IBM extends its deep storage capabilities to Microsoft Azure users

IBM’s storage group’s strategy seeks to support customers in their use of multiple data centers, whether on premises or in public clouds – and whether using IBM products or others. It wants to make it easier and more efficient for them; but to do so it has mastered a number of technical challenges that might stump all but the largest corporate data centers. Today it announces the addition of Microsoft Azure as a fully supported public cloud for its block storage and data protection offerings alongside a number of other introductions and enhancements. As before I was fully briefed ahead of today’s news by Scott Baker (VP of Storage Product Marketing) and others in IBM’s Storage Division.

IBM storage software for Azure

Technically there is no such thing as ‘the cloud’ – rather there are almost as many implementations as there are cloud suppliers, with major differences in the way they handle data and in how customers pay for them. Since IBM wants as many of its customers as possible to take advantage of its storage systems and software, it continues to do the heavy technical lifting to make it easier. Today it announced:

  • Spectrum Virtualize for Public Cloud on Microsoft Azure – the inclusion of Azure with other public clouds (AWS and IBM Cloud) covered by its leading storage hypervisor. The Figure above (based on IBM’s pre-briefing) shows how Spectrum Virtualize is deployed between on-premises and Azure; the key elements are the building of the software using Azure Virtual Machines and the iSCSI storage protocol – although Fibre Channel (FC) and NVMe networking can be used on the user’s site. IBM Storage Insights can be used across both environments for consistent management.
  • Safeguarded Copy on Microsoft Azure – IBM’s advanced security approach, which we covered in July, has been extended to include Microsoft’s cloud platform. Customers can now use Azure as a repository for remote mirror data replication, data validation. Surgical data extraction and cloud recovery.

Long before the emergence of cloud computing, IBM’s storage division tested and certified other vendors’ storage arrays for use with its SVC and Spectrum Virtualize hardware and software. Azure’s integration as a public cloud goes beyond the technical issues to how customers can buy IBM’s software. To deploy it, they

  • Buy it through the IBM web site using Passport Advantage, selecting Azure part numbers for monthly or perennial licenses, and
  • Deploy from the Azure marketplace using an Azure Resource Manager template to create a minimum configuration in only minutes.

In addition they can pay for it with unused license capacity for a number of IBM offerings including the Spectrum Storage Suite, the Virtual Storage Center, or Storage Suite for IBM Cloud Paks. It can also be added to new orders for its FlashSystem.

Higher capacity FlashCore and Turbonomic integration for AI

Beyond Azure IBM’ storage division has made a few other enhancements and introductions. In particular:

  • Enhanced cloud and container support for the Spectrum Protect Suite[1] – it has Red Hat certification for Spectrum Protect Plus, which also enables direct backup of OpenShift and Kubernetes storage data to S3-based clouds such as AWS, Azure Google Cloud and most others.
  • Spectrum Scale high-performance object ingest – it has enhanced the speed at which applications can ingest data from Spectrum Scale: native S3 objects are now accessible at speeds of up to 50GB/second per server node.
  • Higher capacity FlashCore Module for Elastic Storage Systems – it has added the higher capacity 38TB FlashCore module to its Elastic Storage System, giving maxima of 912TB/node, 18.2PB/rack and – a staggering – theoretical 8YB per cluster.
  • Turbonomic integration into FlashSystem monitoring – IBM completed its acquisition of Boston-based Turbonomic, an Application Resource Management (APM) and Network Performance Management (NPM) software provider, for around $1.5b in July 2021, complementing its acquisition of Instana (also an APM software company) in December last year. Both will help it provide better AIOps support for its customers in general; during 4Q21, Turbonomic users will be able to monitor IBM FlashSystem storage resources, model changes, and even automate some data movement.

These announcements shows that IBM continues to expand its support for users of highly-secured, geographically distributed, large data sets; most of which contain large amounts of unstructured (or partially structure) data – increasingly created by container-based applications. Its aim perhaps is to support a ‘single version of the truth’ in operational data, with fast recovery from protected snapshot backups when the need arises.

Differences between the top three public cloud suppliers


IBM launched its Cloud Services Initiative in 2008 – a year in which the total spend on cloud computing was $14 billion. The market has grown 25-fold since then (see my Figure above), validating IBM’s vision, investments and strategy over the years. There are a number of differences between thee three cloud leaders. In particular:

  • Amazon built infrastructure and platform services that are easy for customers to adopt without having to mess with backwards compatibility; it builds most of the hardware systems running in it data centers and has pioneered a number of new protocols such as S3.
  • Microsoft’s software and operating systems were already installed on almost every hardware suppliers’ systems. Its success as a cloud supplier was initially due to its SaaS offerings such as Office 360, but it has also built significant IaaS/PaaS business.
  • IBM Cloud is somewhere in the middle between the others; it has a significant software business of its own, but also the longest history of all in supplying systems to the largest data centers.

Believing in a hybrid multi-cloud future IBM is adjusting its storage offerings to run effectively on the other public cloud platforms, starting with the largest. Beyond the technical challenges of fitting its software to these (often very) different platforms, it will gain new customers in the process. In Azure’s case these will include those with – sometimes very – different profiles to its own.
[1] This packages multiple components of Spectrum Protect and Protect Plus in a single bundle

One Response to “IBM extends its deep storage capabilities to Microsoft Azure users”

Read below or add a comment...

Trackbacks

  1. […] enormously to incorporate and manage these devices (witness my regular assessments of IBM’s announcements). The storage systems hardware market was led by Dell EMC in the year to June, which held a share […]