Workload-adaptive Scheduling for Efficient Use of Parallel File Systems in High-performance Computing Clusters
Explore the source record for details and available documents.
Goponenko, Alexander V.↗
SEARCH · Search NASA
Results for “File System”
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.
Design and implementation of I/O performance prediction scheme on HPC systems through large-scale log analysis
Abstract Large-scale high performance computing (HPC) systems typically consist of many thousands of CPUs and storage units used by hundreds to thousands of users simultaneously. Applications from large numbers of users have diverse characteristics, such as varying computation, communication, memory, and I/O intensity. A good understanding of the performance characteristics of each user application is important for job scheduling and resource provisioning. Among these performance characteristics, I/O performance is becoming increasingly important as data sizes rapidly increase and large-scale applications, such as simulation and model training, are widely adopted. However, predicting I/O performance is difficult because I/O systems are shared among all users and involve many layers of software and hardware stack, including the application, network interconnect, operating system, file system, and storage devices. Furthermore, updates to these layers and changes in system management policy can significantly alter the I/O behavior of applications and the entire system. To improve the prediction of the I/O performance on HPC systems, we propose integrating information from several different system logs and developing a regression-based approach to predict the I/O performance. Our proposed scheme can dynamically select the most relevant features from the log entries using various feature selection algorithms and scoring functions, and can automatically select the regression algorithm with the best accuracy for the prediction task. The evaluation results show that our proposed scheme can predict the write performance with up to 90% prediction accuracy and the read performance with up to 99% prediction accuracy using the real logs from the Cori supercomputer system at NERSC.
97 MATHEMATICS AND COMPUTING↗
Application Crash Consistency and Performance with CCFS
Recent research has shown that applications often incorrectly implement crash consistency. In this paper, we present the Crash-Consistent File System (ccfs), a file system that improves the correctness of application-level crash consistency protocols while maintaining high performance. A key idea in ccfs is the abstraction of a stream. Within a stream, updates are committed in program order, improving correctness; across streams, there are no ordering restrictions, enabling scheduling flexibility and high performance. We empirically demonstrate that applications running atop ccfs achieve high levels of crash consistency. Further, we show that ccfs performance under standard file-system benchmarks is excellent, in the worst case on par with the highest performing modes of Linux ext4, and in some cases notably better. Finally, overall, we demonstrate that both application correctness and high performance can be realized in a modern file system.
97 MATHEMATICS AND COMPUTING↗
VerifyIO: Ensuring Correctness of Consistency Semantics in Parallel I/O
Abstract—High-performance computing (HPC) applications generate and consume substantial amounts of data, typically managed by parallel file systems. These applications access file systems either through the POSIX interface or by using highlevel I/O libraries. While the POSIX consistency model remains dominant in HPC, emerging file systems and popular I/O libraries increasingly adopt alternative consistency models that relax semantics in various ways, creating significant challenges for correctness and portability. This paper addresses these challenges by proposing a trace-driven I/O consistency verification workflow, implemented in our open-source tool, VerifyIO, which collects execution traces, detects data conflicts, and verifies proper synchronization against specified consistency models. Our extensive evaluation of 91 test case executions across three widely used I/O libraries with four I/O consistency models reveals critical consistency issues at both application and implementation levels.
Consistency Semantics↗
Direct-FUSE
The Direct-FUSE framework supports multiple FUSE file systems as well as other, custom user-level file systems in user space without the need to cross the user/kernel boundary into the FUSE kernel module. All layers of Direct-FUSE are in user space, and applications can directly use pre-defined unified file system calls to interact with different user-defined file systems. Our performance results show that Direct- FUSE can outperform some native FUSE file systems by 11.9% on average and does not add significant overhead over backend file systems.
SATO,, KENTO↗
Measurements and analytics of wide-area file transfers over dedicated connections
Distributed scientific and big-data computations are becoming increasingly dependent on access to remote files. Wide-area file transfers are supported by two basic schemes: (i) application-level tools, such as GridFTP, that provide transport services between file systems housed at geographically separated sites, and (ii) file systems mounted over wide-area networks, using mechanisms such as LNet routers that make them transparently available. In both cases, the file transfer performance critically depends on the configuration consisting of host, file, IO, and disk subsystems, which are complex by themselves, as well as on their complex compositions implemented using buffers and IO-network data transitions. We present extensive file transfer rate measurements collected over dedicated 10 Gbps connections with 0-366 ms round-trip times, using GridFTP and XDD file transfer tools, and Lustre file system extended over wide-area networks using LNet routers. Our test configurations are composed of: three types of host systems; XFS, Lustre, and ext3 file systems; and Ethernet and SONET wide-area connections. We present analytics based on the convexity-concavity of throughput profiles which provide insights into throughput and its superior or inferior trend compared to linear interpolations. We propose the utilization-concavity coefficient, a scalar metric that characterizes the overall performance of any file transfer method consisting of specific configuration and scheme. Our results enable performance optimizations by highlighting the significant roles of (i) buffer sizes and parallelism in GridFTP and XDD, and (ii) buffer utilization and credit mechanism in LNet routers.
Rao, Nageswara↗
Profiling the Usage of an Extreme-Scale Archival Storage System
Profiling the archival storage system in scientific computing environments has received much less attention compared to the parallel file system, but is equally important since it stores the final data products safely, for a long duration. In this paper, we analyze eight years worth of data transfer logs for accessing the archival file system (HPSS) in the Oak Ridge Leadership Computing Facility (OLCF), which has been hosting the world's largest supercomputers and file systems. Our analysis encompasses about 135 million data transfer activities to the 80 PB High Performance Storage System (HPSS), between 2010 and 2017. We analyze the logs from several dimensions, including studying the workload characteristics (e.g., access patterns, frequency of accesses and temporal behavior), file system characteristics (e.g., directory depth, file system scaling trends, file types), and scientific user behavior (e.g., domain-specific usage and organization). Based on the analysis, we derive insights into the future evolution of the archive in terms of provisioning, desired features and functionality from the archive software, role and right sizing of the archive tiers, quota management, and the importance of smart and efficient metadata and storage management. We believe our study will prove useful for both operating current archival storage and better provisioning future systems.
Sim, Hyogi↗
Understanding Lustre Internals. Second Edition
The Lustre file system has become a preferred storage resource for systems on the Top500 list, and it is often the file system of choice for small- to medium-sized HPC systems that require parallel shared access to data. Several resources exist to help users deploy and configure Lustre, but the same cannot be said for resources that explain the inner workings of the Lustre source code. A previous ORNL technical report entitled "Understanding Lustre Filesystem Internals" (ORNL/TM-2009/117) provided an excellent summary of Lustre subsystem operations. However, that report is over a decade old and is based on Lustre version 1.6. Since that report was published, Lustre has evolved significantly. Several subsystems underwent significant code changes and many new features have been added to the file system, bringing the current Lustre version up to 2.15.This report aims to document and explain the internal workings of the latest version of the Lustre file system. It will provide more complete and up-to-date information than the previous technical report and should serve as a foundational document for anyone interested in Lustre software development. Key data structures will be described along with the APIs used for interaction among the various Lustre subsystems. Although the Lustre software is constantly being developed, the details in this document should remain relevant for the forseeable future.
97 MATHEMATICS AND COMPUTING↗
Can Applications Recover from fsync Failures?
We analyze how file systems and modern data-intensive applications react to fsync failures. First, we characterize how three Linux file systems (ext4, XFS, Btrfs) behave in the presence of failures. We find commonalities across file systems (pages are always marked clean, certain block writes always lead to unavailability) as well as differences (page content and failure reporting is varied). Next, we study how five widely used applications (PostgreSQL, LMDB, LevelDB, SQLite, Redis) handle fsync failures. Our findings show that although applications use many failure-handling strategies, none are sufficient: fsync failures can cause catastrophic outcomes such as data loss and corruption. Our findings have strong implications for the design of file systems and applications that intend to provide strong durability guarantees.
Computer Science↗
A fully unprivileged CernVM-FS
The CernVM File System provides the software and container distribution backbone for most High Energy and Nuclear Physics experiments. It is implemented as a file system in user-space (Fuse) module, which permits its execution without any elevated privileges. Yet, mounting the file system in the first place is handled by a privileged suid helper program that is installed by the Fuse package on most systems. The privileged nature of the mount system call is a serious hindrance to running CernVM-FS on opportunistic resource and supercomputers. Fortunately, recent developments in the Linux kernel and in the Fuse user-space libraries enabled fully unprivileged mounting for Fuse file systems (as of RHEL 8), or at least outsourcing the privileged mount system call to a custom, external process. This opens the door to several, very appealing new ways to use CernVM-FS, such as a generally usable “super pilot” consisting of the pilot code bundled with Singularity and CernVM-FS, or the on-demand instantiation of unprivileged, ephemeral containers to publish new CernVM-FS content from anywhere. In this contribution, we discuss the integration of these new Linux features with CernVM-FS and show some of its most promising, new applications.
72 PHYSICS OF ELEMENTARY PARTICLES AND FIELDS↗
A Fault-oblivious Extreme-scale Execution Environment
Exascale computing systems will provide a thousand-fold increase in parallelism and a proportional increase in failure rate relative to today's machines. Systems software for exascale machines must provide the infrastructure to support existing applications while simultaneously enabling efficient execution of new programming models that naturally express dynamic, adaptive, irregular computation; coupled simulations; and massive data analysis in a highly unreliable hardware environment with billions of threads of execution. We propose a new approach to the data and work distribution model provided by system software based on the unifying formalism of an abstract file system. The proposed hierarchical data model provides simple, familiar visibility and access to data structures through the file system hierarchy, while providing fault tolerance through selective redundancy. The hierarchical task model features work queues whose form and organization are represented as file system objects. Data and work are both first class entities. By exposing the relationships between data and work to the runtime system, information is available to optimize execution time and provide fault tolerance. The data distribution scheme provides replication (where desirable and possible) for fault tolerance and efficiency, and it is hierarchical to make it possible to take advantage of locality. The user, tools, and applications, including legacy applications, can interface with the data, work queues, and one another through the abstract file model. This runtime environment will provide multiple interfaces to support traditional Message Passing Interface applications, languages developed under DARPA's High Productivity Computing Systems program, as well as other, experimental programming models. We will validate our runtime system with pilot codes on existing platforms and will use simulation to validate for exascale-class platforms. In this final report, we summarize research results from the work done at the Ohio State University towards the larger goals of the project listed above.
97 MATHEMATICS AND COMPUTING↗
I/O Bottleneck Detection and Tuning: Connecting the Dots using Interactive Log Analysis
Using parallel file systems efficiently is a tricky problem due to inter-dependencies among multiple layers of I/O software, including high-level I/O libraries (HDF5, netCDF, etc.), MPI-IO, POSIX, and file systems (GPFS, Lustre, etc.). Profiling tools such as Darshan collect traces to help understand the I/O performance behavior. However, there are significant gaps in analyzing the collected traces and then applying tuning options offered by various layers of I/O software. Seeking to connect the dots between I/O bottleneck detection and tuning, we propose DXT Explorer, an interactive log analysis tool. In this paper, we present a case study using our interactive log analysis tool to identify and apply various I/O optimizations. We report an evaluation of performance improvement achieved for four I/O kernels extracted from science applications.
Bez, Jean Luca↗
Toward a persistent event-streaming system for high-performance computing applications
High-performance computing (HPC) applications have traditionally relied on parallel file systems and file transfer services to manage data movement and storage. Alternative approaches have been proposed that use direct communications between application components, trading persistence and fault tolerance for speed. Event-driven architectures, as popularized in enterprise contexts, present a compelling middle ground, avoiding the performance cost and API constraints of parallel file systems while retaining persistence and offering impedance matching between application components. However, adapting streaming frameworks to HPC workloads requires addressing challenges unique to HPC systems. This paper investigates the potential for a streaming framework designed for HPC infrastructures and use cases. We introduce Mofka, a persistent event-streaming framework designed specifically for HPC environments. Mofka combines the capabilities of a traditional streaming service with optimizations tailored to the HPC context, such as support for massively multicore nodes, efficient scaling for large producer-consumer workflows, RDMA-enabled high-performance network communications, specialized network fabrics with multiple links per node, and efficient handling of large scientific data payloads. Built using the Mochi suite of HPC data service components, Mofka provides a lightweight, modular, and high-performance solution for persistent streaming in HPC systems. We present the architecture of Mofka and evaluate its performance against Kafka and Redpanda using benchmarks on diverse platforms, including Argonne's Polaris and Oak Ridge's Frontier supercomputers, showing up to 8× improvement in throughput in some scenarios. We then demonstrate its utility in several real-world applications: a tomographic reconstruction pipeline, a workflow for the discovery of metal-organic frameworks for carbon capture, and the instrumentation of Dask workflows for provenance tracking and performance analysis.
HPC↗
Multi-Node Program Fuzzing on High Performance Computing Resources
Significant effort is placed on tuning the internal parameters of fuzzers to explore the state space, measured as coverage, of binaries. In this work, we investigate the effects of the external environment on the resulting coverage after fuzzing two binaries with AFL for 24 hours. Parameters such as scaling to multiple nodes, node saturation, and parallel file system type on HPC resources are controlled in order to maximize coverage. It will be shown that employing a parallel file system such as IBM's General Parallel File System offers an advantage for fuzzing operations, since it contains enhancements for performance optimization. When combined with scaling to two and four nodes, while simultaneously restricting the number of coordinated AFL tasks per node on the low end (10-50% of available physical cores), coverage may be enhanced within a shorter period of time. Thus, controlling the external environment is a useful effort.
97 MATHEMATICS AND COMPUTING↗
Data Processing Unit Services Module
The Data Processing Services Module (DPUSM) provides the ability to perform pluggable compression, erasure coding, checksuming and other important file system operations within the Linux kernel. The pluggable provider interface allows for the use of hardware acceleration of those services. In-kernel file systems are then able to use these functions to use these accelerators to perform operations that are normally run on the processor, resulting in improved file system performance. Third parties will register "providers" with the DPUSM to communicate with their respective accelerators. Providers will implement functions with DPUSM API signatures so that the DPUSM can translate the data inputted by users of the DPUSM into data that providers recognize.
Lee, Jason↗
Accelerating Application Bulk Synchronous Writes in HPC Environments
High-bandwidth storage tiers are becoming more common for their capability to absorb high-rate, bursty I/Os. Notably, the designs of these fast storage tiers differ from system to system. The variation of these layers and non-uniform methods of access can pose chal- lenges for applications seeking to run at multiple HPC facilities. Therefore, in this work, we present Spectral, a rapid-output ab- straction library to accelerate application, bulk-synchronous writes on HPC systems. We design Spectral to enable applications to use high-bandwidth storage, such as node-local storage and dis- tributed, write-caches (e.g., burst buffers) transparently without requiring modifications to the application or file system source code. The key idea is to allow applications to spend most of the time performing productive work and to not require any source code changes for maximum portability on different HPC archi- tectures. Spectral internally re-routes write-only files through available, high-performance I/O resources before ultimately mi- grating them to the shared global parallel file system. For instance, on Summit, Spectral transparently places application outputs on node-local storage and then utilizes asynchronous migration to the center-wide GPFS file system. We evaluate Spectral on the Summit HPC system (1024 nodes) using the IOR benchmark and real scientific applications. Spectral shows linear performance scaling, improving application write performance by over an order of magnitude when compared to GPFS.
Khan, Awais↗
Scalable filesystem enumeration and metadata operations
Systems, apparatus, and methods are disclosed for performing scalable operations in a file system, including POSIX-like file systems. Metadata entries in a namespace or directory tree are sharded across multiple file metadata servers. An enumeration operation, such as listing a directory, is parallelized across the multiple file metadata servers, while retaining standard functionality transparently to clients. Other enumeration operations include no-output operations such as changing file attributes or deleting a file, and cumulative operations such as counting total disk space usage. The parallelization is compatible with tree-level parallelization and storage-level parallelization. Disclosed technologies can be applied to other fields requiring scalable enumeration, such as database and network applications.
Grider, Gary A.↗
SLoG: Large-Scale Logging Middleware for HPC and Big Data Convergence
Cloud developers traditionally rely on purpose-specific services to provide the storage model they need for an application. In contrast, HPC developers have a much more limited choice, typically restricted to a centralized parallel file system for persistent storage. Unfortunately, these systems often offer low performance when subject to highly concurrent, conflicting I/O patterns. This makes difficult the implementation of inherently concurrent data structures such as distributed shared logs. Yet, this data structure is key to applications such as computational steering, data collection from physical sensor grids, or discrete event generators. In this paper we tackle this issue. We present SLoG, shared log middleware providing a shared log abstraction over a parallel file system, designed to circumvent the aforementioned limitations. We evaluate SLoG’s design on up to 100,000 cores of the Theta supercomputer: the results show high append velocity at scale while also providing substantial benefits for other persistent backend storage systems.
big data↗