Software-defined storage: Difference between revisions

Content deleted Content added
Citation bot (talk | contribs)
Alter: url. URLs might have been internationalized/anonymized. | You can use this bot yourself. Report bugs here. | Suggested by AManWithNoPlan | All pages linked from cached copy of User:AManWithNoPlan/sandbox3 | via #UCB_webform_linked 1466/2623
rm some namedropping
 
(27 intermediate revisions by 20 users not shown)
Line 1:
{{Short description|Type of computer data storage}}
'''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.<ref>{{Cite web |title= Definition: software-defined storage |author= Margaret Rouse |work= SearchSDN |publisher= Tech Target <!-- reliable? --> |url= http://searchsdn.techtarget.com/definition/software-defined-storage |accessdate= November 7, 2013 }}</ref> The software enabling a software-defined storage environment may also provide policy management for features such as [[data deduplication]], replication, [[thin provisioning]], snapshots, copy-on-write clones, tiering and backup.
 
Software-defined storage (SDS) hardware may or may not also have abstraction, pooling, or automation software of its own. When implemented as software only in conjunction with commodity servers with internal disks, it may suggest software such as a virtual or global [[file system]] or [[distributed block storage]]. If it is software layered over sophisticated large storage arrays, it suggests software such as [[storage virtualization]] or [[storage resource management]], categories of products that address separate and different problems. If the policy and management functions also include a form of [[artificial intelligence]] to automate protection and recovery, it can be considered as intelligent abstraction.<ref>{{Cite web |title=The foundation of clouds: Intelligent abstraction |author=Chris Poelker |date= March 12, 2014 |url= http://www.computerworld.com/article/2476040/cloud-computing/the-foundation-of-clouds--intelligent-abstraction.html}}</ref> Software-defined storage may be implemented via appliances over a traditional [[storage area network]] (SAN), or implemented as [[network-attached storage]] (NAS), or using [[Object storage|object-based storage]]. In March 2014 the [[Storage Networking Industry Association]] (SNIA) began a report on software-defined storage.<ref>{{cite news |url= http://snia.org/sites/default/files/SNIA%20Software%20Defined%20Storage%20White%20Paper-%20v1.0k-DRAFT.pdf/|title=Technical Whitepaper:Software Defined Storage |author= SNIA |date= March 2014 }}</ref>
 
==Software-defined storage industry==
Line 17 ⟶ 18:
{{Cite web |title= The Software-Defined Data Center |work= company web site |publisher= VMware |url= http://www.vmware.com/software-defined-datacenter/ |accessdate= November 7, 2013 }}
</ref>
Other smaller companies then adopted the term "software-defined storage", such as [[Coraid]]Cleversafe (now owned by Coraid founder's new company SouthSuite), [[Scality]] (founded in 2009), [[Inc.|Cleversafe]] (acquired by [[IBM]]), and [[OpenIO]].
 
Based on similar concepts as [[software-defined networking]] (SDN),<ref>{{Cite web |title= Definition: software-defined storage |author= Margaret Rouse |work= SearchSDN |publisher= Tech Target <!-- reliable? --> |url= http://searchsdn.techtarget.com/definition/software-defined-storage |accessdate= November 7, 2013 }}</ref>
interest in SDS rose after [[VMware]] acquired [[Nicira]] for over a billion dollars in 2012.
 
Data storage vendors used various definitions for software-defined storage depending on their product-line. [[Storage Networking Industry Association]] (SNIA), a standards group, attempted a multi-vendor, negotiated definition with examples.<ref>{{cite web | url=http://www.snia.org/sds | title=Technology Focus Areas &#124; SNIA }}</ref>
 
The software-defined storage industry is projected to reach $86 billion by 2023.<ref>{{Cite web|url=https://www.fiercetelecom.com/telecom/flourishing-software-defined-storage-market-will-ramp-up-to-86b-by-2023-report|title=Thriving software-defined-storage market will ramp up to $86B by 2023: report|website=FierceTelecom|date=20 March 2020 |language=en|access-date=2020-03-23}}</ref>
 
Building on the concept of VMware, esurfing cloud has launched a new software-defined storage product called HBlock. HBlock is a lightweight storage cluster controller that operates in user mode. It can be installed on any Linux operating system as a regular application without root access, and deployed alongside other applications on the server. HBlock integrates unused disk space across various servers to create high-performance and highly available virtual disks. These virtual disks can be mounted to local or other remote servers using the standard iSCSI protocol, revitalizing storage resources on-site without impacting existing operations or requiring additional hardware purchases.<ref>{{Cite web |title=第二期观点|天翼云存储资源盘活系统 HBlock,全面释放企业数据价值_云计算_InfoQ精选文章 |url=https://www.infoq.cn/article/HOTjbXqI4WmFSuhVjUDa |access-date=2024-04-16 |website=www.infoq.cn}}</ref>
 
==Characteristics==
Characteristics of software-defined storage may include the following features:<ref name="hype">{{cite web |title=Software-defined storage: The reality beneath the hype |date= March 12, 2013 |author= Simon Robinson |work= Computer Weekly |url= http://www.computerweekly.com/opinion/Software-defined-storage-The-reality-beneath-the-hype |accessdate= November 7, 2013 }}</ref>
* Abstraction of logical storage services and capabilities from the underlying physical storage systems, and in some cases pooling across multiple different implementations. Since data movement is relatively expensive and slow compared to computation and services , pooling approaches sometimes suggest leaving it in place and creating a mapping layer to it that spans arrays. Examples include:
** [[Storage virtualization]], the generalized category of approaches and historic products. External-controller based arrays include storage virtualization to manage usage and access across the drives within their own pools. Other products exist independently to manage across arrays and/or server DAS storage.
** Virtual volumes (VVols), a proposal from [[VMware]] for a more transparent mapping between large volumes and the VM disk images within them, to allow better performance and data management optimizations. This does not reflect a new capability for virtual infrastructure administrators (who can already use, for example, NFS) but it does offer arrays using [[iSCSI]] or [[Fibre Channel]] a path to higher admin leverage for cross-array management apps written to the virtual infrastructure.
** Parallel [[Network File System|NFS]] (pNFS), a specific implementation which evolved within the NFS community but has expanded to many implementations.
** [[OpenStack]] and its Swift, [[Ceph (software)|Ceph]] and Cinder APIs for storage interaction, which have been applied{{by whom|date=August 2014}} to open-source projects as well as to vendor products.
** A number of Object Storage platforms are also examples of software-defined storage implementations examples of this are [[Scality]] RING and the open source swift project.
** Number of distributed storage solutions likefor [[Glusterclustered file system]] or [[distributed block storage]] are good examples of software defined storage.
* Automation with policy-driven storage provisioning with [[service-level agreement]]s replacing technology details. This requires management interfaces that span traditional storage-array products, as a particular definition of separating "control plane" from "data plane", in the spirit of [[OpenFlow]]. Prior industry standardization efforts included the [[Storage Management Initiative – Specification]] (SMI-S) which began in 2000.
* Commodity hardware with storage logic abstracted into a software layer. This is also described{{by whom|date=August 2014}} as a [[clustered file system]] for [[converged storage]].
Line 41 ⟶ 44:
In [[computing]], a '''storage hypervisor''' is a software program which can run on a physical server hardware platform, on a [[virtual machine]], inside a hypervisor OS or in the storage network. It may co-reside with virtual machine [[Supervisory program|supervisors]] or have exclusive control of its platform. Similar to virtual server [[hypervisor]]s a storage hypervisor may run on a specific hardware platform, a specific hardware architecture, or be hardware independent.<ref>{{cite web|title=Comparison of virtualization technologies|url=http://virt.kernelnewbies.org/TechComparison}}</ref>
 
The storage hypervisor software virtualizes the individual storage resources it controls and creates one or more flexible pools of storage capacity. In this way it separates the direct link between physical and logical resources in parallel to virtual server hypervisors. By moving storage management into isolated layer it also helps to increase system uptime and [[High Availability]]. "Similarly, a storage hypervisor can be used to manage virtualized storage resources to increase utilization rates of disk while maintaining high reliability."<ref>{{cite documentjournal |title=Evaluation and design of highly reliable and highly utilized cloud computing systems, Page |page=12 |publisherjournal=Journal of Cloud Computing | date=June 9, 2014 | authorfirst1= Brett |last1=Snyder, |first2=Jordan |last2=Ringenberg, |first3=Robert GreenEmail author,|last3=Green |first4=Vijay |last4=Devabhaktuni and |first5=Mansoor |last5=Alam |volume=4 |doi=10.1186/s13677-015-0036-6|s2cid=17909593|doi-access=free }}</ref>
The storage hypervisor, a centrally-managed supervisory software program, provides a comprehensive set of storage control and monitoring functions that operate as a transparent virtual layer across consolidated disk pools to improve their [[high availability|availability]], speed and utilization.
Line 53 ⟶ 56:
The storage hypervisor manages, virtualizes and controls all storage resources, allocating and providing the needed attributes (performance, availability) and services (automated [[Thin provisioning|provisioning]], [[Snapshot (computer storage)|snapshots]], [[Replication (computer science)|replication]]), either directly or over a storage network, as required to serve the needs of each individual environment.
The term "hypervisor" within "storage hypervisor" is so named because it goes beyond a supervisor,<ref>{{cite web|title=Hypervisor glossary definition|url=http://www.xen.org/files/xen_user_manual.pdf|work=Xen v2.0 for x86 Users' Manual (PDF)|publisher=Xen.org on August 20, 2011|access-date=October 4, 2017|archive-date=October 5, 2011|archive-url=https://web.archive.org/web/20111005054849/http://www.xen.org/files/xen_user_manual.pdf|url-status=dead}}</ref> it is conceptually a level higher than a supervisor and therefore acts as the next higher level of management and intelligence that sits above and spans its control over device-level storage controllers, disk arrays, and virtualization middleware.
 
A storage hypervisor has also been defined as a higher level of storage virtualization <ref>{{cite web|title=SearchStorage.com definition|url=http://searchstorage.techtarget.com/definition/storage-virtualization|publisher=What is storage virtualization? Definition on SearchStorage.com}}</ref> software, providing a "Consolidation and cost: Storage pooling increases utilization and decreases costs. Business availability: Data mobility of virtual volumes can improve availability. Application support: Tiered storage optimization aligns storage costs with required application service levels".<ref>{{cite book|title=IBM SmartCloud Virtual Storage Center|date = 6 March 2015|url=https://books.google.com/books?id=_pbABgAAQBAJ&pg=PA28|publisher=IBM Redbooks|isbn = 9780738440439}}</ref> The term has also been used in reference to use cases including its reference to its role with storage virtualization in disaster recovery <ref>{{cite web|title=SearchDisasterRecovery Article|url=http://searchdisasterrecovery.techtarget.com/news/2240037212/The-cloud-and-virtualization-havent-make-IT-disaster-recovery-plans-obsolete|publisherwebsite=Published in SearchDisasterRecovery.com on |date=June 23, 2011|first=Todd|last=Erickson|access-date=October and4, written2017|archive-date=October by4, Todd Erickson2017|archive-url=https://web.archive.org/web/20171004085129/http://searchdisasterrecovery.techtarget.com/news/2240037212/The-cloud-and-virtualization-havent-make-IT-disaster-recovery-plans-obsolete|url-status=dead}}</ref> and, in a more limited way, defined as a volume migration capability across SANs.<ref>{{cite web |title=ComputerWorld Article |url=http://www.computerworld.com.au/article/368886/compellent_adds_virtualization_hardware_upgrades_its_san/ |publisherdate=Published on November 23, 2010 and|first=Lucas written|last=Mearian by|access-date=October Lucas4, Mearian2017 |archive-date=October 4, 2017 |archive-url=https://web.archive.org/web/20171004140023/https://www.computerworld.com.au/article/368886/compellent_adds_virtualization_hardware_upgrades_its_san/ |url-status=dead }}</ref>
}}</ref>
 
===Server vs. storage hypervisor===
An analogy can be drawn between the concept of a server hypervisor and the concept of a storage hypervisor. By virtualizing servers, server hypervisors ([[VMware ESX]], [[Microsoft Hyper-V]], Citrix Hypervisor, [[Linux KVM]], [[Xen]], [[z/VM]]) increased the utilization rates for server resources, and provided management flexibility by de-coupling servers from hardware. This led to cost savings in server infrastructure since fewer physical servers were needed to handle the same workload, and provided flexibility in administrative operations like backup, failover and disaster recovery.
 
A storage hypervisor does for storage resources what the server hypervisor did for server resources. A storage hypervisor changes how the server hypervisor handles storage I/O to get more performance out of existing storage resources, and increases efficiency in storage capacity consumption, storage provisioning and snapshot/clone technology. A storage hypervisor, like a server hypervisor, increases performance and management flexibility for improved resource utilization.
Line 74 ⟶ 76:
 
[[Category:Cloud storage]]
[[Category:Emerging technologies]]
[[Category:Information technology]]