Random Short Take #46

Welcome to Random Short Take #46. Not a great many players have worn 46 in the NBA, but one player who has is one of my favourite Aussie players: Aron “Bangers” Baynes. So let’s get random.

  • Enrico recently attended Cloud Field Day 9, and had some thoughts on NetApp’s identity in the new cloud world. You can read his insights here.
  • This article from Chris Wahl on multi-cloud design patterns was fantastic, and well worth reading.
  • I really enjoyed this piece from Russ on technical debt, and some considerations when thinking about how we can “future-proof” our solutions.
  • The Raspberry Pi 400 was announced recently. My first computer was an Amstrad CPC 464, so I have a real soft spot for jamming computers inside keyboards.
  • I enjoyed this piece from Chris M. Evans on hybrid storage, and what it really means nowadays.
  • Working from home a bit this year? Me too. Tom wrote a great article on some of the security challenges associated with the new normal.
  • Everyone has a quadrant nowadays, and Zerto has found itself in another one recently. You can read more about that here.
  • Working with VMware Cloud Director and wanting to build a custom theme? Check out this article.

StorCentric Announces Data Mobility Suite

StorCentric recently announced its Data Mobility Suite (DMS). I had the opportunity to talk to Surya Varanasi (StorCentric CTO) about the news, and thought I’d share some of my notes here.

 

What Is It?

DMS is being positioned as a suite of “data cloud services” by StorCentric, with a focus on:

  • Data migration;
  • Data consistency; and
  • Data operation.

It has the ability to operate across heterogeneous storage, clouds, and protocols. It’s a software solution based on subscription licensing and uses a policy-driven engine to manage data in the enterprise. It can run on bare-metal or as a VM appliance. Object storage platform / cloud support if fairly robust, with AWS, Backblaze B2, and Wasabi, amongst others, all being supported.

[image courtesy of StorCentric]

Use Cases

There are a number of scenarios where a solution like DMS makes sense. You might have a bunch of NFS storage on-premises, for example, and want to move it to a cloud storage target using S3. Another use case cited involved collaboration across multiple sites, with the example being a media company creating content in three places, and working in different time zones, and wanting to move the data back to a centralised location.

Big Ideas

Speaking to StorCentric about the announcement, it was clear that there’s a lot more on the DMS roadmap. Block storage is something the team wants to tackle, and they’re also looking to deliver analytics and ransomware alerting. There’s also a strong desire to provide governance as well. For example, if I want to copy some data somewhere and keep it for 10 years, I’ll configure DMS to take care of that for me.

 

Thoughts and Further Reading

Data management means a lot of things to a lot of people. Storage companies often focus on moving blocks and files from one spot to another, but don’t always do a solid job of capturing data needs to be stored where it does. Or how, for that matter. There’s a lot more to data management than keeping ones and zeroes in a safe place. But it’s not just about being able to move data from one spot to another. It’s about understanding the value of your data, and understanding where it needs to be to deliver the most value to your organisation. Whilst it seems like DMS is focused primarily on moving data from one spot to another, there’s plenty of potential here to develop a broader story in terms of data governance and mobility. There’s built-in security, and the ability to apply levels of data governance to data in various locations. The greater appeal here is also the ability to automate the movement of data to different places based on policy. This policy-driven approach becomes really interesting when you start to look at complicated collaboration scenarios, or need to do something smart with replication or data migration.

Ultimately, there are a bunch of different ways to get data from one point to another, and a bunch of different reasons why you might need to do that. The value in something like DMS is the support for heterogeneous storage platforms, as well as the simple to use GUI support. Plenty of data migration tools come with extremely versatile command line interfaces and API support, but the trick is delivering an interface that is both intuitive and simple to navigate. It’s also nice to have a few different use cases met with one tool, rather than having to reach into the bag a few different times to solve very similar problems. StorCentric has a lot of plans for DMS moving forward, and if those plans come to fruition it’s going to form a very compelling part of the typical enterprise’s data management toolkit. You can read the press release here.

Quobyte Announces 3.0

Quobyte recently announced Release 3.0 of its software. I had the opportunity to speak to Björn Kolbeck (Co-Founder and CEO) about the release, and thought I’d share some thoughts here.

 

About Quobyte

If you haven’t heard of Quobyte before, it was founded in 2013 by some ex-Googlers and HPC experts. The folks at Quobyte were heavily influenced by Google’s scale-out software model and wanted to bring that to the enterprise. Quobyte has had software in production since 2016 and has customers across a range of industry verticals, including financial services and media streaming. It’s not really object storage, more a parallel file system or, at a stretch, scale-out NAS.

 

The Tech

Kolbeck describes Quobyte as “storage for Generation Scale-Out” and is focussed on “getting storage out of the ugly corner of specialised appliances”.

Unlimited Performance

  • Linear scaling delivers unlimited performance
  • No bottlenecks – scale from small to 1000s of servers
  • No more NFS – it’s part of the problem

Deploy Anywhere

  • True software storage runs anywhere – bare metal, containers, cloud
  • Almost any x86t server – no appliances

Unconditional Simplicity

  • Anyone can do storage, it’s just another Linux application
  • All in user space, installs in minutes

 

The Announcement

Free Edition

The first part of the announcement is that there’s a free edition (previously there was a 45 day trial on offer). It’s limited in terms of capacity, support, and file system clients, but could be useful in labs and smaller environments.

[image courtesy of Quobyte]

3.0 Release

The 3.0 release is also a big part of Quobyte’s news, with the new version delivering a bunch of new features, most of which are outlined below.

360 Security

  • Holistic data protection
  • End to end AES encryption (in transit / at rest / untrusted storage nodes)
  • Selective TLS support
  • Access keys for the file system
  • X.509 certificates
  • Event stream (metadata, file access)

Policy Engine

Powerful Policy Engine

  • For: Tenant, volume, file, client
  • Control: Layout, tiering, QoS, recoding, caching
  • Dynamic: Runtime re-configurable

Automated

  • Auto file layout: replication + EC and Flash + HDD
  • Auto selection of replication factor, EC schema

Self-Service

Quobyte is looking to deliver a “cloud-like experience” with its self-service capabilities.

Login for users

  • Manage access keys
  • Check resource consumption

Authenticate using access keys

  • S3
  • File system driver
  • K8s / CSI
  • User-space drivers: HDFS, TF, MPI-IO

Multi-Cluster

Data Mover

  • Bi-directional sync (evental consistency)
  • Policy-based data tiering between clusters
  • Recoding

TLS between clusters

More Native Drivers

HDFS

MPI-IO

Benefit of kernel bypass

  • Lower latency
  • Less memory bandwidth

 

Thoughts and Further Reading

One of the challenges with software-defined storage is invariably the constraint that poor hardware choices can put on performance. Kolbeck acknowledged that Quobyte is “as fast as your hardware”. I asked him whether Quobyte provided guidance on hardware choices that worked well with the platform. There is a bunch of recommended (and tested) hardware listed on this page. He did mention that whichever way you decided to go, it was recommended to stick with either Mellanox or Broadcom NICs due to issues observed with other vendors’ Linux drivers. There’re also recommendations on the site for public cloud instance sizing covering AWS, GCP, and Oracle.

Quobyte is being deployed to support scale-out workloads in the enterprise across a number of sectors including financial services, life sciences, media and entertainment, and manufacturing in Europe and Asia. Kolbeck noted that one of the interesting things about the advent of smart everything is that “car manufacturers are suddenly in the machine learning field” and looking for new ways to support their businesses.

There are a lot of reasons to like software-defined storage offerings. You can generally run them on anything, and performance enhancements can frequently be had via code upgrades. That’s not to say that you don’t get that with the big box slingers, but the flexibility of hardware choice has tremendous appeal, particularly in the enterprise market where it can feel like the margin on commodity hardware can be exorbitant. Quobyte hasn’t been around forever, but the folks over there seem to have a pretty solid heritage in software-defined and scale-out storage solutions – a good sign if you’re in the market for a software-defined, scale-out storage solution. Some folks are going to rue the lack of NFS support, but I’m sure Kolbeck and the team would be happy to sit down and discuss with them why that’s no great loss. There’s some pretty cool stuff in this release, and the free edition is definitely worth taking for a spin. I’m looking forward to hearing more from Quobyte over the next little while.

StorONE Q3-2020 Update

StorONE recently announced details of its Q3-2020 software release. I had the opportunity to talk about the announcement with George Crump and thought I’d share some brief thoughts here.

 

Release Highlights

Performance Improvements

One of the key highlights of this release is significant performance improvements for the platform based purely on code optimisations. Crump tells me that customers with Intel Optane and NVMe SSDs will be extremely happy with what they see. What’s also notable is that customers still using high latency media such as hard disk drives will still see a performance improvement of 15 – 20%.

Data Protection

StorONE has worked hard on introducing some improved resilience for the platform as well, with two key features being made available:

  • vRack; and
  • vReplicate.

vRack provides the ability to split S1 storage across more than one rack (or row, for that matter) to mitigate any failures impacting the rack hosting the controllers and disk enclosures. You can now also set tolerance for faults at an enclosure level, not just a drive level.

[image courtesy of StorONE]

vReplicate extends S1:Replicate’s capabilities to provide cascading replication. You can now synchronously replicate between data centres or campus sites and then asynchronously send that data to another site, hundreds of kilometres away if necessary. Primary systems can be an All-Flash Array.next, traditional All-Flash Array, or a Hybrid Array, and the replication target can be an inexpensive hard disk only S1 system.

[image courtesy of StorONE]

There’s now full support for Volume Shadow Copy Service (VSS) for S1:Snap users.

 

Other Enhancements

Some of the other enhancements included with this release are:

  • Improved support for NVMe-oF (including the ability to simultaneously support iSCSI and FC along with NVMe);
  • Improved NAS capability, with support for quotas and NIS / LDAP; and
  • Downloadable stats for increased insights.

 

Thoughts

Some of these features might seem like incremental improvements, but this is an incremental release. I like the idea of supporting legacy connections while supporting the ability to add newer tech to the platform, and providing a way forward in terms of hardware migration. The vRack resiliency concept is also great, and a salient reminder that the ability to run this code on commodity hardware makes some of these types of features a little more accessible. I also like the idea of being able to download analytics data and do things with it to gain greater insights into what the system is doing. Sure, it’s an incremental improvement, but an important one nonetheless.

I’ve been a fan of the StorONE story for some time now (and not just because the team slings a few dollars my way to support the site every now and then). I think the key to much of StorONE’s success has been that it hasn’t gotten caught up trying to be a storage appliance vendor, and has instead focussed on delivering reliable code on commodity systems that results in a performance-oriented storage platform that continues to improve from a software perspective without being tied to a particular hardware platform. The good news is though, when new hardware becomes available (such as Optane), it’s not a massive problem to incorporate it into the solution.

StorONE has always talked a big game in terms of raw performance numbers, but I think it’s the addition of features such as vRack and improvements to the replication capability that really makes it a solution worth investigating. It doesn’t hurt that you can check the pricing calculator out for yourself before you decide to go down the path of talking to StorONE’s sales team. I’m looking forward to seeing what StorONE has in store in the next little while, as I get the impression it’s going to be pretty cool. You can read details of the update here.

Pure Storage Acquires Portworx

Pure Storage announced its intention to acquire Portworx in mid-September. Around that time I had the opportunity to talk about the news with Goutham Rao (Portworx CTO) and Matt Kixmoeller (Pure Storage VP, Strategy) and thought I’d share some brief thoughts here.

 

The News

Pure and Portworx have entered an agreement that will see Pure pay approximately $370M US in cash. Portworx will form a new Cloud Native Business Unit inside Pure to be led by Portworx CEO Murli Thirumale. All Portworx founders are joining Pure, with Pure investing significantly to grow the new business unit. According to Pure, “Portworx software to continue as-is, supporting deployments in any cloud and on-premises, and on any bare metal, VM, or array-based storage”. It was also noted that “Portworx solutions to be integrated with Pure yet maintain a commitment to an open ecosystem”.

About Portworx

Described as the “leading Kubernetes data services platform”, Portworx was founded in 2014 in Los Altos, CA. It runs a 100% software, subscription, and cloud business model with development and support sites in California, India, and Eastern Europe. The product has been GA since 2017, and is used by some of the largest enterprise and Cloud / SaaS companies globally.

 

What’s A Portworx?

The idea behind Portworx is that it gives you data services for any application, on any Kubernetes distribution, running on any cloud, any infrastructure, and at any stage of the application lifecycle. To that end, it’s broken up into a bunch of different components, and runs in the K8s control plane adjacent to the applications.

PX-Store

  • Software-defined storage layer that automates container storage for developers and admins
  • Consistent storage APIs: cloud, bare metal, or arrays

PX-Migrate

  • Easily move applications between clusters
  • Enables hybrid cloud and multi-cloud mobility

PX-Backup

  • Application-consistent backup for cloud native apps with all k8s artefacts and state
  • Backup to any cloud or on-premises object storage

PX-Secure

  • Implement consistent encryption and security policies across clouds
  • Enable multi-tenancy with access controls

PX-DR

  • Sync and async replication between Availability Zones and regions
  • Zero RPO active / active for high resiliency

PX-Autopilot

  • GitOps-driven automation allows for easier platform for non-storage experts to deploy stateful applications, monitors everything about an application, reacts and prevents problems from happening
  • Auto-scale storage as your app grows to reduce costs

 

How It Fits Together

When you bring Portworx into the Pure Storage picture, you start to see that it fits well with the existing Pure Storage picture. In the picture below you’ll also see support for the standard container storage interface (CSI) to work with other vendors.

[image courtesy of Pure Storage]

Also worth noting is that PX-Essentials remains free forever for workloads under 5TB and 5 nodes).

 

Thoughts and Further Reading

I think this is a great move by Pure, mainly because it lends them a whole lot more credibility with the DevOps folks. Pure was starting to make inroads with Pure Storage Orchestrator, and I think this move will strengthen that story. Giving Portworx access to Pure’s salesforce globally is also going to broaden its visibility in the market and open up doors to markets that may have been difficult to get into previously.

Persistent storage for containers is heating up. As Rao pointed out in our discussion, “as container adoption grows, storage becomes a problem”. Portworx already had a good story to tell in this space, and Pure is no slouch when it comes to delivering advanced storage capabilities across a variety of platforms. I like that the messaging has been firmly based in maintaining the openness of the platform and I’m interested to see what other integrations happen as the two companies start working more closely together. If you’d like another perspective on the news, check out Chris Evans’s article here.

Storage Field Day 20 – Wrap-up and Link-o-rama

Disclaimer: I recently attended Storage Field Day 20.  My flights, accommodation and other expenses were paid for by Tech Field Day. There is no requirement for me to blog about any of the content presented and I am not compensated in any way for my time at the event.  Some materials presented were discussed under NDA and don’t form part of my blog posts, but could influence future discussions.

This is a quick post to say thanks once again to Stephen and Ben, and the presenters at Storage Field Day 20. I had a super fun and educational time. For easy reference, here’s a list of the posts I did covering the events (they may not match the order of the presentations).

Storage Field Day 20 – I’ll Be At Storage Field Day 20

Storage Field Day 20 – (Fairly) Full Disclosure

Cisco MDS, NVMe, and Flexibility

Qumulo – Storage Your Way

Pure Storage Announces Second Generation FlashArray//C with QLC

Nebulon – It’s Server Storage Jim, But Not As We Know It

VAST Data – The Best Is Yet To Come

Intel Optane And The DAOS Storage Engine

 

Also, here’s a number of links to posts by my fellow delegates (in no particular order). They’re all very smart people, and you should check out their stuff, particularly if you haven’t before. I’ll attempt to keep this updated as more posts are published. But if it gets stale, the Storage Field Day 20 landing page will have updated links.

 

Jason Benedicic (@JABenedicic)

Nebulon Shadow Storage

 

David Chapa (@DavidChapa)

“High Optane” Fuel For Performance

 

Becky Elliott (@BeckyLElliott)

Guess Who’s Attending Storage Field Day 20?

 

Ray Lucchesi (@RayLucchesi)

Storage that provides 100% performance at 99% full

106: Greybeards talk Intel’s new HPC file system with Kelsey Prantis, Senior Software Eng. Manager, Intel

 

Vuong Pham (@Digital_KungFu)

Storage Field Day 20.. oh yeah!!

 

Keiran Shelden (@Keiran_Shelden)

Let’s Zoom to SFD20

 

Enrico Signoretti (@esignoretti)

An Intriguing Approach to Modern Data Center Infrastructure

Is Scale-Out File Storage the New Black?

 

Paul Stringfellow (@TechStringy)

Storage Field Day and The Direction of Travel

 

Keith Townsend (@CTOAdvisor)

Will the DPU kill the Storage Array?

 

[image courtesy of Stephen Foskett]

Intel Optane And The DAOS Storage Engine

Disclaimer: I recently attended Storage Field Day 20.  My flights, accommodation and other expenses were paid for by Tech Field Day. There is no requirement for me to blog about any of the content presented and I am not compensated in any way for my time at the event.  Some materials presented were discussed under NDA and don’t form part of my blog posts, but could influence future discussions.

Intel recently presented at Storage Field Day 20. You can see videos of the presentation here, and download my rough notes from here.

 

Intel Optane Persistent Memory

If you’re a diskslinger, you’ve very likely heard of Intel Optane. You may have even heard of Intel Optane Persistent Memory. It’s a little different to Optane SSD, and Intel describes it as “memory technology that delivers a unique combination of affordable large capacity and support for data persistence”. It looks a lot like DRAM, but the capacity is greater, and there’s data persistence across power losses. This all sounds pretty cool, but isn’t it just another form factor for fast storage? Sort of, but the application of the engineering behind the product is where I think it starts to get really interesting.

 

Enter DAOS

Distributed Asynchronous Object Storage (DAOS) is described by Intel as “an open source software-defined scale-out object store that provides high bandwidth, low latency, and high I/O operations per second (IOPS) storage containers to HPC applications”. It’s ostensibly a software stack built from the ground up to take advantage of the crazy speeds you can achieve with Optane, and at scale. There’s a handy overview of the architecture available on Intel’s website. Traditional object (and other storage systems) haven’t really been built to take advantage of Optane in quite the same way DAOS has.

[image courtesy of Intel]

There are some cool features built into DAOS, including:

  • Ultra-fine grained, low-latency, and true zero-copy I/O
  • Advanced data placement to account for fault domains
  • Software-managed redundancy supporting both replication and erasure code with online rebuild
  • End-to-end (E2E) data integrity
  • Scalable distributed transactions with guaranteed data consistency and automated recovery
  • Dataset snapshot capability
  • Security framework to manage access control to storage pools
  • Software-defined storage management to provision, configure, modify, and monitor storage pools

Exciting? Sure is. There’s also integration with Lustre. The best thing about this is that you can grab it from Github under the Apache 2.0 license.

 

Thoughts And Further Reading

Object storage is in its relative infancy when compared to some of the storage architectures out there. It was designed to be highly scalable and generally does a good job of cheap and deep storage at “web scale”. It’s my opinion that object storage becomes even more interesting as a storage solution when you put a whole bunch of really fast storage media behind it. I’ve seen some media companies do this with great success, and there are a few of the bigger vendors out there starting to push the All-Flash object story. Even then, though, many of the more popular object storage systems aren’t necessarily optimised for products like Intel Optane PMEM. This is what makes DAOS so interesting – the ability for the storage to fundamentally do what it needs to do at massive scale, and have it go as fast as the media will let it go. You don’t need to worry as much about the storage architecture being optimised for the storage it will sit on, because the folks developing it have access to the team that developed the hardware.

The other thing I really like about this project is that it’s open source. This tells me that Intel are both focused on Optane being successful, and also focused on the industry making the most of the hardware it’s putting out there. It’s a smart move – come up with some super fast media, and then give the market as much help as possible to squeeze the most out of it.

You can grab the admin guide from here, and check out the roadmap here. Intel has plans to release a new version every 6 months, and I’m really looking forward to seeing this thing gain traction. For another perspective on DAOS and Intel Optane, check out David Chapa’s article here.

 

 

VAST Data – The Best Is Yet To Come

Disclaimer: I recently attended Storage Field Day 20.  My flights, accommodation and other expenses were paid for by Tech Field Day. There is no requirement for me to blog about any of the content presented and I am not compensated in any way for my time at the event.  Some materials presented were discussed under NDA and don’t form part of my blog posts, but could influence future discussions.

VAST Data recently presented at Storage Field Day 20. You can see videos of their presentation here, and download my rough notes from here.

 

Feature Progress

VAST Data has come up with a pretty cool solution, and it continues to evolve as time passes (funny how that works). You can see that a whole stack of features has been added to the platform since the 1.0 release in November 2018.

[image courtesy of VAST Data]

The Similarity

One feature that caught my eye was the numbers that VAST Data presented that had been observed with the similarity-based data reduction capability (introduced in 1.2). In the picture below you’ll see a lot of 3:1 and 2:1. It doesn’t seem like that great a ratio, but the data that’s being worked on here is pre-compressed. My experience with applying data reduction techniques to pre-compressed and / or pre-deduplicated data is that it’s usually tough to get anything decent out of it, so I think this is pretty neat.

[image courtesy of VAST Data]

Snap to S3

Another cool feature (added in 3.0) is snap to cloud / S3. This is one of those features where you think, ha, I hadn’t been looking for that specifically, but it does look kind of cool.

[image courtesy of VAST Data]

Replicate snaps to object store

  • Independent schedule, retention

Presented as .remote folder

  • Self service restore (<30 days .snapshots, >30 days .remote)

Large objects

  • Data and metadata
  • Compressed

ReaderVM

  • Presents read-only .remote
  • .ovf, AMI
  • Parallel for bandwidth

 

Thoughts and Further Reading

You’ll notice I haven’t written a lot in this article. This isn’t because I don’t think VAST Data is intriguing, or that I don’t like what it can do. Rather, I think you’d be better served checking out the Storage Field Day presentations yourself (I recommend the presentations from both Storage Field Day 18 and Storage Field Day 20). You can also read my summary of the tech from Storage Field Day 18 here, but obviously things have progressed significantly since then.

As Howard Marks pointed out in his presentation, this is not the first rodeo for many of the folks involved in VAST Data’s rapid development. You can see from the number of features being added in a short time that they’re keen on making more progress and meeting the needs of the market. But it still takes time. SMB failover is cool, but some people might be more interested in seeing vSphere support sooner rather than later. I have no insight into the roadmap, but based on what I’ve seen over the last 20ish months, there’s been some great stuff forthcoming, and definitely more cool things to come. Coupled with the fact that this thing relies heavily on QLC and you’ve got a compelling platform at potentially a super interesting price point upon which you can do a bunch of interesting things, storage-wise. I’m looking forward to seeing what’s in store over the next 20 months.

Nebulon – It’s Server Storage Jim, But Not As We Know It

Disclaimer: I recently attended Storage Field Day 20.  My flights, accommodation and other expenses were paid for by Tech Field Day. There is no requirement for me to blog about any of the content presented and I am not compensated in any way for my time at the event.  Some materials presented were discussed under NDA and don’t form part of my blog posts, but could influence future discussions.

Nebulon recently presented at Storage Field Day 20. You can see videos of the presentation here, and download my rough notes from here. I’d also like to thank Siamak Nazari (CEO), Craig Nunes (COO), and the team for taking the time to do a follow-up briefing with me after the event. I think I took a lot more in when it was done during waking hours.

 

What Is It?

Nebulon defines its offering as “cloud-defined storage (CDS)”. It’s basically an add-in card that delivers “on-premises, server-based enterprise-class storage that consumes no server CPU / memory resources and is defined and managed through the cloud”. This is achieved via a combination of nebulon ON (the cloud management plane) and the Nebulon Services Processing Unit (SPU).

 

The SPU

The SPU is the gateway to the solution, and

  • Runs in any 2RU / 24 drive server;
  • Connects to server SSDs like any RAID card; and
  • Presents local or shared volumes to the application.

A group of SPU-equipped servers is called an “nPod”.

The SPU is built in such a fashion as to deliver some features you may not traditionally associate with host-side storage, including:

  • All-flash performance via a high perfromance 8-core 3GHz CPU and 32GB NVRAM for tier-1 performance with all-flash latencies; and
  • Zero-trust security using hardware-accelerated encryption engines, a token-based “security triangle”, and crypto-authentication chip.

The Nebulon solution is designed to scale out, with support for up to 32 heterogeneous, SPU-enabled servers per nPod connected by 2x 25Gb Ethernet ports. Note that you can scale out your compute independently of your storage needs.

 

[image courtesy of Nebulon]

Key Features

Offloads the full storage software stack from the server.

Enterprise data services

  • Data efficiency: deduplication, compression, thin provisioning
  • Data protection: encryption, erasure coding, snapshots, replication

No software dependencies

  • In-box drivers for all hypervisors and operating systems without managing multi-pathing or firmware dependencies

1.3x more performance from each server

  • Application / VM density advantage vs “restrictive” SDS

isolated fault domain

  • Storage and data services are not impacted by operating system or hypervisor reboots

 

It’s Not On, nebulon ON

Always Up-to-Date Software

The cool thing is the nebulon ON cloud control plane is always being updated and delivered as a service. You can leverage new features instantly, and there’s scope for some AI stuff to be done there too. The SPU runs a lightweight storage OS: nebOS. Nebulon says it’s fast, with infrequent updates and no disruption to service, and scheduling updates is also apparently easy.

 

Thoughts and Further Reading

I tried to come up with a witty title for this post, because the name Nebulon makes me think of Star Trek. But I’ll admit my knowledge of Star Trek runs to “Star Trekkin’” by The Firm, so I can’t really say whether that’s a valid thing. In any case, I didn’t immediately get the value that Nebulon offered, and it wasn’t until the team took me through the offering for a second time (and it wasn’t the middle of the night for me) that I think I started to get the value proposition. Perhaps it was because I’m still working with “traditional” storage vendors on a more regular basis – the exact solution that Nebulon is looking to remove from environments.

“Server storage” is an interesting thing. There are a lot of reasons why it’s a good thing, and well suited to a number of workloads. When coupled with a robust management plane and “enterprise” resilience features, server storage can have a lot of appeal, particularly if you need to scale up or down quickly. One thing that makes the Nebulon solution super interesting is the fact that the management is done primarily from the cloud offering. I confirmed with the team that nothing went bad with the storage itself when the management plane was unavailable for a period of time. I also confirmed that the typical objection handling they were seeing in the field regarding security came down to the need to do a workshop and run through the solution with the security folks to get it over the line.

This solution has a lot of appeal, depending on how you’re consuming your storage resources today. If you’re already down the track of server storage, this may not be as exciting, because you might have already done a lot of the heavy lifting and gotten things working just so. But if you’re still using a traditional storage solution and looking to change things up, Nebulon could have some appeal, particularly as it provides some cloud-based management and some level of grunt on the SPUs. The ability to couple this with your preferred server vendor will also have appeal to the bigger shops looking to leverage their buying power with the bigger server slingers. Enrico covered Nebulon here, and you can read more on cloud-defined storage here.

Pure Storage Announces Second Generation FlashArray//C with QLC

Disclaimer: I recently attended Storage Field Day 20.  My flights, accommodation and other expenses were paid for by Tech Field Day. There is no requirement for me to blog about any of the content presented and I am not compensated in any way for my time at the event.  Some materials presented were discussed under NDA and don’t form part of my blog posts, but could influence future discussions.

Pure Storage recently announced its second generation FlashArray//C – an all-QLC offering offering scads of capacity in a dense form factor. Pure Storage presented on this topic at Storage Field Day 20. You can see videos of the presentation here, and download my rough notes from here.

 

It’s A Box!

FlashArray//C burst on to the scene last year as an all-flash, capacity-optimised storage option for customers looking for storage that didn’t need to go quite as fast the FlashArray//X, but that wasn’t built on spinning disk. Available capacities range from 1.3PB to 5.2PB (effective).

[image courtesy of Pure Storage]

There are a number of models available, with a variety of capacities and densities.

  Capacity Physical
 

//C60-366

 

Up to 1.3PB effective capacity**

366TB raw capacity**

3U; 1000–1240 watts (nominal–peak)

97.7 lbs (44.3 kg) fully loaded

5.12” x 18.94” x 29.72” chassis

 

//C60-494

 

Up to 1.9PB effective capacity**

494TB raw capacity**

3U; 1000–1240 watts (nominal–peak)

97.7 lbs (44.3 kg) fully loaded

5.12” x 18.94” x 29.72” chassis

 

//C60-840

 

Up to 3.2PB effective capacity**

840TB raw capacity**

6U; 1480–1760 watts (nominal–peak)

177.0lbs (80.3kg) fully loaded

10.2” x 18.94 x 29.72” chassis

 

//C60-1186

 

Up to 4.6PB effective capacity**

1.2PB raw capacity**

6U; 1480–1760 watts (nominal–peak)

185.4 lbs (84.1 kg) fully loaded

15.35” x 18.94 x 29.72” chassis

 

//C60-1390

 

Up to 5.2PB effective capacity**

1.4PB raw capacity**

9U; 1960–2280 watts (nominal–peak)

273.2 lbs (123.9 kg) fully loaded

15.35” x 18.94 x 29.72” chassis

Workloads

There are reasons why the FlashArray//C could be a really compelling option for workload consolidation. More and more workloads are “business critical” in terms of both performance and availability. There’s a requirement to do more with less, while battling complexity, and a strong desire to manage everything via a single pane of glass.

There are some other cool things you could use the //C for as well, including:

  • Automated policy-based VM tiering between //X and //C arrays;
  • DR using the //X at production and //C at your secondary site;
  • Consolidating multiple //X array workloads on a single //C array for test and dev; and
  • Consolidating multiple //X array snapshots to a single //C array for long-term retention.

 

It’s a QLC World, Sort Of

The second generation is FlashArray//C means you can potentially now have flash all through the data centre.

  • Apps and VMs – provision your high performance workloads to //X, lower performance / high capacity workloads to //C
  • Modern Data Protection & Disaster Recovery – on-premises production applications on //X efficiently replicated or backed up to //C at DR site
  • User File Shares – User file access with Purity 6.0 via SMB, NFS

QLC nonetheless presents significant engineering challenges with traditionally high write latency and low endurance (when compared to SLC, MLC, and TLC). Pure Storage’s answer to that problem has been to engineer the crap out of DirectFlash to get the required results. I’d do a bad job of explaining it, so instead I recommend you check out Pete Kirkpatrick’s explanation.

 

Thoughts And Further Reading

I covered the initial FlashArray//C announcement here and many of the reasons why this type of offering is appealing remain the same. The knock on Pure Storage in the last few years has been that, while FlashArray//X is nice and fast and a snap to use, it couldn’t provide the right kind of capacity (i.e. cheap and deep) that a number of price-sensitive punters wanted.  Sure, they could go and buy the FlashArray//X and then look to another vendor for a dense storage option, but the motivation to run with a number of storage vendors in smaller enterprise shops is normally fairly low. The folks in charge of technology in these environments are invariably stretched in terms of bodies on the floor to run the environments, and cash in the bank to procure those solutions. A single vendor solution normally makes sense for them (as opposed to some of the larger shops, or specialist organisations that really have very specific requirements that can only be serviced by particular solutions).

So now Pure Storage has the FlashArray//C, and you can get it with some decent density, some useful features (thanks in part to some new features in Purity 6), and integration with the things you know and like about Pure Storage, such as Pure1 and Evergreen storage. It seems like Pure Storage has done an awful lot of work to squeeze performance out of QLC whilst ensuring that the modules don’t need replacing every other week. There’s a lot to like about the evolving Pure Storage story, and I’m interested to see how they tie it all together as the portfolio continues to expand. You can read the press release here, access the data sheet here, and read Mellor’s take on the news here.