The FAIR principles for scientific data (Findable, Accessible, Interoperable, Reusable) are also relevant to other digital objects such as research software and scientific workflows that operate on scientific data. The FAIR principles can be applied to the data being handled by a scientific workflow as well as the processes, software, and other infrastructure which are necessary to specify and execute a workflow. The FAIR principles were designed as guidelines, rather than rules, that would allow for differences in standards for different communities and for different degrees of compliance. There are many practical considerations which impact the level of FAIR-ness that can actually be achieved, including policies, traditions, and technologies. Because of these considerations, obstacles are often encountered during the workflow lifecycle that trace directly to shortcomings in the implementation of the FAIR principles. Here, we detail some cases, without naming names, in which data and workflows were Findable but otherwise lacking in areas commonly needed and expected by modern FAIR methods, tools, and users. We describe how some of these problems, all of which were overcome successfully, have motivated us to push on systems and approaches for fully FAIR workflows.

Persistent memory (PMEM) devices can achieve comparable performance to DRAM while providing significantly more capacity. This has made the technology compelling as an expansion to main memory. Rethinking PMEM as storage devices can offer a high performance buffering layer for HPC applications to temporarily, but safely store data. However, modern parallel I/O libraries, such as HDF5 and pNetCDF, are complicated and introduce significant software and metadata overheads when persisting data to these storage devices, wasting much of their potential. In this work, we explore the potential of PMEM as storage through pMEMCPY: a simple, lightweight, and portable I/O library for storing data in persistent memory. We demonstrate that our approach is up to 2x faster than other popular parallel I/O libraries under real workloads.

The FAIR principles of open science (Findable, Accessible, Interoperable, and Reusable) have had transformative effects on modern large-scale computational science. In particular, they have encouraged more open access to and use of data, an important consideration as collaboration among teams of researchers accelerates and the use of workflows by those teams to solve problems increases. How best to apply the FAIR principles to workflows themselves, and software more generally, is not yet well understood. We argue that the software engineering concept of technical debt management provides a useful guide for application of those principles to workflows, and in particular that it implies reusability should be considered as ‘first among equals’. Moreover, our approach recognizes a continuum of reusability where we can make explicit and selectable the tradeoffs required in workflows for both their users and developers.To this end, we propose a new abstraction approach for reusable workflows, with demonstrations for both synthetic workloads and real-world computational biology workflows. Through application of novel systems and tools that are based on this abstraction, these experimental workflows are refactored to rightsize the granularity of workflow components to efficiently fill the gap between end-user simplicity and general customizability. Our work makes it easier to selectively reason about and automate the connections between trade-offs across user and developer concerns when exposing degrees of freedom for reuse. Additionally, by exposing fine-grained reusability abstractions we enable performance optimizations, as we demonstrate on both institutional-scale and leadership-class HPC resources.

Remote Direct Memory Access (RDMA) capabilities have been provided by high-end networks for many years, but the network environments surrounding RDMA are evolving. RDMA performance has historically relied on using strict ordering guarantees to determine when data transfers complete, but modern adaptively-routed networks no longer provide those guarantees. RDMA also exposes low-level details about memory buffers: either all clients are required to coordinate access using a single shared buffer, or exclusive resources must be allocatable per-client for an unbounded amount of time. This makes RDMA unattractive for use in many-to-one communication models such as those found in public internet client-server situations. Remote Virtual Memory Access (RVMA) is a novel approach to data transfer which adapts and builds upon RDMA to provide better usability, resource management, and fault tolerance. RVMA provides a lightweight completion notification mechanism which addresses RDMA performance penalties imposed by adaptively-routed networks, enabling high-performance data transfer regardless of message ordering. RVMA also provides receiver-side resource management, abstracting away previously-exposed details from the sender-side and removing the RDMA requirement for exclusive/coordinated resources. RVMA requires only small hardware modifications from current designs, provides performance comparable or superior to traditional RDMA networks, and offers many new features. In this paper, we describe RVMA's receiver-managed resource approach and how it enables a variety of new data-transfer approaches on high-end networks. In particular, we demonstrate how an RVMA NIC could implement the first hardware-based fault tolerant RDMA-like solution. We present the design and validation of an RVMA simulation model in a popular simulation suite and use it to evaluate the advantages of RVMA at large scale. In addition to support for adaptive routing and easy programmability, RVMA can outperform RDMA on a 3D sweep application by 4.4X.

Summary Coordinated checkpoint/restart is currently the dominant approach to mitigating the impact of failures on important scientific applications running on large‐scale distributed systems. However, there is widespread evidence that coordinated checkpointing may no longer be viable on next‐generation systems. Uncoordinated checkpoint/restart attempts to address the shortcomings of coordinated checkpoint/restart by allowing application processes to checkpoint their state independently. However, eliminating coordination may significantly degrade application performance. In this paper, we propose an approach that leverages existing coordination in important scientific applications to approximately coordinate checkpoints. Specifically, we propose to extend MPI implementations to force checkpoints to occur immediately after the completion of a collective operation. We evaluate the performance implications of this approach using an existing validated simulation framework. Our results demonstrate that approximately coordinated checkpointing can significantly improve application performance relative to totally uncoordinated checkpointing. We also show that forcing checkpoints to occur following a collective operation has a small impact on the nominal checkpoint interval for several important workloads. As a whole, the results presented in this paper demonstrate that approximately coordinated checkpointing may provide significant performance benefits without significantly increasing the cost of failure recovery.

Scientific data collections grow ever larger, both in terms of the size of individual data items and of the number and complexity of items. To use and manage them, it is important to directly address issues of robust and actionable provenance. We identify three key drivers as our focus: managing the size and complexity of metadata, lack of a priori information to match usage intents between publishers and consumers of data, and support for campaigns over collections of data driven by multi-disciplinary, collaborating teams. We introduce the Hoarde abstraction as an attempt to formalize a way of looking at collections of data to make them more tractable for later use. Hoarde leverages middleware and systems infrastructures for scientific and technical data management. Through the lens of a select group of challenging data usage scenarios, we discuss some of the aspects of implementation, usage, and forward portability of this new view on data management.

Next-generation applications increasingly rely on in situ analytics to guide computation, reduce the amount of I/O performed, and perform other important tasks. Scheduling where and when to run analytics is challenging, however. This paper quantifies the costs and benefits of different approaches to scheduling applications and analytics on nodes in large-scale applications, including space sharing, uncoordinated time sharing, and gang scheduled time sharing.

Scientific workloads running on current extreme-scale systems routinely generate tremendous volumes of data for postprocessing. This data movement has become a serious issue due to its energy cost and the fact that I/O bandwidths have not kept pace with data generation rates. In situ analytics is an increasingly popular alternative in which post-simulation processing is embedded into an application, running as part of the same MPI job. This can reduce data movement costs but introduces a new potential source of interference for the application. Using a validated simulation-based approach, we investigate how best to mitigate the interference from time-shared in situ tasks for a number of key extreme-scale workloads. This paper makes a number of contributions. First, we show that the independent scheduling of in situ analytics tasks can significantly degradation application performance, with slowdowns exceeding 1000%. Second, we demonstrate that the degree of synchronization found in many modern collective algorithms is sufficient to significantly reduce the overheads of this interference to less than 10% in most cases. Finally, we show that many applications already frequently invoke collective operations that use these synchronizing MPI algorithms. Therefore, the syncronization introduced by these MPI collective algorithms can be leveraged to efficiently schedule analytics tasks with minimal changes to existing applications. This paper provides critical analysis and guidance for MPI users and developers on the importance of scheduling in situ analytics tasks. It shows the degree of synchronization needed to mitigate the performance impacts of these time-shared coupled codes and demonstrates how that synchronization can be realized in an extreme-scale environment using modern collective algorithms.

Next-generation systems face a wide range of new potential sources of application interference, including resilience actions, system software adaptation, and in situ analytics programs. In this paper, we present a new model for analyzing the performance of bulk-synchronous HPC applications based on the use of extreme value theory. After validating this model against both synthetic and real applications, the paper then uses both simulation and modeling techniques to profile next-generation interference sources and characterize their behavior and performance impact on a selection of HPC benchmarks, mini-applications, and applications. Lastly, this work shows how the model can be used to understand how current interference mitigation techniques in multi-processors work.

Fault-tolerance poses a major challenge for future large-scale systems. Active research into coordinated, uncoordinated, and hybrid check pointing systems has explored how the introduction of asynchrony can address anticipated scalability issues. However, few insights into selection and tuning of these protocols for applications at scale have emerged. In this paper, we use a simulation-based approach to show that local checkpoint activity in resilience mechanisms can significantly affect the performance of key workloads, even when less than 1% of a local node's compute time is allocated to resilience mechanisms (a very generous assumption). Specifically, we show that even though much work on uncoordinated check pointing has focused on optimizing message log volumes, local check pointing activity may dominate the overheads of this technique at scale. Our study shows that local checkpoints lead to process delays that can propagate through messaging relations to other processes causing a cascading series of delays. We demonstrate how to tune hierarchical uncoordinated check pointing protocols designed to reduce log volumes to significantly reduce these synchronization overheads at scale. Our work provides a critical analysis and comparison of coordinated and uncoordinated check pointing and enables users and system administrators to fine-tune the check pointing scheme to the application and system characteristics.

The past decade has witnessed a major paradigm shift in high performance computing with the introduction of accelerators as general purpose processors. These computing devices make available very high parallel computing power at low cost and power consumption, transforming current high performance platforms into heterogeneous CPU-GPU equipped systems. Although the theoretical performance achieved by these hybrid systems is impressive, taking practical advantage of this computing power remains a very challenging problem. Most applications are still deployed to either GPU or CPU, leaving the other resource under- or un-utilized. In this paper, we propose, implement, and evaluate a performance aware scheduling technique along with optimizations to make efficient collaborative use of CPUs and GPUs on a parallel system. In the context of feature computations in large scale image analysis applications, our evaluations show that intelligently co-scheduling CPUs and GPUs can significantly improve performance over GPU-only or multi-core CPU-only approaches.

Pathology is a medical subspecialty that practices the diagnosis of disease. Microscopic examination of tissue reveals information enabling the pathologist to render accurate diagnoses and to guide therapy. The basic process by which anatomic pathologists render diagnoses has remained relatively unchanged over the last century, yet advances in information technology now offer significant opportunities in image-based diagnostic and research applications. Pathology has lagged behind other healthcare practices such as radiology where digital adoption is widespread. As devices that generate whole slide images become more practical and affordable, practices will increasingly adopt this technology and eventually produce an explosion of data that will quickly eclipse the already vast quantities of radiology imaging data. These advances are accompanied by significant challenges for data management and storage, but they also introduce new opportunities to improve patient care by streamlining and standardizing diagnostic approaches and uncovering disease mechanisms. Computer-based image analysis is already available in commercial diagnostic systems, but further advances in image analysis algorithms are warranted in order to fully realize the benefits of digital pathology in medical discovery and patient care. In coming decades, pathology image analysis will extend beyond the streamlining of diagnostic workflows and minimizing interobserver variability and will begin to provide diagnostic assistance, identify therapeutic targets, and predict patient outcomes and therapeutic responses.

Modern HPC applications, for example adaptive mesh refinement and multi-physics codes, have dynamic communication characteristics which result in poor performance on current MPI implementations. Current MPI implementations do not change transport protocols or allocate resources based on the application characteristics, resulting in degraded application performance. In this paper, we describe PRO-MPI, a protocol reconfiguration and optimization system for MPI that we are developing to meet the needs of dynamic modern HPC applications. PRO-MPI uses profiles of past application communication characteristics to dynamically reconfigure MPI protocol choices. We show that such dynamic reconfiguration can improve the performance of important MPI applications significantly when exact communication profiles are known. We also present preliminary data showing that profiles from past application runs with different (but related) inputs can be used to optimize the performance of later application runs.

Abstract In the early 1990s, researchers at Sandia National Laboratories and the University of New Mexico began development of customized system software for massively parallel ‘capability’ computing platforms. These lightweight kernels have proven to be essential for delivering the full power of the underlying hardware to applications. This claim is underscored by the success of several supercomputers, including the Intel Paragon, Intel Accelerated Strategic Computing Initiative Red, and the Cray XT series of systems, each having established a new standard for high‐performance computing upon introduction. In this paper, we describe our approach to lightweight compute node kernel design and discuss the design principles that have guided several generations of implementation and deployment. A broad strategy of operating system specialization has led to a focus on user‐level resource management, deterministic behavior, and scalable system services. The relative importance of each of these areas has changed over the years in response to changes in applications and hardware and system architecture. We detail our approach and the associated principles, describe how our application of these principles has changed over time, and provide design and performance comparisons to contemporaneous supercomputing operating systems. Copyright © 2008 John Wiley & Sons, Ltd.

Current and emerging large-scale HPC applications face daunting I/O challenges. In existing codes, problems arise both from large data volumes and from the need to perform complex online data manipulations, including data staging, reorganization, and transformation. We describe three related techniques for enabling, encouraging, and exploiting latent I/O asynchrony in HPC applications: data taps, IOgraphs, and Metabots.

Efficient data movement is an important part of any high-performance I/O system, but it is especially critical for the current and next-generation of massively parallel processing (MPP) systems. In this paper, we discuss how the scale, architecture, and organization of current and proposed MPP systems impact the design of the data-movement scheme for the I/O system. We also describe and analyze the approach used by the lightweight file systems (LWFS) project, and we compare that approach to more conventional data-movement protocols used by small and mid-range clusters. Our results indicate that the data-movement strategy used by LWFS clearly outperforms conventional data-movement protocols, particularly as data sizes increase

Today's high-end massively parallel processing (MPP) machines have thousands to tens of thousands of processors, with next-generation systems planned to have in excess of one hundred thousand processors. For systems of such scale, efficient I/O is a significant challenge that cannot be solved using traditional approaches. In particular, general purpose parallel file systems that limit applications to standard interfaces and access policies do not scale and will likely be a performance bottleneck for many scientific applications. In this paper, we investigate the use of a "lightweight" approach to I/O that requires the application or I/O-library developer to extend a core set of critical I/O functionality with the minimum set of features and services required by its target applications. We argue that this approach allows the development of I/O libraries that are both scalable and secure. We support our claims with preliminary results for a lightweight checkpoint operation on a development cluster at Sandia

No abstract available.

There is substantial interest in using SOAP (simple object access protocol) in distributed applications' interprocess communications due to its promise of universal interoperability. The utility of SOAP is limited, however, by its inefficient implementation. Our research aims to make SOAP useful for high end or resource-constrained applications. The resulting SOAP-bin communication protocol exhibits substantially improved performance compared to regular SOAP communications. Gains are particularly evident when the same types of parameters are exchanged repeatedly, examples including transactional applications, remote graphics or visualization, and distributed scientific codes. A further improvement to SOAP-bin, termed SOAP-binQ, addresses resource-constrained applications like distributed media codes, where scarce communication bandwidth, for example, may prevent end users from interacting in real-time. SOAP-binQ offers quality management functions that permit SOAP to reduce parameter sizes dynamically, as and when needed. The methods used in size reduction are provided by end users and/or by applications, thereby enabling domain-specific tradeoffs in quality vs. performance. An adaptive use of SOAP-binQ's quality management techniques presented significantly reduces the jitter experienced in two sample applications like remote sensing and remote visualization.

We explore the use of compression methods to improve the middleware-based exchange of information in interactive or collaborative distributed applications. In such applications, good compression factors must be accompanied by compression speeds suitable for the data transfer rates sustainable across network links. Our approach combines methods that continuously monitor current network and processor resources and assess compression effectiveness, with techniques that automatically choose suitable compression techniques. The resulting network- and user-aware compression methods are evaluated experimentally across a range of network links and application data, the former ranging from low end links to homes, to wide-area Internet links, to high end links in intranets, the latter including both scientific (binary molecular dynamics data) and commercial (XML) data sets. Results attained demonstrate substantial improvements of this adaptive technique for data compression over non-adaptive approaches, where better compression methods are used when CPU loads are low and/or network links are slow, and where less effective and typically, faster compression techniques are used in high end network infrastructures.

We present a mechanism for providing differential data protection to publish/subscribe distributed systems, such as those used in peer-to-peer computing, grid environments, and others. This mechanism, termed "security overlays", incorporates credential-based communication channel creation, subscription and extension. We describe a conceptual model of publish/subscribe services that is made concrete by our mechanism. We also present an application, active video streams, whose reimplementation using security overlays allows it to react to high-level security policies specified in XML without significant performance loss or the necessity for embedding policy-specific code into the application.

In this paper, we argue that an exclusively inactive interface to directory services can hinder server scalability and indirectly restrict the behavior of potential applications. We propose to extend directory services' interfaces with a proactive mode by which clients can express their interest in (and be notified of) changes in the environment. These notification channels can be subsequently customized on a per-client basis through client-specified filters. Finally, in order to simplify the handling of client/server failures we adopt a leasing model for client registration to (and customization of) a notification channel. To validate our approach, we have designed and implemented the Proactive Directory Service (PDS).

Common to computational grids and pervasive computing is the need for an expressive, efficient, and scalable directory service that provides information about objects in the environment. We argue that a directory interface that ‘pushes’ information to clients about changes to objects can significantly improve scalability. This paper describes the design, implementation, and evaluation of the Proactive Directory Service (PDS). PDS’ interface supports a customizable ‘proactive’ mode through which clients can subscribe to be notified about changes to their objects of interest. Clients can dynamically tune the detail and granularity of these notifications through filter functions instantiated at the server or at the object’s owner, and by remotely tuning the functionality of those filters. We compare PDS’ performance against off-the-shelf implementations of DNS and the Lightweight Directory Access Protocol. Our evaluation results confirm the expected performance advantages of this approach and demonstrate that customized notification through filter functions can reduce bandwidth utilization while improving the performance of both clients and directory servers.

Summary form only given. An increasing number of distributed applications aim to provide services to users by interacting with a correspondingly growing set of data-intensive network services. To support such requirements, we believe that new services need to be customizable, applications need to be dynamically extensible, and both applications and services need to be able to adapt to variations in resource availability and demand. A comprehensive approach to building new distributed applications can facilitate this by considering the contents of the information flowing across the application and its services and by adopting a component-based model to application/service programming. It should provide for dynamic adaptation at multiple levels and points in the underlying platform; and, since the mapping of components to resources in dynamic environment is too complicated, it should relieve programmers of this task. We propose Active Streams, a middleware approach and its associated framework for building distributed applications and services that exhibit these characteristics.

High-performance computing faces considerable change as the Internet and the Grid mature. Applications that once were tightly-coupled and monolithic are now decentralized, with collaborating components spread across diverse computational elements. Such distributed systems most commonly communicate through the exchange of structured data. Definition and translation of metadata is incorporated in all systems that exchange structured data. We observe that the manipulation of this metadata can be decomposed into three separate steps: discovery, binding of program objects to the metadata, and marshaling of data to and from wire formats. We have designed a method of representing message formats in XML, using datatypes available in the XML Schema specification. We have implemented a tool, XMIT that uses such metadata and exploits this decomposition in order to provide flexible run-time metadata definition facilities for an efficient binary communication mechanism. We also demonstrate that the use of XMIT makes possible such flexibility at little performance cost.

The definition and translation of metadata is incorporated in all systems that exchange structured data. We observe that the manipulation of this metadata can be decomposed into three separate steps: discovery, binding of program objects to the metadata, and marshaling of data to and from wire formats. We have designed a method of representing message formats in XML, using data types that are available in the XML schema specification. We have implemented a tool called xml2wire that uses such metadata and exploits this decomposition in order to provide flexible metadata definition facilities for an efficient binary communications mechanism. We also observe that the use of xml2wire makes possible such flexibility without intolerable performance effects.

It is our belief that network applications relying on globally distributed shared resources will increasingly adopt meta-level descriptions to describe the data streaming in the application at runtime. Our group has developed the notion of computational data streams to describe and act upon such data flows. Our work conceptualizes the data flows as database relations over which useful operations, such as querying, can be performed. This paper shows how one can make the step from a meta-level description of data flows to an actual implementation using CORBA-style event channels and binary I/O for data transport. 1

High performance computing is being increasingly utilized in non-traditional circumstances where it must interoperate with other applications. For example, online visualization is being used to monitor the progress of applications, and real-world sensors are used as inputs to simulations. Whenever these situations arise, there is a question of what communications infrastructure should be used to link the different components. Traditional HPC-style communications systems such as MPI offer relatively high performance, but are poorly suited for developing these less tightly-coupled cooperating applications. Object-based systems and meta-data formats like XML offer substantial plug-and-play flexibility, but with substantially lower performance. We observe that the flexibility and baseline performance of all these systems is strongly determined by their `wire format', or how they represent data for transmission in a heterogeneous environment. We examine the performance implications of different wire formats and present an alternative with significant advantages in terms of both performance and flexibility.