Tag Archives: Storage

Random Short Take #109

Posted on by

Random short take #109. Looking for news that’s old? You’ve come to the right place. Let’s get random.

  • This opinion piece would be hilarious, if it wasn’t so alarmingly accurate. We weren’t allowed to trust Wikipedia articles but whatever AI tells us is just fine? Let’s pump the brakes and reflect, please.
  • And speaking of art and artifice, this post from John Birmingham on digital media versus holding the art in your hand was great.
  • Need to store that digital media somewhere? Backblaze knows a bit about that. Read the Q2 2026 Drive Stats report here.
  • Speaking of storage, this article over at El Reg on the Domesday Project was pretty cool.
  • And while we’re talking long-term storage, in press release news, check out what Cerabyte has been up to.
  • In people news, Preston de Guise is looking for a job. Hire him if you need someone smart who can really help your organisation.
  • I’m not looking to return to the world of home lab builds, but if you’re looking for a primer, this article covers off some of the basics really well.
  • And this is just … ugh.

VMware Cloud on AWS – TMCHAM – Part 16 – Remove External Datastores

Posted on by

In this edition of Things My Customers Have Asked Me (TMCHAM), I’m going to do a (really) short post on removing external datastores on the VMware-managed VMware Cloud on AWS platform. It might seem like a simple thing, but it can be confusing for customers who haven’t done it before.

 

The Problem

If you’ve used an external datastore (either FSx for NetApp ONTAP or the artist formerly known as VMware Cloud Flex Storage) with your SDDC, you’ll know that management of the connection of the datastores to your SDDC is done via the Cloud Console, and not via vCenter. There might come a time when you want to remove that storage, either because you’ve bought more nodes and want to use vSAN, or you’re possibly shrinking your footprint. Whatever the reason, if you do try to remove the datastore via vCenter (assuming you’ve removed your workloads from the datastore), you’ll get an error about insufficient privileges. At first it might seem like it’s because you have VMs still present on the datastore (and if this is the case, contact Global Support to help you if you can’t find them). It’s more likely, however, that you’re not in the right place to do this activity.

 

The Solution

The solution is simple. In the VMware Cloud Console, go to the Storage tab of the SDDC you want to remove the datastore from, and click on Detach Datastore. And that’s it. It won’t work if you’ve got workloads running there, so make sure you’ve cleaned everything up / moved it somewhere else before you attempt this one.

Random Short Take #106

Posted on by

Random short take #106. Looking for old news? I’ve got just the thing. Let’s get random.

Random Short Take #105

Posted on by

Random short take #105. Want news? Sorry I’ve had a bit on. Let’s get random.

Random Short Take #104

Posted on by

Random short take #104. Want news? This is where news goes to die. Let’s get random.

Nexsan Announces Unity NV4000

Posted on by

Nexsan recently announced the Nexsan Unity NV4000. I had the chance to speak to Vincent Phillips (CEO), Judy Kaldenberg (SVP Sales & Marketing), and Mircea Turcu (VP of Engineering) about it, and thought I’d share some thoughts here.

 

What Is It?

[image courtesy of Nexsan]

I’ve said it before (a little over a year ago as it happens), and I’ll say it again … in the immortal words of Silicon Valley: “It’s a box“. And a reasonably powerful one at that, coming loaded with the following specifications.

Supported Protocols SAN (Fibre Channel, iSCSI), NAS (NFS, SMB 1.0 to 3.0, FTP), Object (S3)
Disk Bays | Rack U 24 | 4RU
Supported Expansion Systems N/A
Maximum Raw Capacity 576 TB Raw
System Memory (DRAM) per controller up to 128GB
FASTier 2.5″ SSD Drives (TB) 1.92 | 3.84 | 7.68 | 15.36
SAS SSDs (TB) 1.92 | 3.84 | 7.68 | 15.36 | 30.72
SAS HDDs (TB) 4 | 6 | 8 | 10 | 12 | 14 | 16| 18 | 20 | 22 | 24
Host Connectivity 16/32Gb FC | 10/25/40/100 GbE

It’s a dual-controller platform, with dual Intel Xeon Silver 4309Y CPUs (2.8 GHz CPU, 8 cores, 16 threads) per controller, and you get access to the following features included as part of the platform license:

  • Nexsan’s FASTier® Caching – Use solid-state to accelerate the performance of the underlying spinning disks
  • Nexsan Unity software version 7.2, with important enhancements to power, enterprise-class security, compliance, and ransomware protection
  • Enhanced Performance – Up to 1,000,000 IOPs
  • Third-Party Software Support – Windows VSS, VMware, VAAI, Commvault, Veeam Ready Repository, and more
  • Multi-Protocol Support – SAN (Fibre Channel, iSCSI), NAS (NFS, CIF, SMB1 to SMB3, FTP), Object (S3), 16/32GB FC, 10/25/40/100 GbE
  • High Availability – No single point-of-failure architecture with dual redundant storage controllers, redundant power supplies and RAID

 

Other Features (via Unity 7.2)

Along with the standard Unity features, with the release of Unity 7.2, you also get access the new Cloud Connector, offering bi-directional sync for files / directories with cloud storage solutions such as AWS S3, Azure Blob, and Unity Object store. As well as this, an SMB Access Audit capability has been introduced with the new version of Unity, given administrators the ability to view file access by user via the SMB protocol.

 

Use Cases

The Unity NV series of arrays is well suited to the following workloads:

  • Media and entertainment;
  • Hypervisor storage (primary and secondary);
  • Digital video surveillance; and
  • Backup and recovery storage.

 

Thoughts and Further Reading

I wrote about the NV6000 when it was announced in January last year. At the time I wondered, somewhat tongue in cheek, who really cared about storage array announcements in this age of hyperscaler storage solutions. But I think plenty of people should and do care that these kinds of products are out there in the market. As with the other Unity solutions, the NV4000 provides users some degree of flexibility when it comes to the protocols they want to use (SMB, NFS, iSCSI) and the transport mechanisms (FC, Ethernet). For every pundit pronouncing the death of fibre channel, there are any number of folks out in the field still running those networks from a bygone era, and quite happily to boot. The smaller footprint of the NV4000 may not be that exciting for those of us craving large scale storage capacities and high performance arrays to do the heavy lifting for our critical workloads, but these kinds of solutions are great at the edge or in places where you can’t simply roll in a full rack of storage. As well as this, in a society increasingly fascinated with surveillance, solutions like the NV4000 provide a robust solution for the storage of CCTV footage, and Nexsan also seems very focussed on solving this problem efficiently and cost-effectively.

Hybrid storage is still a real thing in the wild, because people are still looking for ways to store more and more data in a cost effective manner, and providing support for spinning disks in storage arrays, with some clever caching / storage tiering up front, is one way of doing that. Particularly when that storage needs to happen at the edge, or in places where clouds can’t be easily reached. While it seems like an odd concept for people working with data in places where big fibre runs exist to their hyperscaler of choice, there are plenty of places (like most of Australia) where moving large volumes of data between the cloud and local workloads still isn’t a terribly fast (or cheap) activity. Until that problem is somehow solved (and I know it’s being worked on by many companies) midrange, on-premises storage isn’t dead yet.

StorPool Announces Disaster Recovery Engine For KVM-based Clouds

Posted on by

StorPool recently announced its Disaster Recovery Engine (DRE) for KVM-based clouds. I had the chance to talk to Dave Henry and Beth Caltagirone about the announcement and thought I’d share some thoughts here.

 

What It Does

According to StorPool, the DRE “can protect environments with tens to thousands of VMs. It simplifies disaster recovery scenarios like 1:1, many to 1, many to many, and delivers industry-leading recovery point objective (RPO) and recovery time objective (RTO) of virtual machines in disaster scenarios, to ensure uninterrupted business operations.” At its core, it’s a way to get data from point A to point B, and use that data in the event that point A goes away.

[image courtesy of StorPool]

It supports asynchronous replication down to a 15-minute recovery point objective, and, according to Henry, can deliver some pretty snappy recovery times (depending on the hardware deployed at the recovery site).

Supported Topologies

There are several topologies supported, including:

  • Active – Passive;
  • Bi-directional;
  • Many-to-one; and
  • Multi-site mesh.

The choice of topology will be dependent on various environmental factors, such as distance between sites, the type of disaster you’re looking to recover from, etc. But I’m sure this level of flexibility would be appreciated by StorPool customers.

Licensing

Existing and new customers will be pleased to hear that the StorPool Disaster Recovery Engine is fully integrated with StorPool’s existing licensing – no additional license is required.

 

Who’s Using It?

“StorPool customers are IT service providers building public, private and hybrid clouds – Managed Service Providers, Hosting Service Providers, Cloud Service Providers, Enterprises and SaaS vendors. The StorPool Storage platform is a Storage-as-a-Service (STaaS) offering, with a bring your own server model.” CloudSigma has been using the product to offer DR capabilities to its customers across 31 sites globally.

 

Thoughts

In terms of hypervisors, I’ve been working almost exclusively with VMware for the last twenty years (well, except for that weird XenServer dalliance in the early 2000s), so my knowledge of KVM-based solutions, particularly when it comes to disaster recovery and orchestration, is fairly limited. That said, I’ve been following StorPool for some time, and have seen the offering develop over time to meet the needs of service providers looking to do cool stuff with software-defined storage solutions. The ability to provide some useful DR replication and orchestration seems to be a logical progression for a company that helps service providers deliver storage and cloud services to its customers, and that slo offers its own managed services as part of the arrangement. The company has always been focussed on an API-first approach, and this carries over to the DRE offering as well. I’m always happy to hear about companies working to deliver more DR capabilities to their products, particularly when there’s been a bit of a gap in the capability previously. And I’m especially happy to hear that this capability is being included at no additional cost.

The StorPool DRE is now in public beta and GA is scheduled for Q2 2025. I’m keen to hear how the solution is being adopted with StorPool customers and looking forward to hearing more about how StorPool are innovating in the market.

NetApp Cloud Storage Announcements – February 2025

Posted on by

 

NetApp recently made some announcements around its cloud storage capabilities as part of a larger series of product announcements. I thought I’d share some notes and some brief thoughts here.

 

Azure NetApp Files

New capability – Cost optimisation

New capability – Resilience

There’s also preview support for backing up large volumes and performing single file restores from backup.

 

Amazon FSx for NetApp ONTAP

There’s a new AWS CloudFormation package Amazon FSx for NetApp ONTAP that unlocks advanced ONTAP operations, including:

  • SnapMirror creation (including ClusterPeer and SVM Peer);
  • iSCSI (LUN and iGroup) creation;
  • CIFS Share creation;
  • Snapshot and Snapshot Policy creation; and
  • Advanced Volume creation.

[image courtesy of NetApp]

 

Thoughts

Why am I writing about FSx for NetApp ONTAP on my weblog? And why should you be interested in a few minor enhancements to this and to Azure NetApp Files? How do these technologies even make sense when the benefits of public cloud are right there? Why would organisations not just use the hyperscaler-native storage solutions available to them? It’s for the same reason that numerous organisations have adopted VMware Cloud on AWS and Azure VMware Solution, sometimes your applications need something different to what the public cloud can offer, at least until those applications have undergone some kind of transformation. I have plenty of customers in my day job taking advantage of FSx for NetApp ONTAP with their VMware Cloud on AWS deployments because the linear scalability of hyper-converged infrastructure has proven problematic for storage-heavy workloads. Additionally, they get the benefit of being able to leverage the NetApp tools they know and love, just in a cloud context. It’s great to see NetApp continue to add features and functionality to both Azure NetApp Files and FSx for NetApp ONTAP. If you’re struggling with shifting storage-dense workloads to public cloud while needing to stay within the confines of a VMware-based solution, these offerings can really help the economics make sense, even for enterprises, not just the t-shirt and jeans cloud-native types.

NetApp Enterprise Storage Announcements – February 2025

Posted on by

NetApp recently made some announcements about enhancements to its enterprise storage line of products. I thought I’d share a few notes here.

 

ASA A-Series

If you’re unfamiliar with the ASA line of devices, it’s NetApp’s all-flash SAN solution. There are a few models already available (see the data sheet here) and now there are three more A-series models: the ASA A50, ASA A30, and ASA A20.

[image courtesy of NetApp]

Major Hardware Innovations

There’s an integrated compression offload engine with Intel® QAT®1 (this is not available with the ASA A20), delivering the following benefits:

  • Next generation CPU architecture provides increased performance for resource-intensive tasks and processing
  • Always-on enhanced storage efficiency without impacting performance
  • Up to 160Gb/s data compression

It takes advantage of PCIe Gen 5, as well as DDR 5 Memory. According to NetApp, this delivers twice the clock speed for faster memory access and 50% more throughput for large data transfers. Additionally, there’s support for 64Gb Fibre Channel and NVMe/FC, as well as 100GbE and NVMe/TCP support. They use Titanium power supplies with a 95% efficiency rating.

Storage efficiency

You also get access to “always on, inline, data reduction”. There’s no cold data policy to manage, nothing for customers to configure and, according to NetApp, it’s easy to meet Storage Efficiency Guarantee ratios. Thanks to QAT, you’ll never need to choose between performance or efficiency.

 

New FAS and EF-Series systems

FAS50

The FAS50 delivers quite a punch in a small 2U form factor, with:

  • Optimised I/O flexibility;
  • Improved price/performance value;
  • Enabled for future security hardware offload; and
  • Released with ONTAP 9.16.1 GA.

[image courtesy of NetApp]

EF Series

NetApp has also announced the availability of new EF-Series arrays – the EF300C and EF600C.

[image courtesy of NetApp]

A few things to note with the EF300C and EF600C:

  • The default configuration is a single Dynamic Disk Pool (DDP) which is automatically created using all available drives for simplicity in management
  • There is support for 30TB or 60TB capacities (but by default there is no mixing of capacities in a shelf)
  • There is only support for QLC NVMe SSDs (No mixing of QLC or TLC NVMe SSDs allowed)
  • Available with a minimum of 12x up to a maximum of 24x QLC NVMe SSDs (No expansion shelf capability)

 

Thoughts

It’s been a minute since I wrote about a storage company’s product announcement. The news from NetApp piqued my interest, however, as it seems like the portfolio of products and options has developed into something broader than what NetApp was traditionally known for (and I know there’s even more news that was announced that I haven’t covered here). Whilst I talk a lot about how useful the cloud is in my day job, it’s hard to dispute that I/O (and storage in general) can still have a massive impact on the performance and usability of critical and not so critical applications alike. Additionally, having options in terms of scalability, performance, and transport mechanisms, is key to delivering the right storage platforms to the applications (and end users) that need them. But while I’ve dumped a lot of speeds and feeds into this article, ultimately the success of a storage company like NetApp is more about how the platforms can be managed, integrated with other systems, and upgraded from one version to the next.

If you’re running everything in one of the hyperscalers (or more than one, for that matter). you’re not going to be terribly interested in what NetApp is selling. Despite the fact that NetApp is doing some cool stuff with hyperscalers as well (that I’ll cover in a separate post). But a large percentage of the world’s critical workloads are still hosted on what we might term “traditional” storage appliances hosted in data centres across the globe and accessed and managed in a similar fashion to the way they were 5 or 10 years ago. But that doesn’t mean that companies like NetApp aren’t innovating. Indeed, looking at some of the specs and numbers being touted here, it’s easy to see why companies are sticking with the likes of NetApp when it comes to hosting enterprise workloads on enterprise storage. I don’t think the end of traditional storage is here by any stretch, and I’m pretty sure there’s more innovation to come.

Random Short Take #100

Posted on by

Welcome to Random Short Take #100. I’m back from my holiday, but it’s only a few weeks until Christmas, so this is likely the last Random Short Take for the year. Let’s get random.

Bonus round. vExpert applications are now open for next year. Read more about that here. In the meantime, stay safe, don’t get shot, and see you in the new year.