File and Object Storage

 View Only

IBM Storage Ceph 7.1 - A new milestone, offering block storage through NVMe/TCP for non-Linux clients

By Marcel Hergaarden posted Tue June 11, 2024 07:00 AM

  

IBM announces IBM Storage Ceph 7.1, enabling VMware users to access Ceph block storage through NVMe/TCP.


IBM announces IBM Storage Ceph version 7.1, planned general availability on June 14, 2024
This new Storage Ceph 7.1 release enables IBM clients with an ability to address new use cases with Storage Ceph, supported by IBM.
This blogpost contains a summary of new features, functions and enhancements made available with IBM Storage Ceph version 7.1.

About IBM Storage Ceph
IBM Storage Ceph is IBM’s for enterprise usage supported distribution of the open-source Ceph platform that provides massively scalable object, block, and file storage from a single solution.

IBM Storage Ceph is an Open Source-Software Defined Storage (SDS) solution which provides enterprise level client support, delivered by IBM.
Storage Ceph is software-defined storage and part of IBM’s storage portfolio since January of 2023. IBM Storage Ceph runs on industry-standard x86 server hardware and can start in a small minimum setup of four nodes and then scale-out to Petabytes scale, according to business needs.

IBM Storage Ready Nodes for Ceph
IBM optionally offers IBM Storage Ready Nodes for IBM Storage Ceph clients.
Ready Nodes concern server hardware, which has been fully tested and qualified for IBM Storage Ceph implementation with ease.
Storage Ready Nodes for Ceph enable IBM clients for a complete storage software- and hardware solution experience.

IBM Storage Ceph could of course also be installed on industry standard server hardware though IBM Storage Ready Nodes are certified for installation of IBM Storage Ceph, to ensure a quick and easy installation process without hardware specific configuration guesswork upfront.

Storage Ready node feature- and capacity choices are available with multiple configuration choices and options, depending on a client’s use cases and use patterns. On-site implementations are feasible by IBM expert labs consulting services.


New in IBM Storage Ceph 7.1


Fast network-based Block storage for VMware
IBM Storage Ceph 7.1 offers NVMe/TCP access to Ceph (RBD) block storage.
This functionality is now general available.

NVMe/TCP enables access to Ceph RBD block storage with non-Linux clients.
This functionality enables VMware vSphere to connect with IBM Storage Ceph and use Ceph block storage as VMware VMFS storage resourcing.

The methodology looks similar in scope and design to Ceph RBD over iSCSI in the past.
However, this is a complete new management layer via a Ceph-NVMe/TCP daemon.
This deamon coordinates configuration of NVMe/TCP targets, across multiple cluster nodes, including fail-over capabilities.

The NVMe/TCP functionality has no Linux kernel dependencies and is based on SPDK code for a solid performance. This results in a similar performance outcome as with native Ceph RBD block access.

The functionality enables VMware clients to decouple their VM storage from ESX compute nodes, which implies that storage resourcing can now independently scale from ESX compute resources. That can be helpful to save on license cost for VMware nodes and to obtain more storage flexibility within infrastructure environment layouts. A vSphere plugin is available for IBM Storage Ceph as well. 
A concise technical video demo of the NVMe/TCP capability combined with VMware is available here.

NFS support for CephFS
CephFS filesystem access for non-native Ceph clients.
This functionality has now general availability status, with IBM Storage Ceph 7.1

Clients can now create, edit, and delete NFS exports from within the Ceph dashboard after configuring the Ceph filesystem. 
CephFS namespaces can be exported over NFS protocol, using NFS v4.1 or NFS v3 service.

IBM Storage Ceph Linux clients can natively mount CephFS volumes, while the driver for CephFS is default integrated in the Linux kernel. Non-Linux clients can now also access CephFS volumes, by using the NFS v4.1 including failover functionality or make use of the NFS v3 protocol.
In practice this allows non-Linux clients to access CephFS resources through the NFS protocols.

Ceph Object storage, RADOS Gateway (RGW) updates
New supported S3-compatible platforms, for RGW object data transitioning, of to public cloud.
This functionality was already existing for Amazon AWS S3 and Microsoft Azure*.
With IBM Storage Ceph 7.1, the transitioning functionality has been extended to IBM Storage COS and IBM Storage Ceph clusters, as new candidate target locations.

* For Azure, multi-cloud gateway (MCG) is additionally required, to translate S3 protocol object data into Azure Blob.

Intel QAT acceleration, GA supported
Intel QAT, Quick Assist Technology
, provides extended acceleration for IBM Storage Ceph internal checksums, encryption and compression services.

By offloading those typical types of CPU requests to hardware, the actual CPU performance improves. More efficiency in terms of cost and power, compared to general-purpose CPUs, with those types of compute-intensive workloads. Object encryption and compression can now be enabled without additional cost to regular CPU workloads, while using Intel QAT.


Multi-site object replication
Multi-site object replication is now selectable with bucket-level granularity.

This provides IBM clients with an ability to replicate a selected bucket or group of buckets, to a different IBM Storage Ceph cluster(s) on secondary location(s). Active-active replication can be established between sites, with this new bucket level granularity and also bi-directional replication of selected buckets.

This functionality could be helpful with (but not limited to) use cases like remote offices, branch offices, of edge locations where only the selected data is relevant for replication, rather than entire namespace replications.
New, highly flexible sync policies can be built with IBM Storage Ceph 7.1

Object access with NFS, fronting RADOS Gateway backend
This functionality is general available in IBM Storage Ceph version 7.1
NFS with RGW back-end integration provides an easy way to ingest existing enterprise data from Linux and Windows clients into a Ceph object store via dropping file on an easy to access NFS fileshare.

It enables regular IT users on other platforms to ingest or access result data on an IBM Storage Ceph Object store. Object data can be thus made available to applications or platforms which do not have capabilities for S3 object access natively. In example, older legacy applications or industrial hardware equipment which does not yet support S3 object communication
.

Object archive zone - general availability

The object archive zone can store immutable versions of selected, business relevant buckets and objects. This can become helpful functionality in a case of ransomware or eventual virus attacks. Immutable objects can be copied back to production easily. An archive zone can be configured for selected object buckets, with a bucket level granularity.


New IBM Storage Ceph Dashboard UI functionalities
New capabilities for IBM Storage Ceph related interactions and administrative insights.

Cephadm RADOS Gateway deployment
Full bootstrapping of the RGW service for both single-site and multi-site deployments.

About Cephadm
The Cephadm utility deploys and manages a Ceph storage cluster. Tightly integrated with both the command-line interface (CLI) and the IBM Storage Ceph Dashboard web interface. Cephadm can manage IBM Storage Ceph clusters from within either environment.
With the release of IBM Storage Ceph 7.1, Cephadm has a capability to execute complete bootstrapping for IBM Storage Ceph RGW service. It could now instantly deploy RGW into a production-ready state for immediate client usability.


New CephFS NFS monitoring and management options
Create Exports/Shares rate-limits, in example: set/list/edit/delete
Monitor alarms for NFS services status and rate-limiting.
Including monitoring for utilization, capacity, IO-usage , average read&writes in MB/sec, connected clients and IP addresses.


RGW Object level UI interactions
RADOS Gateway (RGW) UI Labeled performance counters.
A capability to retrieve information on the operations happening per RGW.
The functionality offers an ability to get RGW performance counters with a per user, or per bucket granularity. Including the total number of operations and numbers of operations and different users, accessing a bucket.

UI based volume and snapshot management for CephFS
Ability to manage the entire lifecycle of CephFS volumes, subvolumes and snapshots all through the Ceph Dashboard UI.
Monitor health status, current and historical capacity utilization for CephFS volumes and subvolumes.

Manage filesystem snapshots via CephFS snapshot scheduler.
Create, schedule, delete or list snapshots and remote mirroring management tasks.
Creation, setting options, and deletion of filesystem volumes, subvolumes and user access management and see where a particular filesystem or volume is consumed by layered services such as NFS Ganesha.


IBM Storage Ceph hardware monitoring
This functionality enables IBM Storage Ceph to provide an inventory of the cluster server hardware, identify hardware issues and perform specific actions.

Hardware status healthy/unhealthy, hardware type/model
Firmware version and/or revision
High-level hardware status insight is available on the IBM Storage Ceph UI dashboard landing page.

A few example actions on IBM Storage Ready nodes:

·     Light the chassis and/or HDD LEDs
Perform firmware upgrades
Manage power on|off, reboot nodes.

Please note that this functionality is mainly implemented for IBM Storage Ready nodes, but may or may not be fully functional on other industry-standard hardware yet.

Technology preview

The below mentioned functionalities/features are currently introduced in Technology Preview state and can become general available in future releases of IBM Storage Ceph.

Technology Preview features are not supported with IBM production service level agreements (SLAs), might not be functionally complete, and IBM does not recommend using them for production. These features provide early access to upcoming product features, enabling customers to test functionality and provide feedback during the development process.


Tech Preview
:
Ceph volume management, performance and health monitoring with VMware vCenter
This functionality enables IBM Storage Ceph volume management through the vCenter for vSphere admins. It
covers all life cycle operations of the Ceph block volumes as well as health and performance monitoring. Available for VMware 7 update 3 and higher.

Summary
IBM has announced IBM Storage Ceph 7.1 GA, with new features and functionalities.
The main significant highlights of this 7.1 release are:

·     Fully supported unified storage release
IBM Storage Ceph 7.1 supports File, Block and Object storage.
This enables IBM clients for new use cases, supported by IBM.

·     Fast network-based Block storage for VMware
NVMe/TCP to access Ceph RBD block storage for non-Linux clients.
With native RBD performance, reliable and proven technology for over a decade.

·     NFS access to CephFS volumes and subvolumes
Export CephFS volumes via NFS v4.1 or v3
For servers and applications which cannot natively access CephFS in the way a native Linux kernel driver does.

·     RADOS Gateway (RGW) updated features
Archive Zone - general availability
Bucket granular sync replication and other functional enhancements

·     New dashboard user interface management capabilities
Significant UX improvements for more ease of use.
In example, snapshot management, log filtering and viewing.
Ability to upgrade IBM Storage Ceph clusters via the UI dashboard.

IBM Storage Ceph resources
find out more about IBM Storage Ceph

IBM Storage Ceph website:  https://www.ibm.com/products/ceph  
IBM Storage Ceph documentation:  http://docs.ceph.blue  
IBM Storage Ceph video demos: http://easy.ceph.blue


IBM Redpaper publications

-       Concepts and architecture of IBM Storage Ceph:
   https://www.redbooks.ibm.com/abstracts/redp5721.html 

-       Step-by-step implementation of some of the common scenarios with IBM Storage Ceph:
   https://www.redbooks.ibm.com/abstracts/redp5715.html


Why IBM?

Data matters. 
When planning a data strategy for new or existing applications it’s easy to focus on compute resources and applications without proper planning for the data that will drive the results for the applications. 

Our products are all about solving hard problems faster with data. IBM helps customers achieve business value with a clear data strategy. 
Our strategy is simple, unlock data to speed innovation, de-risk data to bring business resilience and help customers adopt value based data to bring cost and energy efficiencies.

Value needs to be delivered by connecting organizational data sources with business drivers to create business value that matters to the organization. Many organizations focus on a single driver with a storage solution, but the best solution is driven by an infrastructure strategy than can accomplish most if not all drivers for maximum benefits.

Our story is not just about another storage product, but is about innovation and a comprehensive storage portfolio that is helping businesses drive more value throughout the organization.

Contact IBM
https://www.ibm.com/contact 



#Featured-area-2-home


#Highlights
#Highlights-home
#ibmtechxchange-ai
0 comments
36 views

Permalink