Parallel I/O

Last updated

Parallel I/O, in the context of a computer, means the performance of multiple input/output operations at the same time, for instance simultaneously outputs to storage devices and display devices. [1] It is a fundamental feature of operating systems. [2]

Contents

One particular instance is parallel writing of data to disk; when file data is spread across multiple disks, for example in a RAID array, one can store multiple parts of the data at the same time, thereby achieving higher write speeds than with a single device. [3] [4]

Other ways of parallel access to data include: Parallel Virtual File System, Lustre, GFS etc.

Features

Scientific computing

It is used for scientific computing and not for databases. It breaks up support into multiple layers including High level I/O library, Middleware layer and Parallel file system. [5] Parallel File System manages the single view, maintains logical space and provides access to data files. [6]

Storage

A single file may be stripped across one or more object storage target, which increases the bandwidth while accessing the file and available disk space. [7] The caches are larger in Parallel I/O and shared through distributed memory systems. [8] [9] [10] [11]

Breakthroughs

Companies have been running Parallel I/O on their servers to achieve results with regard to price and performance. Parallel processing is especially critical for scientific calculations where applications are not only CPU but also are I/O bound. [12]

See also

Related Research Articles

In computing, a block, sometimes called a physical record, is a sequence of bytes or bits, usually containing some whole number of records, having a maximum length; a block size. Data thus structured are said to be blocked. The process of putting data into blocks is called blocking, while deblocking is the process of extracting data from blocks. Blocked data is normally stored in a data buffer, and read or written a whole block at a time. Blocking reduces the overhead and speeds up the handling of the data stream. For some devices, such as magnetic tape and CKD disk devices, blocking reduces the amount of external storage required for the data. Blocking is almost universally employed when storing data to 9-track magnetic tape, NAND flash memory, and rotating media such as floppy disks, hard disks, and optical discs.

Programmed input–output is a method of data transmission, via input/output (I/O), between a central processing unit (CPU) and a peripheral device, such as a Parallel ATA storage device. Each data item transfer is initiated by an instruction in the program, involving the CPU for every transaction. In contrast, in direct memory access (DMA) operations, the CPU is uninvolved in the data transfer.

Lustre is a type of parallel distributed file system, generally used for large-scale cluster computing. The name Lustre is a portmanteau word derived from Linux and cluster. Lustre file system software is available under the GNU General Public License and provides high performance file systems for computer clusters ranging in size from small workgroup clusters to large-scale, multi-site systems. Since June 2005, Lustre has consistently been used by at least half of the top ten, and more than 60 of the top 100 fastest supercomputers in the world, including the world's No. 1 ranked TOP500 supercomputer in November 2022, Frontier, as well as previous top supercomputers such as Fugaku, Titan and Sequoia.

In computer science, asynchronous I/O is a form of input/output processing that permits other processing to continue before the transmission has finished. A name used for asynchronous I/O in the Windows API is overlapped I/O.

GPFS is high-performance clustered file system software developed by IBM. It can be deployed in shared-disk or shared-nothing distributed parallel modes, or a combination of these. It is used by many of the world's largest commercial companies, as well as some of the supercomputers on the Top 500 List. For example, it is the filesystem of the Summit at Oak Ridge National Laboratory which was the #1 fastest supercomputer in the world in the November 2019 TOP500 list of supercomputers. Summit is a 200 Petaflops system composed of more than 9,000 POWER9 processors and 27,000 NVIDIA Volta GPUs. The storage filesystem called Alpine has 250 PB of storage using Spectrum Scale on IBM ESS storage hardware, capable of approximately 2.5TB/s of sequential I/O and 2.2TB/s of random I/O.

In computer science, storage virtualization is "the process of presenting a logical view of the physical storage resources to" a host computer system, "treating all storage media in the enterprise as a single pool of storage."

Count key data (CKD) is a direct-access storage device (DASD) data recording format introduced in 1964, by IBM with its IBM System/360 and still being emulated on IBM mainframes. It is a self-defining format with each data record represented by a Count Area that identifies the record and provides the number of bytes in an optional Key Area and an optional Data Area. This is in contrast to devices using fixed sector size or a separate format track.

In IBM mainframe operating systems OS/360 and its successors, a Unit Control Block (UCB) is a memory structure, or a control block, that describes any single input/output peripheral device (unit), or an exposure (alias), to the operating system. Certain data within the UCB also instructs the Input/Output Supervisor (IOS) to use certain closed subroutines in addition to normal IOS processing for additional physical device control.

An access method is a function of a mainframe operating system that enables access to data on disk, tape or other external devices. Access methods were present in several mainframe operating systems since the late 1950s, under a variety of names; the name access method was introduced in 1963 in the IBM OS/360 operating system. Access methods provide an application programming interface (API) for programmers to transfer data to or from device, and could be compared to device drivers in non-mainframe operating systems, but typically provide a greater level of functionality.

In computing, virtualization or virtualisation is the act of creating a virtual version of something at the same abstraction level, including virtual computer hardware platforms, storage devices, and computer network resources.

The National Center for Computational Sciences (NCCS) is a United States Department of Energy (DOE) Leadership Computing Facility that houses the Oak Ridge Leadership Computing Facility (OLCF), a DOE Office of Science User Facility charged with helping researchers solve challenging scientific problems of global interest with a combination of leading high-performance computing (HPC) resources and international expertise in scientific computing.

<span class="mw-page-title-main">Input/output</span> Communication between an information processing system and the outside world

In computing, input/output is the communication between an information processing system, such as a computer, and the outside world, possibly a human or another information processing system. Inputs are the signals or data received by the system and outputs are the signals or data sent from it. The term can also be used as part of an action; to "perform I/O" is to perform an input or output operation.

Input/Output Control System (IOCS) is any of several packages on early IBM entry-level and mainframe computers that provided low level access to records on peripheral equipment. IOCS provides functionality similar to 1960s packages from other vendors, e.g., File Control Processor (FCP) in RCA 3301 Realcom Operating System, GEFRC in GECOS, and to the later Record Management Services (RMS) in DEC VAX/VMS

Blue Whale Clustered file system (BWFS) is a shared disk file system made by Tianjin Zhongke Blue Whale Information Technologies Company in China.

Software-defined storage (SDS) is a marketing term for computer data storage software for policy-based provisioning and management of data storage independent of the underlying hardware. Software-defined storage typically includes a form of storage virtualization to separate the storage hardware from the software that manages it. The software enabling a software-defined storage environment may also provide policy management for features such as data deduplication, replication, thin provisioning, snapshots and backup.

A distributed file system for cloud is a file system that allows many clients to have access to data and supports operations on that data. Each data file may be partitioned into several parts called chunks. Each chunk may be stored on different remote machines, facilitating the parallel execution of applications. Typically, data is stored in files in a hierarchical tree, where the nodes represent directories. There are several ways to share files in a distributed architecture: each solution must be suitable for a certain type of application, depending on how complex the application is. Meanwhile, the security of the system must be ensured. Confidentiality, availability and integrity are the main keys for a secure system.

Computation offloading is the transfer of resource intensive computational tasks to a separate processor, such as a hardware accelerator, or an external platform, such as a cluster, grid, or a cloud. Offloading to a coprocessor can be used to accelerate applications including: image rendering and mathematical calculations. Offloading computing to an external platform over a network can provide computing power and overcome hardware limitations of a device, such as limited computational power, storage, and energy.

<span class="mw-page-title-main">Dell Technologies PowerFlex</span> Software-defined storage product

Dell Technologies PowerFlex, is a commercial software-defined storage product from Dell Technologies that creates a server-based storage area network (SAN) from local server storage using x86 servers. It converts this direct-attached storage into shared block storage that runs over an IP-based network.

In the high-performance computing environment, burst buffer is a fast intermediate storage layer positioned between the front-end computing processes and the back-end storage systems. It bridges the performance gap between the processing speed of the compute nodes and the Input/output (I/O) bandwidth of the storage systems. Burst buffers are often built from arrays of high-performance storage devices, such as NVRAM and SSD. It typically offers from one to two orders of magnitude higher I/O bandwidth than the back-end storage systems.

Compute Express Link (CXL) is an open standard for high-speed, high capacity central processing unit (CPU)-to-device and CPU-to-memory connections, designed for high performance data center computers. CXL is built on the serial PCI Express (PCIe) physical and electrical interface and includes PCIe-based block input/output protocol (CXL.io) and new cache-coherent protocols for accessing system memory (CXL.cache) and device memory (CXL.mem). The serial communication and pooling capabilities allows CXL memory to overcome performance and socket packaging limitations of common DIMM memory when implementing high storage capacities.

References

  1. ↑ "Parallel I/O" (PDF). Johns Hopkins University.
  2. ↑ "Introduction to Parallel I/O" (PDF). Oak Ridge National Laboratory.
  3. ↑ "Introduction: The Parallel I/O Stack" (PDF). Cornell University.
  4. ↑ "Introduction to Parallel I/O". The University of Texas at Austin.
  5. ↑ "Parallel I/O". Scientific Computing Department.
  6. ↑ "A Comprehensive Look at High Performance Parallel I/O". Berkeley Lab.
  7. ↑ http://calcul.math.cnrs.fr/Documents/Manifestations/CIRA2011/2011-01_haefele_parallel_IO-workshop_Lyon.pdf [ bare URL PDF ]
  8. ↑ https://www.olcf.ornl.gov/wp-content/uploads/2013/05/OLCF-Data-Intro-IO-Gerber-FINAL.pdf [ bare URL PDF ]
  9. ↑ "A Comprehensive Look at High Performance Parallel I/O".
  10. ↑ "Parallel I/O – Why, How, and Where to?". 2015-04-09.
  11. ↑ Teng Wang; Kevin Vasko; Zhuo Liu; Hui Chen; Weikuan Yu (2016). "Enhance parallel input/output with cross-bundle aggregation". The International Journal of High Performance Computing Applications. 30 (2): 241–256. doi:10.1177/1094342015618017. S2CID   12067366.
  12. ↑ Laghave, Nikhil; Sosonkina, Masha; Maris, Pieter; Vary, James P. (2009-05-25). "Benefits of Parallel I/O in Ab Initio Nuclear Physics Calculations". Computational Science – ICCS 2009. Lecture Notes in Computer Science. Vol. 5544. pp. 84–93. doi:10.1007/978-3-642-01970-8_9. ISBN   9783642019692. S2CID   28279330.


This page is based on this Wikipedia article
Text is available under the CC BY-SA 4.0 license; additional terms may apply.
Images, videos and audio are available under their respective licenses.