25 hours at CLEUR

Last week I was lucky enough to get a chance to attend Cisco Live in Berlin. I have been at this venue before but this is my first Cisco event and I have to say I was impressed. Hosted at the Berlin Messe it didn’t feel overly crowded yet with over 12,000 people involved in the conference it barely used up a third of the 26 halls available for events there. My reason for attending was a FlexPod round-table to be hosted jointly by people from NetApp and Cisco. I was in attendance as the voice of Arrow ECS Europe and, as the UK distributor involved in the most FlexPods, I thought it was important not only to give my feedback at this event but also to hear the messaging coming out directly from the vendors and pass this back to our reseller partners in the UK and also back to Arrow ECS.

file-28-02-2017-16-31-29

Sadly no AAA

I attended the conference in both a virtual and physical capacity; virtually reviewing the content available from keynotes and also by being there on an Explorer pass. This basically got me to everything but breakout sessions. Being immersed in the Cisco community was a refreshing experience and one I would recommend. Even without attending sessions there is a huge amount of information available to gather, not just from Cisco but also from some of their strategic partners including and not limited to Veeam, F5 and Citrix.

At the round-table it was great to hear was the rate of growth from a FlexPod perspective. A partnership just over five years old and it’s an over $7 billion business, and the number 1 integrated infrastructure. It also great to see that they are not resting on their laurels with a new CVD released that week covering how to deploy FlexPod Datacenter with Docker Datacenter for Container Management and with more in the pipeline narrowing the gap between private and hybrid clouds I would have to say that this is a partnership with plenty left in the tank.

file-28-02-2017-16-36-07

Ready for the DevOps community

I swung by the NetApp stand after and heard about another exciting FlexPod project the All Flash 3D FlexPod, and for anyone who attended the UK partner academy last June might recall a presentation on an older version of this project. We often talk about FlexPod being more than a sum of all the constituent parts and this is one case where this statement truly shines. Being used in anything from the medical profession to 4k content creation to geological applications this is a true monster and I doubt we have really scratched the surface of the areas this solution could be applicable I would suggest checking out here for more information.

It was also great to see that a MetroCluster FlexPod was running the event, by swinging by the NOC you could see the statistics in real time like the 20GB (yes Gigabytes) of internet traffic flowing around the campus supported by 968 access points (they added 75 on Tue night to improve the experience) yet with everything going on the AFF8060 never really was taxed as seen on the Graphana dashboard.

file-28-02-2017-16-37-17

Monitoring one half of the NOC AFF FlexPod

What did hit me whilst wandering around the many halls was the vast plethora of Cisco products and how this company has evolved. I knew some of different areas they but there’s so much more than the routing and switching that the business was born out of. I talked to many interesting people covering varying business units from IOT to Digital Finance Services to business transformation. If I had had more time there there’s so much more I would have like to have done, like sat down for a few hours and run through a self-passed lab, attend one of the many of what looked like sessions in the DevOps area or even give 20 minutes of my time to their charitable cause to Rise Against Hunger. One thing that hit me was this was a company with employees understanding the companies vision, it’s like an 8 person coxed crew perfectly in time with one another lifting their stroke rate above 32 still creating perfect puddles and yet not breaking a sweat. The slogan for the event was Your Time Is Now and I have to say that we are definitely in a Cisco Era.

The A700s Killing it in the Storage Market

On the 31st of Jan NetApp released a new box the A700s and it’s a game changer in more ways than you might know. On the same day they also released its SPC-1 benchmark results to show how pearly white its teeth are.

Hardware

Let’s start with the physical, this All Flash Array is a 4U chassis containing controllers and storage. NetApp has been producing integrated disk and controller systems for many years now but these have always been aimed at the SMB market. The A700s is the first designed with the enterprise market in mind and boy what a first. Housing 24 internal SSDs and expandability up to 216 per HA pair, with the ability to scale out to 12 SAN nodes or 24 NAS. This gives you some serious expandability options.

Each controller has four on-board 40Gbe ports with four PCIe slots for a vast array of additional functionality including more 40Gbe, maybe 32Gb FC, or even expanding to my personal favourite a MetroCluster.

Operating System

As expected this system is running the world’s number one branded operating system, ONTAP which provides a unified SAN and NAS architecture to meet the most demanding workloads. Consisting of anything from consolidated virtualisation, enterprise applications to design and engineering workloads, all delivered from a truly scale out architecture, ONTAP can manage up to 24 nodes as a single system with 13 PB of all flash storage.

SPC-1

The SPC-1 benchmark to quote their website – “consists of a single workload designed to demonstrate the performance of a storage subsystem while performing the typical functions of business critical applications. Those applications are characterized by predominately random I/O operations and require both queries as well as update operations. Examples of those types of applications include OLTP, database operations, and mail server implementations.”

This benchmark is a way to allow customers to compare various storage vendors. NetApp previously under took this benchmark back in April 2015 with the FAS8080AE and by way of comparison to how things have changed in just under two years I have put some of the more relevant results into a table. (For more detail see here and check out exec summary)

spc1table

SPC-1 Highlights

As you can see the FAS8080AE; which was before NetApp were required to adopt Gartner’s naming scheme for it to qualify as an All Flash Array; performed admirably giving us just over 685,000 IOPS at 1.2 milliseconds. This 8 node cluster was placed fifth in the top ten table but by the end of 2016 it had slowly been pushed out to 8th which is still very impressive. The most recent results published Jan 2017 show that the A700S delivered 2.4 million IOPS at roughly 0.7ms in a 12 node cluster. This huge number not only shows us the improvements in hardware that have occurred over the last couple of years but also the advances that NetApp has made with its ONTAP operating system. Even if you don’t need to scale out to 12 nodes a HA pair can deliver over 400,000 IOPS in under 0.3ms and when you consider you can stack it with 15.3TB SSDs giving you an effective capacity of a petabyte in a 4U enclosure delivering 650,000 (IOPS) in under a millisecond!

So what do these numbers actually mean? NetApp have reduced the physical capacity down from 2 full racks (84RU) to just over half a rack (26RU) whilst upping the node count by 50%. Yet by doing this they have greatly increased the throughput. So they’ve quartered your rack space requirements halved the latency and quadrupled the IOPS and this isn’t the box pushed to its max.

What’s in a name?

The S in the name has been said to represent slim but some have said it could well stand for sexy (and what a beast it is), or sport, but unlike a comparison made to hot hatchbacks, I would say this model is more the Aston Martin DB11 Intrepid Sport in Cinnabar Orange. This V12 monster is turning heads everywhere it goes, people are noticing her before they even get the chance to set eyes on her. Dropping into 3rd place in the SPC-1 is no mean feat but to do this with kit that occupies roughly half a rack is phenomenal!

The A700s is not trying to sneak around the corner no this AFA has all the capabilities we’ve come to love from ONTAP whether that be the data efficiencies with dedupe and compaction, SnapMirror, FlexClone, per Volume Encryption, Secure Multi tenancy, can form part of your Data Fabric solution, the list goes on. Remember this is an OS that keeps going from strength to strength as we can see from the addition of FlexGroup providing massively scalable next generation data containers (for more information see @NFSDudeAbides post here ) and this is hardware that marries the advances in technology beautifully.

Conclusion

All I can say is that if you are in the storage market this will have made you sit up and take note, and if you’re one of the many flash start-ups this has probably got you scared. No matter how you slice it this box delivers in all respects and is a deadly addition to any environment; just like Old MacHeath’s jack-knife.

The NetApp, They Are A-Changin’

 

A lot of people criticize NetApp for not moving with the times. Some of the newer start-ups like to claim that NetApp is a legacy company not in touch with today’s marketplace. Yet we all know the company has a rich and deep heritage which spans nearly a quarter of a century with over 20 of those years spent on the NASDAQ; so they must be doing something right.

They also like to say NetApp are not in touch with today’s data centre requirements. I would question that. Today NetApp launches the start of a whole new line for the FAS and All Flash FAS side of the portfolio. They have announced three new FAS models: the FAS2600, the FAS8200, and the FAS9000. And on the all-flash side, another two new models. These systems are designed with the data centre of the future in mind, and these enterprise products again deliver an industry first (NetApp were the first to support 15.3TB SSD drives), with next-generation networking in the form of 40Gbe and 32GB FC.

The FAS9000 is the new flagship of the line, and introduces a new modular design similar to what we have seen Cisco adopt to great success in the UCS line. This system has 10 PCI slots per controller which, when combined with the ability of either of the next-gen networking previously mentioned, gives HUGE amounts of bandwidth to either flash and NL-SAS drives. It also has a dedicated slot for NVMe SSD to help with read caching (aka Flash Cache) for those workloads that benefit from a read boost, and has the ability to swap out the NVRAM and controller modules separately, which is to allow for expansion upgrades in the years to come. Here are some of the numbers associated with the FAS9000: it can scale up to 14 PB (Petabytes) per high availability pair (HA pair) or up to 172PB for a 24 node (12 HA pairs) in a NAS environment. Yes, that’s up to 172PB of flash storage managed as a single entity!!

They also announced the arrival of the FAS8200, the new workhorse for enterprise workloads, delivering six 9s or greater of availability. It carries 256GB of RAM—that’s equivalent to today’s FAS8080, or 4x of what’s found in a FSA8040—with 1TB of NVMe M.2 Flash Cache as standard (which frees up a PCIe slot) and can scale to 48TB of flash per HA pair when combined with Flash Pool technology. The FAS8200 also has 4x UTA2 and 2x 10T ports on board. This system is ready to go, and if you need to add 40Gbe or 32Gb FC, this chassis will support the addition of those via cards. This 3U chassis will support up to 4.8PB and can scale out to 57PB, meeting any multi-protocol or multi-application workload requirements.

Another new member to the FAS family is the FAS2600, which replaces the ever popular FAS2500 series. For this market space, disk and controllers contained within the same chassis are prevalent, and the trend that started with the original FAS2000 (maybe even the good ole StoreVault) is still here today, with the FAS2600 offering similar options as the FAS2500 but now with SAS3 support. We have the FAS2620, which supports large form factor drives, whilst the FAS2650 supports the smaller variants. Something that is new to the FAS2000 series is the inclusion of Flash Cache, and the FAS2600 has received the gift of NVMe with 1TB standard per HA pair. Also, changes to the networking have been made. No longer do we have dedicated Gbe ports. Instead, they have change them to 10Gbe, which are for cluster interconnects, scaling up to 8 nodes in this range, and can now use all 4 UTA2 ports for data connectivity. And if you still require 1Gbe, it can be achieved via SFPs for these UTA2 ports (X6567-R6 for optical and X6568-R6 for RJ45).

NetApp, a company that, for some, may not be known for its flash portfolio, yet has sold north of 575PB of the stuff, have also announced two new controllers for the All-Flash Array (AFA) space; the A300 and the A700. These systems are designed purely for flash media, and it shows with the A300 supporting 256GB of RAM whilst the A700 runs with a terabyte of RAM (1024GB)! This huge jump will allow for a lot more processing from the 40Gb and 32Gb networks whilst still delivering microsecond response times. For this ultra-low latency, we are looking at either products like the Brocade X6 director for FC or Cisco’s 3132Q-V for Ethernet to meet these ever-increasing demands.

These new systems will support the world’s number one storage OS: ONTAP version 9.1 and beyond, with this new release also announced today. ONTAP 9.1 in itself has some improvements over the previous versions. We have seen some major boosts to performance, especially in the SME space with the FAS2600 gaining a 200% performance improvement over the previous generation with the FAS8200, and with the FAS9000, about 50% better than their predecessor. The new stellar performer in AFA space is the A700. This new AFA has been reported to handle practically double the workload of an AFF8080 running an Oracle database which is another huge leap in performance.

There are a couple of other nice new features in ONTAP 9.1 which I will mention here, but won’t go into too much detail on. The first would be FlexGroups, which is a single namespace spanning multiple controllers scaling all the way to 20PB or 400 billion files (think infinite volumes but done a lot better). Then there’s cloud tiering: the ability of an AFA to utilise an S3 object store for its cold data—now that’s H. O. T. HOT! ONTAP 9.1 also brings us volume-level encryption, which will work with any type of drive and only encrypt the data that needs it. The Data Fabric also gets an upgrade, with the inclusion of ONTAP Cloud for Azure, which has been a while behind the cloud version for AWS but is worth the wait. And finally we also get the ability with the Enterprise products running ONTAP 9.1 to scale to 12 nodes within a single SAN cluster(that’s the ability to add another 4 nodes).

On another note, NetApp did launch another new box just a couple of weeks ago; the new E2800 sporting the SANtricity OS 8.30, also available in AFA variants and delivering over 300,000 IOPS in a box designed for small and mid-sized businesses. Which like the SolidFire side of the portfolio should not be over looked if it meets all of your desired requirements.

So come gather round people, writers and critics alike. Take a good look. I think we can safely say, that NetApp is a keeping, itself in the game and delivering platforms that go beyond tomorrow’s requirements.

But the big question everyone wants to know is, “What does it look like?” For the answer to that, you should be at NetApp Insight!

ONTAP 9 A new flavour with plenty of features

 

Name change

NetApp recently announce the upcoming release of their flagship operating system for their FAS and AFF product lines. ONTAP 9 as you can glean from the name is the ninth iteration for this OS which like a fine wine keeps getting better with age. Some of you will also have noticed the simplification of the name, no more clustered, or data, just simply ONTAP. The reality is clustering is the standard way to deploy controllers which store data, so it’s not really necessary to repeat that in the name, a bit like Ikea telling you the things you can put inside a Kullen (or Hemnes or Trysil which are all improvements over the Hurdal). But the most important thing about this change is the numeral at the end, 9. This is the next new major release of the operating system providing all the features that were available in 7-mode but also so much more.

So now that we have got that out of the way let’s see what else
has changed….

New features

Let’s take a quick look at some of the new features; so grab a pen (or for the millennials your phone camera):

  • Firstly I think I should mention you can now get ONTAP in three different varieties dependant on your use case. The appliance based version, ONTAP; the Hyperscaler version, ONTAP Cloud; and the software only version ONTAP Select. This should allow for management of data where ever it exists.
  • SnapLock – Yes the feature that everybody wanted to know where it had gone when comparing cDOT with Data ONTAP 7-mode, yet less than 5% of systems worldwide used (according to ASUP) is back. WORM functionality to meet retention and compliance requirements.
  • Compaction – A storage efficiency technology that when
    combined with NetApp’s inline deduplication and compression allows you to fit even more into each storage block. More on this technology in a later post.
  • MetroCluster – Ability to scale out to up to 8 nodes. We can now have 1, 2 or 4 nodes per site as supported configurations. NetApp have also added the ability to have non mirrored aggregates on a MetroCluster.
  • On-board Key manager – which removes the need for an off box key manager system when encrypting data.
  • Windows Workgroups – Another feature making a return is the ability to setup a CIFS/SMB workgroup so now we don’t need an Active Directory infrastructure to carry out simple file sharing.
  • RAID –TEC – Triple Erasure Coding expanding on the protection provided by RAID –DP. Allowing us to add triple parity support to our RAID groups, this technology is going to be crucial as we expand to SATA drives in excess on 8TB and SSD beyond 16TB.
  • 15TB SSD support – Yes you read that right, NetApp are one of, if not the first, major storage vendor to bring you 15.3TB SSDs to market. We can utilise these with an AFF8080 giving you 1PB of guaranteed effective capacity in a 2U disk shelf!!! To continue that train of thought we could scale out to 367TB effective AFF capacity within a single cluster. This will radically change the way people think about and design datacentres of the future. By shrinking the required hardware footprint we in turn reduce the power and cooling requirements, lowering the overall OPEX for the datacentres of the future; this will lead to a hugely reduced timeframe for return on investment on this technology, which in turn will drive adoption.
  • AFF deployments – With ONTAP 9 NetApp are introducing the ability to rapidly deploy applications to use the storage within 10 minutes with one simple input screen and this wizard follows all the best practices for the selected application.

Upgrade concerns

One of the worries people previously had with regards to NetApp FAS systems was how to upgrade to a new version of the OS for your environment, especially if you had systems at both primary and DR.

Version independent SnapMirror which arrived with 8.3 is great if you have a complex system of bidirectional water-falling relationships as planning an upgrade prior to this
needed an A1 sized PERT chart to plan the event. Now NetApp allow for an automated rolling upgrade around a cluster, it should mean for those customers out there who have gone for a scale out approach to tackling their storage requirements (and I salute you on your choice) it’s the same steps if you have 2 or 24 controllers. Today you can undertake this with three commands for complete cluster upgrades which is such a slick process, heck you can even call up the API from within a PowerShell script.

How does it look?

If you look below I have a few screen shots showing some of the new interface including the new performance statics that OnCommand System Manager can now display.

Notice the new menu along the top. This helps to make moving around a lot easier.

Here we can see some of the performance figures for a cluster, as this is a sim I didn’t really drive too much IO at it, but it will be very useful once in production giving you insight to how your cluster is performing at 15 second intervals.

Another nice feature of the latest release is the search ability, which I think will come into its own in larger multi-protocol installations of several PB, helping hone in on the resource you are after quicker.

First impressions

For this article I am using a version in a lab environment and from its slick new graphical interface (see above) to the huge leaps made under the covers this OS keeps getting stronger. The GUI is fast to load even on a sim, the wizards methodical, the layout intuitive and once you start using this and have to jump back onto an 8.x version, as I did, you will appreciate the subtle differences and refinements that have gone into ONTAP 9.

Overall takeaways

With the advent of ONTAP 9 NetApp have also announce a 6 month cadence for future releases making it easier to plan for upgrades and improvements which is good news for those shops who like to stay at the forefront of technology. The inclusion of the features above and the advancements made under the cover should hopefully illustrate to you that NetApp is not a company who rests on their laurels but strives for innovation. The ability to keep adding more and more features yet making it simpler to manage, monitor and understand is a remarkable trait; and with this new major software release we get a great understanding of what this company hopes to achieve in the coming years.

This is also
an exciting upgrade for the Data fabric, as mentioned above ONTAP 9 is now available in 3 separate variants – engineered for FAS & AFF; ONTAP Select for Software Defined Storage currently running on top of vSphere or KVM; and ONTAP Cloud running in AWS and soon Azure. Businesses can now take even greater control of their data as they move to a bimodal method of IT deployment. As more and more people move to a hybrid multi-cloud model we will see people adopting these three options in varying amounts to provide the data management and functionality that they will require. As companies mix all three variations we get
what I like to call the Neapolitan Effect, which is probably one of the best of all ice-cream flavours; to their storage strategy, delivering the very best data storage and management wherever needed which is thanks to the ability of ONTAP to run simply anywhere.

So go out and download a copy today!