggregation. Fast task switching enables processors to tolerate communication and synchronization delays; aggregation sustains high bandwidth on low injection rate networks.

Source record↗

Single-pass parallel prefix scan with dynamic look back

One embodiment of the present invention performs a parallel prefix scan in a single pass that incorporates variable look-back. A parallel processing unit (PPU) subdivides a list of inputs into sequentially-ordered segments and assigns each segment to a streaming multiprocessor (SM) included in the PPU. Notably, the SMs may operate in parallel. Each SM executes write operations on a segment descriptor that includes the status, aggregate, and inclusive-prefix associated with the assigned segment. Further, each SM may execute read operations on segment descriptors associated with other segments. In operation, each SM may perform reduction operations to determine a segment-wide aggregate, may perform look-back operations across multiple preceding segments to determine an exclusive-prefix, and may perform a scan seeded with the exclusive prefix to generate output data. Advantageously, the PPU performs one read operation per input, thereby reducing the time required to execute the prefix scan relative to prior-art parallel implementations.

Merrill, Duane↗

Computing rank‐revealing factorizations of matrices stored out‐of‐core

This paper describes efficient algorithms for computing rank-revealing factorizations of matrices that are too large to fit in main memory (RAM), and must instead be stored on slow external memory devices such as disks (out-of-core or out-of-memory). Traditional algorithms for computing rank-revealing factorizations (such as the column pivoted QR factorization and the singular value decomposition) are very communication intensive as they require many vector-vector and matrix-vector operations, which become prohibitively expensive when data is not in RAM. Randomization allows to reformulate new methods so that large contiguous blocks of the matrix are processed in bulk. The paper describes two distinct methods. The first is a blocked version of column pivoted Householder QR, organized as a “left-looking” method to minimize the number of the expensive write operations. The second method results employs a UTV factorization. It is organized as an algorithm-by-blocks to overlap computations and I/O operations. As it incorporates power iterations, it is much better at revealing the numerical rank. Numerical experiments on several computers demonstrate that the new algorithms are almost as fast when processing data stored on slow memory devices as traditional algorithms are for data stored in RAM.

97 MATHEMATICS AND COMPUTING↗

Performance Models for Split-execution Computing Systems

Split-execution computing leverages the capabilities of multiple computational models to solve problems, but splitting program execution across different computational models incurs costs associated with the translation between domains. We analyze the performance of a split-execution computing system developed from conventional and quantum processing units (QPUs) by using behavioral models that track resource usage. We focus on asymmetric processing models built using conventional CPUs and a family of special-purpose QPUs that employ quantum computing principles. Our performance models account for the translation of a classical optimization problem into the physical representation required by the quantum processor while also accounting for hardware limitations and conventional processor speed and memory. We conclude that the bottleneck in this split-execution computing system lies at the quantum-classical interface and that the primary time cost is independent of quantum processor behavior.

multiprocessor↗

Foundations for the Future Power System: Inverter-Based Resource Interconnection Standards

As power systems evolve to become reliant on solar, wind, batteries, and other inverter-based resources (IBRs), it is essential that those resources meet certain minimum performance and capability criteria designed to ensure the power system operates stably and reliably. Because those criteria, as enshrined in interconnection standards, take years to develop and are very long lived, they need to account for not only the present state of the power system but also its expected future state over the lifetime of the power plants to which they will apply. In addition, they need to be specific enough to ensure reliability without over specifying and, thereby, impeding innovation. Because power systems are shifting rapidly from a state where IBRs make up a small to medium portion of generation to one where the generation will, at times, come predominantly from IBRs, interconnection standards are especially challenging to develop today. Good interconnection standards can make the power system more reliable and less costly to operate, whereas poorly designed standards can lead to major problems, like the famous German 50.2-Hz problem, which introduced a risk of losing many gigawatts of solar on a frequency excursion and resulted in many IBRs being retrofitted at great cost to mitigate a major system reliability risk. Readying the power system by specifying forward-looking technical minimum functional capabilities for IBRs can be an effective approach to avoid future retrofits.

data models↗

Point-to-Point Multicast Communications Protocol

This paper describes a protocol to support point-to-point interprocessor communications with multicast. Dynamic, cut-through routing with local flow control is used to provide a high-throughput, low-latency communications path between processors. In addition multicast transmissions are available, in which copies of a packet are sent to multiple destinations using common resources as much as possible. Special packet terminators and selective buffering are introduced to avoid a deadlock during multicasts. A simulated implementation of the protocol is also described.

PROTOCOLS↗

Quantifying Scheduling Challenges for Exascale System Software

The move towards high-performance computing (HPC) ap- plications comprised of coupled codes and the need to dra- matically reduce data movement is leading to a reexami- nation of time-sharing vs. space-sharing in HPC systems. In this paper, we discuss and begin to quantify the perfor- mance impact of a move away from strict space-sharing of nodes for HPC applications. Specifically, we examine the po- tential performance cost of time-sharing nodes between ap- plication components, we determine whether a simple coor- dinated scheduling mechanism can address these problems, and we research how suitable simple constraint-based opti- mization techniques are for solving scheduling challenges in this regime. Our results demonstrate that current general- purpose HPC system software scheduling and resource al- location systems are subject to significant performance de- ciencies which we quantify for six representative applica- tions. Based on these results, we discuss areas in which ad- ditional research is needed to meet the scheduling challenges of next-generation HPC systems.

C.1.2 [Multiprocessors]: Parallel Processors↗