Distributed block storage: Difference between revisions

Revision as of 13:44, 24 March 2024

Latest revision as of 02:49, 9 May 2024

(4 intermediate revisions by 2 users not shown)

Line 1:

{{Short description|Computer data storage architecture}}

{{Orphan|date=December 2023}}

'''Distributed block storage''' is a [[computer data storage]] architecture that the data is stored in volumes (known as blocks:, a term dating back to [[IBM 7030 Stretch|Project Stretch]]<ref>

[[Gerrit Anne Blaauw|Blaauw]], Gerrit Anne; [[Frederick Phillips Brooks, Jr.|Brooks, Jr., Frederick Phillips]]; [[Werner Buchholz|Buchholz, Werner]] (1962), "[http://archive.computerhistory.org/resources/text/IBM/Stretch/pdfs/Buchholz_102636426.pdf 4: Natural Data Units]" (PDF), in [[Werner Buchholz|Buchholz, Werner]] (ed.), Planning a Computer System – Project Stretch, [[McGraw-Hill Book Company, Inc.|McGraw-Hill Book Company, Inc]]. / The Maple Press Company, York, PA., pp. 39–40, LCCN [https://lccn.loc.gov/61-10466 61-10466], [https://web.archive.org/web/20170403014651/http://archive.computerhistory.org/resources/text/IBM/Stretch/pdfs/Buchholz_102636426.pdf archived] (PDF) from the original on 2017-04-03, retrieved 2017-04-03, […]Quote: Terms"Block usedrefers hereto the number of words transmitted to describeor thefrom structurean imposedinput-output unit in response to a single input-output instruction. Block size is a structural property of an input-output unit; it may have been fixed by the machine design, inor additionleft to [[bit]],be arevaried listedby the belowprogram.

[[Byte]] denotes a group of bits used to encode a [[Character (computing)|character]], or the number of bits transmitted in parallel to and from input-output units. A term other than character is used here because a given character may be represented in different applications by more than one code, and different codes may use different numbers of bits (i.e., different byte sizes). In input-output transmission the grouping of bits may be completely arbitrary and have no relation to actual characters. (The term is coined from bite, but respelled to avoid accidental mutation to bit.)

A [[Word (computer architecture)|word]] consists of the number of data bits transmitted in parallel from or to memory in one memory cycle. [[Word size]] is thus defined as a structural property of the memory. (The term [[Catena (unit)|catena]] was coined for this purpose by the designers of the [[Groupe Bull|Bull]] [[Bull Gamma 60|GAMMA 60]] [fr] computer.)

Block refers to the number of words transmitted to or from an input-output unit in response to a single input-output instruction. Block size is a structural property of an input-output unit; it may have been fixed by the design or left to be varied by the program.

</ref>) across multiple physical servers, as opposed to other storage architectures like file systems which manages data as a file hierarchy, and object storage which manages data as objects. A common distributed block storage system is a [[Storage area network|Storage Area Network]] (SAN).

Line 25 ⟶ 22:

Storage blocks are generally accessed by [[iSCSI]], [[Fibre Channel]] or [[Fibre Channel over Ethernet]] (FCoE) protocols. Block storage provides high performance for mission-critical applications and can provide high I/O performance and low latency. it is commonly used in [[Storage area network|Storage Area Network]] environments in place of file storage.

In recent years, with the creation of [[NVM Express|NVMe]] standrard, [https://www.simplyblock.io/nvme-storage NVMe-oF] protocol has gained popularity, in particular driving higher adoption of NVMe/TCP (Non-Volatile Memory Express over TCP).

* NVMe/TCP defines the mapping of NVMe queues, NVMe-oF capsules, and data delivery over the IETF Transport Control Protocol (TCP). This innovative transport protocol offers several advantages:

* NVMe/TCP capitalizes on the inherent low latency of NVMe storage devices. By encapsulating NVMe commands and data within TCP datagrams, it ensures efficient communication between hosts and storage media.

* NVMe/TCP enables a larger number of queues and queue paths for data transport compared to traditional protocols like iSCSI. This scalability results in a significant increase in throughput and a reduction in latency.

* Unlike other NVMe over Fabrics (NVMe-oF) transports, such as NVMe/FC (Fibre Channel), NVMe/TCP runs over Ethernet. This makes it accessible and widely deployable across existing network infrastructures.

Emergence of NVMe-oF standard allows for creation of next-generation [[software-defined storage]] technologies, such as [https://www.simplyblock.io/ simplyblock] that has fully embraced NVMe/TCP as a fundamental component of their container storage system. Leveraging the NVMe/TCP transport, simplyblock provides a seamless and high-performance solution for [[Kubernetes]] environments.

==References==