Hello, everyone. Welcome to the Breakout session BKO4004: Ensuring Business Continuity with NetApp SnapActive Sync and MetroCluster.

Very excited to be here.

My name is Anjali Munjal.

I'm a director of product management, looking after data protection, focused on business continuity and disaster recovery.

I've been with NetApp for the past four years, and in today's session would like to showcase w- how far we have come, where we are heading, and most important, listening, you know, to our most important customers.

Happy to say I'm sharing the stage with Senthil Kumar Ramakrishnan.

Hello, everyone.

My name is Senthil Kumar Ramakrishnan.

I am focused for the business continuity solutions, S- SnapMirror Active Sync and MetroCluster.

I'm here with the NetApp for the last two years and looking forward for the session.

Thank you.

Yeah.

And also excited to have Erik from NORRIS NETWORK AG to talk about real-world, life examples.

Yeah, Erik.

Yeah.

So my name is Erik.

I'm at NORRIS NETWORK a Principal IT Systems Engineer.

We are doing MSP business from co-location to service, full service experience, so we are also the NetApp Pinnacle Partner, and later on I will show some blueprints how we create our data service very reliable.

Okay.

So before we start, just our usual confidentiality notice, right?

Everything we talk is proprietary, and it's under NDA, and you need any written consents from our side to share it further.

So without further ado, let's start, right?

Setting the stage.

Why are we all here, right?

So as you heard in the keynote and the feature sessions, data is rapidly expanding.

It's growing exponentially.

And zero downtime is no longer a luxury or a choice.

It is the necessity to save the business.

And this session will go deep dive into how our powerful NetApp, portfolio can help you protect mission-critical data to have it up and running three sixty-five days a year and seven days a week.

So agenda for today, we'll start with, you know, I would say, quick recap on why zero downtime is imperative in today's digital age.

We'll go deeper into, NetApp SnapActive Sync, and then Senthil will talk about NetApp MetroCluster, a proven technology.

And with Erik on, you know, it's actually proven in practice, and he's gonna talk about his real-life configurations and environment.

We'll end the session with key takeaways and then followed by a Q&A, so hold on to your questions.

We'll have the time towards the end for question and answers.

So without further ado, yeah, let's begin.

Right.

So a quick show of hands, how many have you seen disruptions in your business in the past, last year, maybe starting January?

Couple of people.

Show of hands.

Okay.

Couple of them.

Yeah.

Right.

A couple of them, yes.

Okay.

Yeah.

So, so I would say, like, disasters, whether they're natural or man-made, you know, say hurricanes, earthquakes, wildfire, floods, right?

Can happen anytime.

There's little we can do to prevent them, but sure, we can protect from them.

And the business impacts are growing.

Like the data is growing, the impact on the business could range from, you know, thousands of dollars per hour to millions of dollars, and if you see the amount of recovery time, you can do the maths, you know, on the impact.

And speed of recovery matters.

As per the studies, ninety-three percent of the companies that do not have a disaster recovery plan or any plans to recover, they're gonna shut down in five years, and I'm sure none of us would be like to be in that situation, and every dollar that you spend today, you'll be saving seven dollars in the ROIs for recovery.

So again, looking at the maths, the ROI looks very, you know, impressive.

So, and we'll sure case how, right, NetApp's, SnapActive Sync and MetroCluster can help you save those, you know, dollars.

This is a slide, like, let's look at the overall picture.

I'm sure you must have again seen it at, you know, all the featured, you know, sessions and the keynotes.

We at NetApp want to provide you a very comprehensive, you know, NetApp platform.

It doesn't matter what the kind of data is, whether we are, you, looking at the AI agentic data or the, you know, in-cloud applications born in the cloud, or we're looking at enterprise applications, you know, or SQL, Oracle, you name it, running on virtualizations or whether they're running on on-prem, cloud, edge, you know, or any of the hyperscalers, we have you covered.

But the data becomes secure and available with all the data resiliency, and that's where the business continuity, as you can see, comes into play.

So let's talk about your data, right?

Every business has a wide variety of data, and every data has a different needs.

So you might have, you know, mission-critical datas, but you might have test dev data and so forth.

So let's talk in terms of how do you protect your data, and there are two factors.

What we call is the rep- RPO, which is recovery point objective, is the amount of data loss your applications can afford.

And also the RTO, which is the recovery time object, is that how long does it takes to recover your b- data.

So like every, you know, workload has a different RPO and RTO needs, so this slide shows you a comprehensive, you know, variety of the needs, how they can be fit.

And one size does not fit all, so this slide, if you look at the RPOs are getting stricter from left to right.

On the extreme left is the backup recovery, where you have RPOs in terms of hours or days.

The center is a disaster recovery, where you can afford some kind of downtime, you know, and you can be affordable between five to thirty minutes.

But on the extreme right is the business continuity, where your applications cannot afford any downtime, which means zero data loss and zero, you know, it has to be instantaneous.

And that's where our portfolio comes in.

So let's deep dive into how NetApp SnapMirror ActiveSync can help your data to be always available and always protected.

So again, a show of hands, two questions for everyone here.

Firstly, are you aware of SnapMirror ActiveSync?

Show of hands.

Oh.

Oh, wow, quite a lot.

Great to hear.

And how many of you are aware of SnapMirror ActiveSync for NAS, which was just released around the corner?

Oh, great.

I think that that's, that's good to know.

So for those of you who are not aware of SnapMirror ActiveSync, you know, I would say a refresher here.

So SnapMirror ActiveSync is a software-defined zero-RPO and zero-RTO solution based for the granular applications.

Again, the outcome is our application that the workloads needs to run seamlessly, and you can choose exactly which applications to be protective.

We don't need any fancy infrastructures out there, right?

No extra hardware gear.

As long as you have the IP networks providing us the ten-millisecond round trip time, we have it covered.

So you can go as far as you want, as long the ten-millisecond RTT is adhered.

As I mentioned, right, it is at a granular level solution, so you can choose exactly which applications, which workloads you need to protect.

And all that comes together with our storage constituents called the consistency group.

So let's say you have a SQL instance, right?

You need to protect your SQL logs and database together, and you can put them together in the consistency group, and you will get a consistent write order.

So every write, right, that is committed will be, you know, protected.

We started the journey with SnapMirror ActiveSync with our, for SAN workloads.

Then we last year, or, or I would say year and a half, we introduced called active-active configuration.

What that means is that your read/write is on the both the nodes, or I would say both the sides of the clusters.

So if you look at in the figure, right, domain A and domain B can simultaneously have the read/write on both the sides, which not only gives you the business continuity, but also gives you load balancing.

So you can use your resources very effectively in addition to having your zero downtime policy.

Alongside, you have the flexibility to choose how do you need the host configuration.

We have two flavors.

We call it uniform host configuration and non-uniform.

With uniform host configurations, you can have a host on site A access to site A as well as site B.

And likewise, a host on site B can have access to both the sites.

And with host proximity, you can define which host get the shortest path to the access.

So again, very flexible to choose your configuration to, to meet your business needs.

As I mentioned, right, we want zero disruption, and that's where our mediator comes in, which provides you the automatic, you know, unplanned failover so that, again, your mission-critical data is continuously up and running.

It avoids the split brain syndrome in the sense if the site goes down or the network goes down or their net connectivity is impacted, it helps make sure that you have the access to your data, whether it's site A or site B.

And we introduced what we call the cloud mediator last year.

So earlier, we had the ONTAP on-prem mediator, then we introduced the cloud mediator.

Now, that helps you reduce your TCO cost because we maintain the cloud mediator at NetApp, and then you don't have to worry about those upgrades and, you know, to be on the latest revs, and we have you covered.

So again, a lot of choices there.

And in addition, right, I mentioned that not one fit, you know, or one size fits all.

So along with business continuity, I'm sure many of you have, you, backup needs for your archival, compliance or the other objectives.

So we do have the support for a third-site protection using SnapMirror Asynchronous protected under SnapCenter, which gives you a holistic data protection for not only your business continuity, but for your archival compliance and backup needs.

So now that, right, we have covered what SnapMirror ActiveSync is, now let's go talk about the SnapMirror ActiveSync for NAS.

We just released this with our ONTAP nine dot 19 dot one just about a, you know, a, I would say month and a couple of months back.

Same objectives.

It's gonna provide you zero-RPO and zero protection at a granular level, and this time, the granularity at the storage virtual machine or the SVM level.

Again, all automatic transparent failover.

So as long as you have the NFS, we have all versions covered, and SMB3, CA shares.

So all the, you know, failovers will be transparent, no disruption to your applications.

For non-CA shares, however, you will see a disruption, and that's because we are limited by the technology that does not support it.

So as and when the CA sh- you know, non-CA shares or others will support, we will be able to support it.

Also, the configuration is gonna be active-passive, or we call it active-standby, whatever you want to call it, which means that your read/writes are happening on the one site, you know, site A, while your site B is dormant.

So it will only become active after a failover.

But if you need to do, you know, run your dev test or anything, you know, some kind of analytics on your data, you do have an option, and that is by spinning a flex clone out of the dormant site so that you can just do whatever you need to do it.

Again, right, as always, or with the SnapMirror ActiveSync, you don't need any dedicated hardware, extra switches or anything.

As long as the RTT is under ten milliseconds, again, the distance is gonna be, not a problem.

So let's take a little more deeper into the SnapMirror ActiveSync for NAS architecture.

Right.

Great, we have the data continuity, but what about the other, right, IP settings, protocols, and, you know, other, informations?

And I think some people had that question.

So with this, we ensure that your IP settings and all protocol settings remain as is after the failover with our configuration replication services, so that, again, you get a very seamless experience.

We do understand data is growing and so is the scale.

We have today support for two-node and four-node architecture, and yes, we want to grow, and I'll talk about more that in our roadmap slides.

But for this purpose, we have the two-node AFF cluster just in the diagram.

Good news, we support both SAN and NAS, you know, can coexist in the same cluster, so you don't need a dedicated cluster for SAN or a dedicated cluster for NAS.

So as you can see on the top of the figure, right, you have SVM A1 running on the NAS workloads and SVM A2 running the SAN workloads, which is your consistency group.

And after failover, right, both of them gets failover to corresponding SVMs.

Then in addition, you can have further, you know, I would say reduce, total cost of ownerships.

We have both unprotected and protected SVMs in the same cluster, and only the protected ones will be failed over.

So like in the below screen, right, we have the SVM A2, which has been protecting with SnapMirror ActiveSync, will, you know, failover with SVM A2, once you have the, you know, failover.

Another important one that I would like to highlight, and it's not shown in the diagram, but, hear me out.

Although we have an active-passive or active, I would say standby combination, we do have support kind of a pseudo active-active.

And what we mean by that is that you can have active SVMs serving different workloads on different sites simultaneously.

So for example, if you have an SVM 3, you know, on Site A is having a read/write on Site A with a dormant on Site B, then simultaneously you can have SVM 4 on Site B having read/write and dormant site as Site A.

So you can have those kind of a pseudo active-active, which means you can have active SVMs, but there has to be different SVMs on the both sites.

I think we covered a lot around, you know, what it is, so let's look at the demo.

It's gonna show how quickly we can configure it, and then also we'll run a manual failover to show, you know, how seamless it is.

So in this case, we are gonna talk about we're gonna protect a NAS volume, which is gonna be running on the NAS SVM.

It has two volumes underneath, which we would like to protect.

So right now it is not protected, as you can see.

So you wanna protect under SMAS.

So we'll go into the Replication, in Overview tab.

We select under the Business Continuity, we select the, the NAS option.

And you have a choice to protect all the volumes under it.

In this case, we have two volumes, we'll protect it.

But you have a choice to exclude the volumes, or you can have also, you know, only selectively which volume to select.

In this case, again, as I mentioned, we have two volumes.

We will be protecting both.

So it's all protected with Cluster A and Cluster B and having a cloud mediator.

So now let's see how we can, you know, stimulate a failover and, you know, we'll see how seamless the reads and the IOs are.

So this, both the volumes are protected with SnapMirror ActiveSync for NAS, and it will show, right, that both the volumes have, you know, Clusters A and B connected with the cloud mediator, so we're good to go.

And again, we're just checking on the Cluster B, right?

They are in the relationship with SnapMirror ActiveSync, and again, for the both the volumes.

Now, we have a script that will be generating a lot of IOs to it, so we'll show, you know, how the IOs, the latency, and throughput is getting impacted on the Cluster A.

And then after failover, how it, you know, it seamlessly transfers to Cluster B.

So right now you can see, right, the IOs are starting to spike up.

So you can see all the throughput, IOPS, and latencies are, are changing.

Now, what we want to do is we want to generate an on-demand, you know, manual failover just for the purpose of the demo.

In case of a real-life scenarios, these are all gonna be automated.

So we go, and choose the option for the manual failover under the Replication tab, and then we stimulate, the failover.

And now we'll go and check how the IOPS are behaving on the, the secondary site, on the Cluster B.

So right now it has been very dormant because all the IOPS has been going to the Cluster A.

And after the failover, now you will see that now the spikes are coming up and now they're being accessed on the Cluster B.

So all the information, now it is like seamlessly directed.

Instead of Cluster A, now they are being directed over to the Cluster B with all the spikes, you know, with the data that you see.

Yeah.

Okay.

It's almost there.

Yeah.

So, again, this was just a quick demo.

If you want to try, you know, hands-on, we have labs on demand and, you know, we can, or you can reach out to me, Senthil, and Eric.

We can always help you get up a setup to try your hands, you know, directly there.

Now, we had SAN, then we introduces the NAS workloads.

And this is, again, a comparison chart.

I'm not gonna go line by line.

But our aim is to have compatibility, I would say, or feature consistently with both SAN and NAS.

A couple of things that I want to highlight, right?

In terms of, yes, we do understand, right, as I mentioned, the scale is important, so we want to grow from four nodes to six, eight, you know, however twelve nodes we need.

And along the sides on the volume size, the number of consistency group, number of STFs that we support.

And it's a journey, and I'll show in the roadmap slides, right, where we are heading and how much, you know, and it will be a continuous journey, I would say, for some time.

Before I talk about the roadmaps, let's take a quick look, right?

It's great we have the technology, but without the tools to help it, I think it's not gonna do the job.

So good news, right?

With SnapMirror ActiveSync, it will integrate directly into your ecosystem.

So for example, you have, two sites, you know, one running at Berlin and one at Hamburg connected between SnapMirror ActiveSync for the virtualization workloads and your vSphere Metro Storage Cluster is handling, you know, the, the VMs, level.

With ONTAP, tools for VMware, it's directly integrate so you can provision and manage end-to-end life cycle through that for your, you know, entire management of the VM under SnapMirror ActiveSync.

In addition, right, if you want to add up a third site for a backup copy, say let's in Munich, and we just connected with SnapMirror Asynchronous, again, we have you covered with SnapCenter.

So with VMware LiteSite Recovery, and again, with your ONTAP tools for VMware, again, we have you holistically covered.

So not only just for the business continuity, but your other workloads that needs, you know, other, RPOs and RTOs requirements.

So I would say I think, this is probably the, I would say, the end of the SnapMirror ActiveSync section.

So we have come a long way, right?

We introduces the NAS support with ONTAP nine dot nineteen dot one.

A quick touch base, right?

I know still how many of you have AIX systems today?

Quick show of hands.

So there are a couple, right.

So good news, we have the transparent failover for your AIX systems.

Also, right, security is imperative, so we are investing in the journey with security, with providing support with anti-ransomware protection.

It's been added for the SAN workloads.

We'll continue to add for NAS.

With ONTAP nine dot twenty dot one, we will be having the support for FlexGroup for NAS, and also the choice to use Fibre Channel as a replication transport.

If you want to do that, you know, now you have a choice to use IP or FC for those FC infrastructures environment.

Gonna increase the scale to six hundred, then continue to increase the scale to thousand with our next release mid-next year.

And then also on the tamper-proof side that we want to provide again your protection from, you know, any accidental data loss, all the immutable copies for your NAS and SAN workloads.

So again, right, with SnapMirror ActiveSync, you'll see how seamless you can protect your workloads, whether you're talking with SAN, NAS or, you know, unprotected, you know, all together and just have your business continuity, you know, provide you the zero RPO and the zero RTO.

With that, I'll hand it over to Senthil.

Thank you, Anjani.

So now let's look on the NetApp's MetroCluster.

This is another business continuity solution designed to keep data and applications available in the si- in the site-level failures, right?

So MetroCluster combines the high availability within the site as well as synchronous replications across the site, which helps the customers or work you especially in the ensuring the zero data loss in the event of disaster.

Let's see some of the key principles which makes it more unique, the MetroCluster, right?

The first one is the simplicity.

Once it's been configured, the protection happens automatically in the background, so there is no, need of your teams to run your replication schedules or make sure that only your last cycle success, write has been successfully committed or not.

So it is a kind of solution where you can simply set and forget it kind of things.

The second one is that the data continuous data availability.

This question comes always with our customer discussions, right?

What is the toughest thing to explain to your CIO in the event of disaster?

Is it a couple of minutes of application downtime or the couple of minutes of lost data?

Most of our customers says it's the lost data.

That's what MetroCluster helps with the synchronous replications where the sing- similar copy is being copied in the both the sites.

So the write has been getting committed before it get acknowledged to the applications.

So in the practical terms, if you remember, if you see one site has been suddenly disappears, the o- other site has identical data over there.

The third principle is the continuous application availability.

Either you have been performing an planned failover or as some, or you are been experiencing a disaster, MetroCluster helps here to keep your applications running with the minimal disruptions.

Let's be honest, right?

Your best failover is the one which your application owner does not even notice that.

The last one is that operation simplicity.

So MetroCluster combines the high availability as well as a disaster solution in a single one, and it also supports all-flash and hybrid solution in the same architecture.

So the key message here is that Metro Cluster has been proven solution for more than two decades.

This is not a solution for slide representations or for the lab demos.

Our customers deploy this one in the mission-critical environments where every second or a minute of downtime costs them a lot of revenue loss, operation impact, or their customer experiences.

Okay, let's see how Metro Cluster has been helping with the, these kind of, things with the architecture.

If you can see this one is this, high-level Metro Cluster four node, architecture.

You can see that one, there are two independent ONTAP clusters has been kept in the two sites.

Typically, they have been deployed in the two different data centers, right?

And these two data centers have synchronous replications across them and keep the data intact.

Let's focus on the Site A, where you can see there is a node A1 and n-node A2.

They have maintained a HA partner relationship with each other.

That's the first level of protection we have been having in the Metro Cluster.

Similarly, you can see there in this clus- site B, where the node B1 and B2, they also maintain the same HA partnership relationship.

And you can see there is a arrow across the se, sites.

That makes Metro Cluster more unique.

So every node in the site A maintains a DR partnership in the site B.

So typically, a node maintains two relationship here.

One is a local partner, HA partner, the second one is a DR partner in the remote site.

Our engineering folks always ca- jokes on this one, right?

So Metro Cluster nodes maintain the social relationship better than any of us because they maintain the relationship all the time very much intact.

Let's do a small failure exercise here.

Let's assume that when the node A1 fails in the site A.

Who protects the data workload over here?

The node A2.

This is our first level of protection, the business continuity without the involvement of the second site.

Likewise, the whole Site A fails over here.

Who protects the workload here?

Obviously, site B, with the help of the Metro Cluster switchover functionality.

So the message here is that Metro Cluster has been designed to handle the smaller as well as the bigger failures over here.

Let's see that on how the switchover functionality works for the Metro Cluster.

Let's start with the situation, right?

It's a Sunday morning, two AM, and, one of your site has been completely down due to a power cut.

Who wants to be a person to receive that call, joining some bridges, and run the DR playbooks and do the manual switchover over there?

Obviously, nobody would like to be there in the situation.

That's where Metro Cluster ASO bui- is been designed for.

Automated unplanned switchover has been designed to take that switchover decision safely and automatically without administrator to be logged on.

So let's say that when ASO has been designed for site fail- failures, right?

In what are the scenarios it'll be, it'll be get triggered?

The first requirement for us is that the disaster site should be down.

If your administrator or someone accidentally reboots a node or someone unplug the cable in the back end, the Metro Cluster does not want to do a switchover decision over here.

All right?

It has to be a real site-level failures.

So the ASO will not be get triggered over there.

It should not be triggered for any of, the operational hiccups there.

The second requirement is the surviving site should be up.

This may look some, so obvious, but this is very important for the ASO functionality to work.

Metro Cluster makes sure that one surviving site is being capable, fully capable, before making that ASO or switchover decision over there.

The third requirement here is that mediator is reachable.

Now let's talk about the hero of the slide, ONTAP Mediator.

Anjali has been covered about the mediator on the SMAS, sorry, SnapMirror clustering.

The same mediator has been applicable for the Metro Cluster also here.

So mediator acts as a third point witness here.

One of the key challenge on the design or the problem for the customers is that if the two site loses the communication completely is the split brain situation, right?

Each sites thinks that one, they owns the data.

That's where mediator comes as a third pa- party reference or as a referee in a match.

There is a conflict arises, referee helps them to the team who possess the ball.

The similar functionality where the Metro Cluster uses the mediator to identify which site is being truly live and keep the possession of the data over there.

And the last one is the requirement here is that NVRAM sync has to be success.

This is one of the key requirement for us to keep our data, the customer data to be intact.

The fastest recovery is always good, but with the loss of recovery is not always.

So data integrity is one of the key principle for the Metro Cluster design, so NVRAM sync is being successful is one of the key requirement for Metro Cluster to make the ASO to, switch over for the other site.

And let's see that on what are the typical trigger points for the Metro Cluster ASO.

You can see that one is site-level failures of temperature, thermal or power failures or multiple node failures in the s- sites.

You would have been notice a pattern over there.

It is all site-level failures.

It's not a simple operation failures like node reboot or cable mismatch or so on.

So ASO has been typically designed to handle the site-level failures, not on the operational hiccups over here.

Right?

Okay, a quick knowledge check over here.

So how many of you are aware about the HTTPS communication for the ONTAP Mediator?

Oh, great.

Thank you.

Quite a few.

So in nine nineteen dot one, we introduced the HTTPS-based communication for the ONTAP, right?

Historically, we have been supporting for the iSCSI-based mailbox communication.

Don't take me wrong, it has been working fine, but it came with a lot of baggages.

One of the biggest thing was the SCSD kernel module, which has been depending on the iSCSI communication over there.

So in nine nineteen dot one, we introduced HTTPS-based communication where communica- where customers can benefit over the standard encryption support communication between the ONTAP and Mediator.

So there is no SCSD dependency by this, so you are having a faster OS upgrades or the kernel updates because of that.

And, we have been reducing a lot of platform dependency by removing the SCSD qualification cycles also over there.

Okay.

So as we conclude, I really want to spend some couple of minutes on our Metro Cluster innovation journey.

Over the years, we have been focused on the three key areas.

One is the security, the second is the scale, and the third one is the operational simplicity.

On nine seventeen dot one, we completed our end-to-end encryption journey for all the ONTAP, sorry, Mediator-supported platforms, and we also increased the scale for our medium as well as the high-level platforms there.

On nine eighteen dot one, we focused more on the operational simplicity, especially on the Mediator side.

We introduced the Oracle Linux host OS support, and we completed our IPv6 journey end-to-end completion by communication enablement between the Mediator and ONTAP.

The third one is the scale improvement, where we have been enhanced our scale from five to ten, for where the Mediator can manage-- one single Mediator m- can manage up to ten Metro Clusters.

In nine nineteen dot one, we introduced the HTTPS-based communications, and then we introduced the flow control capa- flow control management capabilities.

We are to help our customer who has the undersized ISL on the back end, right?

That is, in the level of, yeah, managing from the, yeah, is the setting levels, I would say.

All right?

And then we introduced some of the network monitoring capabilities and the Ansible scripts for the head upgrades over there.

In nine twenty dot one, we have been currently working on the increasing the volumes as well as the SVM for the active-passive.

That's a starting point for us to increase the scale.

The second is that we have been working on g- bridging the gap between the ONTAP capabilities and the Metro Cluster, specifically for the S3, SnapMirror support, as well as the tenant-aware capacity management for the Metro Cluster.

The last one is that we want to complete our flow control management capabilities, where we want to introduce the adaptive flow control for the nine twenty dot one.

In nine twenty dot one, we have been more focused to introduce the cloud Mediator support, where the SnapMirror active currently has it.

Along with that, we want to introduce a switchless Metro Cluster, which will work for a campus environment kind of things, where we want to have the back-to-back connectivity between the controllers to avoid the whole deployment of the switches.

Our POC shows very good progress on this one.

We saying that one, we do not want to say or lose any of the resilience or the failure capabilities of the Metro Cluster.

So we have been targeting this one for the upcoming releases.

And then we want to increase the volume as well as scaling limit for active-active of the Metro Cluster deployment, right, as well as introducing the new switch supporters from the Arista as well as for the Cisco.

So the key message here is that Metro Cluster is not a static solution.

We continue to invest on the cloud side to help the customers to reduce their CapEx and OpEx on the switchless, Metro by introducing the Metro Cluster and so on.

Okay.

Let's come to, come to the, last slide of mine, right?

So we, we discussed about SnapMirror ActiveSync, and we discussed about the Metro Cluster.

Yeah, very common question I receive from most of the customer is that what, which solution I have to select, SnapMirror ActiveSync or the Metro Cluster?

My answer is always the same.

The question is wrong.

What I'm trying to protect?

That's the right question here, right?

The good news is that both the solutions support the synchronous replication, supports the, zero RPO and RTO over here.

But what is it you want to protect, that's where you have to select the solution.

You can see that on SnapMirror ActiveSync work on the application level as well as the dataset.

It gives a great flexibility for the administrator to select what they really want to protect, right, and keep the others unprotected if they say it.

And this also supports active-active.

That means they don't write, read-write in the both the sites, as well as the faster, I/O resumption times, which is excellent choice for the mission-critical environments, right?

Where the, write has been mirrored in the both the sites.

On the other side, Metro Cluster has been working on the site level.

It has been working on the aggregate level, where the whole site level, failure will be helping for the customers.

The other major significant point is that the, now you have the flexibility to select what platforms on the SnapMirror ActiveSync.

You can select the different platforms in the site A as well as the site B.

That helps a heterogeneous kind of environment to help in the SnapMirror Active.

But in the Metro Cluster, the requirement is more on the TP, in the similar kind of platforms or the DR partner so that it can work on the aggregate level.

On the scale side, Metro Cluster has the good amount of scale it supports, and it also support the SAN, NAS, and S3 workloads in the same architecture.

So the whole choice of yours is based on what you are really want to protect and what is your use cases you want to protect over here.

So by that comes to an end of my, presentation here on the Metro Cluster.

We discussed about the Metro Cluster overview of the ca- business continuity capabilities as well as the innovation journey.

Nothing better than that when you hear directly from the customer who handles this one in the first hand of the all these capabilities over here.

I'm pleased to hand over to Mr. Erik from the Noris Network, who will share the Metro Cluster journey and their Metro Cluster deployment.

Most importantly, how Metro Cluster helped them in the real-life scenario to keep the business running.

Erik, thanks for joining with us.

The stage is yours.

Yeah.

Thanks for having me here.

So we have two additional points, when to choose SnapMirror ActiveSync and when to choose Metro Cluster.

I think everyone is aware of it, and everyone should cover the usage or the solution for us customers as well or us partners internally.

So you have to decide when to use which.

So we have the real world examples now in place.

Here you can see one of our beautiful multi-node Metro Clusters of our customers.

And yeah, at first we will start with a disaster I was faced a few weeks ago.

So my colleagues called me on the weekend and say, "Hey, Erik, how should we handle that?"

And yeah, the cause of this event, I'm not allowed to speak in detail to it, but we had an event in our data center which caused a multiple disk failure.

And even with RAID tech, we were unable to, you know, be online.

So even un-fail of a disk won't help.

So yeah, bad thing.

Aggregat was down.

The good thing was the customer was aware about this data, and he set it up in SnapMirror application.

Only one volume was unrecoverable, but customer decides that this volume is not, recoverable or is okay if it's in data loss.

So they accepted that one volume don't need protection.

All other volumes we were able to protect really fast with the SnapMirror, and then everything was up and running in I think at least one hour everything was restored.

So yeah, this was one disaster we have faced very recently, and we were happy that SnapMirror helped us.

And yeah, if our customers are asking, "Hey, how to protect our data?"

We have one solution, use a Metro Cluster.

So we put the customer data on a Metro Cluster, and everything is replicated with an RTO, RPO zero to the second site.

We connect this data service via BGP into our customer networks.

That's how it makes it a little bit easier, and the requirement for stretching the networks over a long distance is not, more any- there anymore.

So, the IP was simply announced in case of a switchover to the other side, and that even help us to have a fast and reliable network connection to it.

If the customer don't trust Metro Cluster, we are currently using SVMDR with an RPO, RTO of fifty minutes.

And, in the end, it's the same back end.

It's connected via BGP to the customer file space, and we are able to choose a disaster recovery plan or action plan or disaster recovery plan to do an failover very granularly.

So that's the second blueprint we offer our customers.

We have the solution to do it from Nuremberg to München or in Nuremberg to Nuremberg or to on third site so the customer could choose if they use Metro Cluster or SVMDR.

And if the customer's asking for more protection, we are use, we usually use an SnapMirror Acti- the Metro Cluster replication or the SVMDR, and we attach an additional vault to the data to a third site so that we can work with some short-term snapshot schedule on the Metro Cluster SVMDR solution and with a long-term, long-term protection schedule for more snapshots on the vault on the third site.

That are the three basic blueprints we are using.

Later on, we are switching to SnapMirror ActiveSync.

Currently, we are fine with SVMDR.

So yeah, we are looking forward to use the SnapMirror ActiveSync as well.

So that was a quick overview how we are doing our, solutions to protect our customer data.

And yeah, that's where we go to the key takeaways.

For me, the most important key takeaways, know your data and protect it.

If you don't know your data, you can't protect them efficiently.

So that's my key takeaway.

Yes.

I will hand over to you.

Thank you.

Yes.

Thanks, Erik.

That was, you know, really great to know, right, how it, we can help protect in the real life, and I think that's an, you know, testament to it.

Thanks for sharing your, you know, example.

So I would like to take three takeaways, you know, from this session before we open up for Q&A.

First thing, yes, zero data loss is achievable, right?

We showed you with our SnapMirror ActiveSync in Metro Cluster how seamlessly you can, you know, protect and have zero downtime for your mission-critical applications so that they can run three sixty five days a year, seven days a week.

Second, failover stays transparent, right?

With our, mediator, whether you're talking about on-prem or cloud or active/active, we ensure that any disruption is not impacting your application access, and they're seamlessly up and running as you need.

And every write that is, you know, committed is protected, and again, you can have a zero downtime with it.

Last but very important, you need to know your data in order to protect it, right?

Every data has its own needs in terms of, you know, how critical it is to your business and in terms of RPO and RTO.

So knowing your data in order to protect it is very important.

And then you can sca- you know, scope it out and design it based on, you know, whether SnapMirror ActiveSync is the right s- you know, perfect fit for you or is more of the Metro Cluster.

But always remember, right, we at NetApp are here to help you design scope so that you can have a very successful business continuity and a good night's sleep so that you don't have to worry about any, you know, disasters out there.

So with that, I'll open up for question and answers.

Are you going to invest in the long term on both solutions, or are you just going to focus on one solution?

So good question.

The question is, are we going to invest on one solution or both solution long term?

And the answer is we are going to invest in both the solutions.

As you can see, both of them have their own value propositions.

They are the swim lanes.

We are investing in Metro Cluster IP with the cloud mediator HTTP and, you know, switch lists, you know, configuration to help reduce your cost.

And then on the SnapMirror ActiveSync side, right, we are growing on the workload side.

We have SAN, then NAS, and then kind of bringing it with, fiber channel as a replication.

So, short answers, both of them are, you know, kind of complementary to each other.

They serve different use case, and we'll continue to invest in them.

S3 support for ActiveSync, is it on the roadmap?

It is further down.

Sorry, sorry.

The question is the S3 support for SnapMirror ActiveSync.

It is further down on the roadmap.

Right now, our focus is on scale and security, so we want to finish those journeys with, you know, expanded scale and ARP and TPS protection.

Yes, that's something we are

And also, we are also focusing on the ecosystem, so we are integrating with Veeam, you know, as you saw on one of the slides.

So we want to finish those before we further expand on the workloads.

Yeah.

Any plans for S3 and Active?

Sorry?

Any plans for S3 and Active?

So I think it's the same question, like, S3 support, right?

Yeah, S3 support and also about if it's ActiveSync Metro Cluster

What-

Are you able to use Active for those websites?

Oh, for S, SMAS, NAS for Active-Active, is that one?

For S3.

For S3.

S3.

Not in the near term, I would say.

Any other questions?

There's some there.

Yep.

Do you plan to provide HA to install the mediator?

Could you just repeat that?

Sorry.

To install the mediator, you need the VN radar and so on.

Do you plan to provide an OVA like you had for for mediator OVA?

Yeah.

Yes.

So the question is around the mediator OVA.

You wanna answer it?

So, so we introduced that, HTTPS-based mediator communication.

That's where we've been hear from our most of the customers saying that one, that solves the problem of the SCSD ka- kernel modules overhead.

So I, I do hear that OVA kind of, ask over there, but the roadmap is more on the side of the on-prem cloud-native mediator support, which we have been planning to bring it in the next cycle, right?

So that also will help you to scale up as well as scale down based on your, the Metr- Metro Cluster or the SnapMirror ActiveSync instances, which is going to be there, as well as the HA also will come as part of that one.

So the OVA is down the line on the roadmap, but I would say that one is in the backlog.

Any further question?

I think we'll just take the last question.

Yeah.

I saw on the roadmap that, Metro Cluster tenant capacity management.

Can you tell us when that's coming?

So the question is on Metro Cluster tenant capacity management-

Right SPM quota.

Yeah.

So that's been, getting, released or we have been targeting that one for nine twenty dot one, right?

So that's a part of the roadmap, and we have been getting delivered for the next release.

So it's like limit on the SPM quota.

Right.

Yes.

Exactly.

Okay.

Yeah.

So I think we are already over time, and we'll happy to answer questions, you know, after the, right after the session.

But just quick related resources.

For additional information, you can reach out to me, Senthil, or Eric, you know, whenever you have more questions.

And thank you, everyone, for joining this session.

Thank you.

And, and please, provide your surveys and feedback so that it'll help us know, and you can earn your points as well.

Thank you, and have a wonderful evening.

Yeah.