Search NASA⌕ Search

SEARCH · Search NASA

Results for “INFINIBAND”

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 127 records · Page 7

Cheetah: A Framework for Scalable Hierarchical Collective Operations

Collective communication operations, used by many scientific applications, tend to limit overall parallel application performance and scalability. Computer systems are becoming more heterogeneous with increasing node and core-per-node counts. Also, a growing number of data-access mechanisms, of varying characteristics, are supported within a single computer system. We describe a new hierarchical collective communication framework that takes advantage of hardware-specific data-access mechanisms. It is flexible, with run-time hierarchy specification, and sharing of collective communication primitives between collective algorithms. Data buffers are shared between levels in the hierarchy reducing collective communication management overhead. We have implemented several versions of the Message Passing Interface (MPI) collective operations, MPI Barrier() and MPI Bcast(), and run experiments using up to 49, 152 processes on a Cray XT5, and a small InfiniBand based cluster. At 49, 152 processes our barrier implementation outperforms the optimized native implementation by 75%. 32 Byte and one Mega-Byte broadcasts outperform it by 62% and 11%, respectively, with better scalability characteristics. Improvements relative to the default Open MPI implementation are much larger.

99 GENERAL AND MISCELLANEOUS↗

GlusterFS One Storage Server to Rule Them All

GlusterFS is a Linux based distributed file system, designed to be highly scalable and serve many clients. Some reasons to use GlusterFS are: No centralized metadata server, Scalability, Open Source, Dynamic and live service modifications, Can be used over Infiniband or Ethernet, Can be tuned for speed and/or resilience and Flexible administration. It's useful for enterprise environments - virtualization; high performance computing (HPC) and it works with Mac, Linux and Windows clients. Conclusions are: (1) GlusterFS proved to have widespread capabilities as a virtual file system; (2) Scalability is very dependent upon the underlying hardware; (3) Lack of built-in encryption and security paradigm; and (4) Best suited in a general purpose computing environment.

99 GENERAL AND MISCELLANEOUS↗

Assessing the Performance and Scalability of a Novel Multilevel K-Nomial Allgather on CORE-Direct Systems

In this paper, we propose a novel allgather algorithm, Reindexed Recursive K-ing (RRK), which leverages flexibility in the algorithm's tree topology and ability to make asynchronous progress coupled with Core-Direct communication offload capability to optimize the MPI-Allgather for Core-Direct enabled systems. In particular, the RRK introduces a reindexing scheme which ensures contiguous data transfers while adding only a single additional send and receive operation for any radix, k, or communicator size, N. This allows us to improve algorithm scalability by avoiding the use of a scatter/gather elements (SGE) list on InfiniBand networks. The implementations of the RRK algorithm and its evaluation shows that it performs and scales well on Core-Direct systems for a wide range of message sizes and various communicator configurations.

99 GENERAL AND MISCELLANEOUS↗

Optimizing Blocking and Nonblocking Reduction Operations for Multicore Systems: Hierarchical Design and Implementation

Many scientific simulations, using the Message Passing Interface (MPI) programming model, are sensitive to the performance and scalability of reduction collective operations such as MPI Allreduce and MPI Reduce. These operations are the most widely used abstractions to perform mathematical operations over all processes that are part of the simulation. In this work, we propose a hierarchical design to implement the reduction operations on multicore systems. This design aims to improve the efficiency of reductions by 1) tailoring the algorithms and customizing the implementations for various communication mechanisms in the system 2) providing the ability to configure the depth of hierarchy to match the system architecture, and 3) providing the ability to independently progress each of this hierarchy. Using this design, we implement MPI Allreduce and MPI Reduce operations (and its nonblocking variants MPI Iallreduce and MPI Ireduce) for all message sizes, and evaluate on multiple architectures including InfiniBand and Cray XT5. We leverage and enhance our existing infrastructure, Cheetah, which is a framework for implementing hierarchical collective operations to implement these reductions. The experimental results show that the Cheetah reduction operations outperform the production-grade MPI implementations such as Open MPI default, Cray MPI, and MVAPICH2, demonstrating its efficiency, flexibility and portability. On Infini- Band systems, with a microbenchmark, a 512-process Cheetah nonblocking Allreduce and Reduce achieves a speedup of 23x and 10x, respectively, compared to the default Open MPI reductions. The blocking variants of the reduction operations also show similar performance benefits. A 512-process nonblocking Cheetah Allreduce achieves a speedup of 3x, compared to the default MVAPICH2 Allreduce implementation. On a Cray XT5 system, a 6144-process Cheetah Allreduce outperforms the Cray MPI by 145%. The evaluation with an application kernel, Conjugate Gradient solver, shows that the Cheetah reductions speeds up total time to solution by 195%, demonstrating the potential benefits for scientific simulations.

Gorentla Venkata, Manjunath↗

IBTopo

IBTopo is a set of perl modules and scripts that enable one to determine the network topology of an Infiniband network.

Moody, A. T.↗

OpenSM Monitoring System

The OpenSM Monitoring System includes a collection of diagnostic and monitoring tools for use on Infiniband networks. The information this system gathers is obtained from a service, which in turn is obtained directly from the OpenSM subnet manager.

Meier, T. A.↗

Single Sided Messaging v. 0.6.6

Single-Sided Messaging (SSM) is a portable, multitransport networking library that enables applications to leverage potential one-sided capabilities of underlying network transports. It also provides desirable semantics that services for highperformance, massively parallel computers can leverage, such as an explicit cancel operation for pending transmissions, as well as enhanced matching semantics favoring large numbers of buffers attached to a single match entry. This release supports TCP/IP, shared memory, and Infiniband.

Curry, Matthew Leon↗

Announcing Supercomputer Summit

Summit is the next leap in leadership-class computing systems for open science. With Summit we will be able to address, with greater complexity and higher fidelity, questions concerning who we are, our place on earth, and in our universe. Summit will deliver more than five times the computational performance of Titan’s 18,688 nodes, using only approximately 3,400 nodes when it arrives in 2017. Like Titan, Summit will have a hybrid architecture, and each node will contain multiple IBM POWER9 CPUs and NVIDIA Volta GPUs all connected together with NVIDIA’s high-speed NVLink. Each node will have over half a terabyte of coherent memory (high bandwidth memory + DDR4) addressable by all CPUs and GPUs plus 800GB of non-volatile RAM that can be used as a burst buffer or as extended memory. To provide a high rate of I/O throughput, the nodes will be connected in a non-blocking fat-tree using a dual-rail Mellanox EDR InfiniBand interconnect. Upon completion, Summit will allow researchers in all fields of science unprecedented access to solving some of the world’s most pressing challenges.

97 MATHEMATICS AND COMPUTING↗

Portals Reference Implementation v. 1.0

The Portals reference implementation is based on the Portals 4.X API, published by Sandia National Laboratories as a freely available public document. It is designed to be an implementation of the Portals Networking Application Programming Interface and is used by several other upper layer protocols like SHMEM, GASNet and MPI. It is implemented over existing networks, specifically Ethernet and InfiniBand networks. This implementation provides Portals networks functionality and serves as a software emulation of Portals compliant networking hardware. It can be used to develop software using the Portals API prior to the debut of Portals networking hardware, such as Bull’s BXI interconnect, as well as a substitute for portals hardware on development platforms that do not have Portals compliant hardware. The reference implementation provides new capabilities beyond that of a typical network, namely the ability to have messages matched in hardware in a way compatible with upper layer software such as MPI or SHMEM. It also offers methods of offloading network operations via triggered operations, which can be used to create offloaded collective operations. Specific details on the Portals API can be found at http://portals4.org.

Grant, Ryan E.↗

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↗

IBCheck v0.1

IBCheck is a utility that can help data center, especially large High Performance Supercomputing centers to manage and troubleshoot Infiniband related issues.

QIN, YONG↗

ANALYSIS OF COMMUNICATION PERFORMANCE DEGRADATION OF THE RADIATION TRANSPORT CODE PIDOTS ON HIGH-UTILIZATION, MULTI-USER HPC SYSTEMS

The PIDOTS radiation transport code implements a spatially decomposed Integral Transport Matrix Method (ITMM) solver, which is intended to fully utilize the capabilities of modern, massively parallel high-performance computing (HPC) systems. While the code shows promising results, there was also an unexpected loss of parallel efficiency on the test systems as the number of participating processors grew and the parallelization grain size got finer. We seek to identify and characterize the communication-latency based effects that contribute to this slowdown. Through the creation of a high-level, parameterized representation of the workload of the code, the communication methodology was identified as the likely reason. From there, low-level, InfiniBand communication performance data measured on the Falcon HPC at Idaho National Laboratory (INL) was used to produce a discrete-event model of the communication scheme. The data, as well as the model, show that infrequent, large spikes in send/receive latency drive an unexpectedly substantial increase in projected runtime of a code. Furthermore, these effects are more pronounced under the high-utilization, scattered-workload environment of Falcon. Based on results from the model, it can be shown that a high-communication volume, small message-size regime can easily lead to slowdowns of ~5-10x, when compared to ideal results, as the processor count increases.

97 MATHEMATICS AND COMPUTING↗

The CMS Event-Builder System for LHC Run 3 (2021-23)

The data acquisition system (DAQ) of the CMS experiment at the CERN Large Hadron Collider (LHC) assembles events of 2MB at a rate of 100 kHz. The event builder collects event fragments from about 750 sources and assembles them into complete events which are then handed to the highlevel trigger (HLT) processes running on O(1000) computers. The aging eventbuilding hardware will be replaced during the long shutdown 2 of the LHC taking place in 2019/20. The future data networks will be based on 100 Gb/s interconnects using Ethernet and Infiniband technologies. More powerful computers may allow to combine the currently separate functionality of the readout and builder units into a single I/O processor handling simultaneously 100 Gb/s of input and output trac. It might be beneficial to preprocess data originating from specific detector parts or regions before handling it to generic HLT processors. Therefore, we will investigate how specialized coprocessors, e.g. GPUs, could be integrated into the event builder. We will present the envisioned changes to the event-builder compared to today’s system. Initial measurements of the performance of the data networks under the event-building trac pattern will be shown. Implications of a folded network architecture for the event building and corresponding changes to the software implementation will be discussed. Corresponding author: remigius.mommsen@cern.c

72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗

Overlapping Computation and Communication: Barrier Algorithms and ConnectX-2 CORE-Direct Capabilities

This paper explores the computation and communication overlap capabilities enabled by the new CORE-Direct hardware capabilities introduced in the InfiniBand (IB) Host Channel Adapter (HCA) ConnectX-2. These capabilities enable the progression and completion of data-dependent communications sequences to progress and complete at the network level without any Central Processing Unit (CPU) involvement. We use the latency dominated nonblocking barrier algorithm in this study, and find that at 64 process count, a contiguous time slot of about 80 percent of the nonblocking barrier time is available for computation. This time slot increases as the number of processes participating increases. In contrast, CPU based implementations provide a time slot of up to 30 percent of the nonblocking barrier time. This bodes well for the scalability of simulations employing offloaded collective operations. These capabilities can be used to reduce the effects of system noise, and when using nonblocking collective operations have the potential to hide the effects of application load imbalance.

99 GENERAL AND MISCELLANEOUS↗

ConnectX2 In niBand Management Queues: New support for Network Of oaded

This paper introduces the newly developed InfiniBand (IB) Management Queue capability, used by the Host Channel Adapter (HCA) to manage network task data flow dependancies, and progress the communications associated with such flows. These tasks include sends, receives, and the newly supported wait task, and are scheduled by the HCA based on a data dependency description provided by the user. This functionality is supported by the ConnectX-2 HCA, and provides the means for delegating collective communication management and progress to the HCA, also known as collective communication offload. This provides a means for overlapping collective communications managed by the HCA and computation on the Central Processing Unit (CPU), thus making it possible to reduce the impact of system noise on parallel applications using collective operations. This paper further describes how this new capability can be used to implement scalable Message Passing Interface (MPI) collective operations, describing the high level details of how this new capability is used to implement the MPI Barrier collective operation, focusing on the latency sensitive performance aspects of this new capability. This paper concludes with small scale benchmark experiments comparing implementations of the barrier collective operation, using the new network offload capabilities, with established point-to-point based implementations of these same algorithms, which manage the data flow using the central processing unit. These early results demonstrate the promise this new capability provides to improve the scalability of high-performance applications using collective communications. The latency of the HCA based implementation of the barrier is similar to that of the best performing point-to-point based implementation managed by the central processing unit, starting to outperform these as the number of processes involved in the collective operation increases.

99 GENERAL AND MISCELLANEOUS↗