Welcome to the Episode 383, part of the continuing series called “Behind the Scenes of the NetApp Tech ONTAP Podcast.”
Databases traditionally are run in on-premises data centers, but as they scale larger and larger and require more performance, admins need to start evaluating whether to purchase new hardware to manage or move to a lease-based model in the cloud.
Amazon FSxN for NetApp ONTAP offers a way for database adminstrators to confidently move their workloads to the cloud without sacrificing performance or the feature-richness of NetApp ONTAP.
Semion Mazor (semion.mazor@netapp.com) and Pedro Fernandes (pedro.fernandes@netapp.com) join us to discuss how Amazon FSxN for NetApp ONTAP improves the life of database administrators considering a move to the cloud.
For more information:
Finding the Podcast
You can find this week’s episode here:
I’ve also resurrected the YouTube playlist. Now, YouTube has a new podcast feature that uses RSS. Trying it out…
You can find this week’s episode here in the RSS feed:
You can also find the Tech ONTAP Podcast on:
I also recently got asked how to leverage RSS for the podcast. You can do that here:
http://feeds.soundcloud.com/users/soundcloud:users:164421460/sounds.rss
Transcription
The following transcript was generated using Descript’s speech to text service and then further edited. As it is AI generated, YMMV.
Tech ONTAP Podcast Episode 383 – Using Databases on Amazon FSxN for ONTAP
===
Justin Parisi: This week on the Tech ONTAP podcast, we talk about databases in the cloud using Amazon FSx for NetApp ONTAP.
Podcast Intro/Outro: [Intro]
Justin Parisi: Hello and welcome to the Tech ONTAP podcast. My name is Justin Parisi. I’m here in the baSemiont of my house and with me today, I have a couple of special guests to talk to us all about databases in the cloud. To do that, we have Pedro Fernandes. So Pedro, what do you do and how do we reach you?
Pedro Fernandes: Hello, Justin. Welcome to your baSemiont. My name is Pedro Fernandes and you may reach me via email pedro.fernandes@netapp.com. So my role at the moment is cloud solutions architects and I’m helping customers on cloud journey, and I’m optimizing workloads in terms of performance and cost optimization.
Justin Parisi: All right. Also with us today, we have Semion Mazor. So Semion, what do you do and how do we reach you?
Semion Mazor: Hi, Justin. Thank you for having me. So, yeah, I’m Semion Mazor. I’m a product evangelist here at NetApp in the Cloud Innovation Center, and we focus on several workloads for FSx for ONTAP. My role is to let people know about the different things they can do with this great service.
Justin Parisi: And we’ll get to that in a little bit. But today we’re here specifically to talk about databases and databases in the cloud. So let’s kick that off with traditional database workloads and how they’ve always been on-prem. Why are people moving those workloads to the cloud?
Pedro Fernandes: Most of the organizations are taking those databases because there are other services related that can be integrated on the workloads, right? So in that way, some customers are giving up regarding the hurdles that they are having with on-prem databases in terms of cost, in terms of management. Some of them are using managed services. Some of them are using self managed databases, and they are replicating what they have in on-prem systems.
Justin Parisi: So you mentioned cost. What sort of costs are greater in on-prem versus the cloud? Because a lot of people think about cloud and they think, oh, it’s expensive. But, in reality, there’s a cost associated with on-prem as well. So can you break down why a database might cost more to do on-prem than in the cloud?
Semion Mazor: I would like to add ONTAP of what Pedro said that databases are essential and crucial part of the organizational IT. And now the question about the cloud is not there anymore. Organization are moving to the cloud and adopting it for several reasons. It’s for the scalability that provides the simplicity of operation moving from CapEx to OpEx. And the ability to your specific question is the ability to choose exactly what you want on a specific time. And the agility that you can get from the cloud is one of the major reasons for organization to move to the cloud and be able to adapt the need and pay for what you actually need and the time that you need it, and not waiting for shipping time until you get the hardware installed and configure and everything.
Justin Parisi: So Pedro, you talked about on-prem versus the cloud and people are moving from on-prem to the cloud because of cost concerns. What sort of costs are associated with on-prem that you don’t necessarily see in the cloud?
Pedro Fernandes: I mean, it’s not only cost related, right? It’s also the topic that Sam and talk about is, which is the flexibility and also to construct a high availability database infrastructure, it’s easier to do it in cloud. In a matter of seconds, you can launch and multi AZ.
You can have replication across regions. It’s not only about costs, but also flexibility and agility on that type workloads, right? And in this case, typically this work goes on on-prem. They are very static. They are very fixed. And you need to upfront size the database or size your storage.
And when you move to the cloud, those costs are flexible. You can just increase when you want, when you need. So in that way I think the cost is a topic, but I think it’s also other topics to consider on this.
Justin Parisi: Well, it’s not just the cost of the storage and the compute itself.
It’s also the cost of the man hours, the data center, right? So you’re saving that money when you lease the equipment versus basically owning it. And, essentially a cloud is a lease, right? You’re leasing space. You’re leasing a part of the data center in the cloud to give your databases the capacity they need and the compute they need.
So that’s really, I think, where the main cost savings come from, in my opinion.
Pedro Fernandes: It’s the lease. Yeah. It’s the lease also. Yeah.
Semion Mazor: I thought you didn’t know the answer from the beginning.
Justin Parisi: I did, and you guys weren’t getting there.
Semion Mazor: So glad to have you here.
Pedro Fernandes: So glad to have you here, Justin.
Justin Parisi: Yeah, like I asked it and Semion didn’t answer like, damn it, Semion, answer the question like a politician. All right. So, I heard the words managed in self managed databases and those words have a lot of weight in a lot of cases, especially when you talk about cloud, because when you say self managed versus managed in the cloud, it’s entirely different meaning.
So. Tell me about what managed and self-managed is with concern to databases.
Semion Mazor: Yeah, so when somebody comes to architect their databases on the cloud, in particular on AWS, but basically on each cloud, they have two options, and not only for databases, but we’re here talking specifically on databases.
They have a fully managed service, and in AWS, there’s a bunch of different services for different databases. And basically, when you choose that option, you get it all managed. You don’t need to carry with anything. You’re just using it without caring for the underlying infrastructure. The other option is with Self managed.
In this case, you’re using a compute again. If it’s AWS, Amazon EC twos and a storage service coupled with that, and you manage itself. So there are different consideration choosing between those two options. So for managed service again, it’s managed infrastructure. The scalability is elastic.
You pay only for what you use. And if you know the capacity that you need and it fixed so you can also reserve instances for the databases. With self managed, you can make more control. You need to manage it by yourself, but you can find in the configuration and support any kind of databases because you install it by yourself and not only the supported version that the managed service have. Also databases requires licenses.
So with the self managed license, you can invoke the same license you already have on-premises, for example, and use it in the cloud, and there’s no limitation on instances. So it depends on the situation and preferences of each organization. And both ways are valid with different consideration for each of them.
Justin Parisi: So essentially, you’re talking about databases as a service versus just bringing your own database into the cloud.
Semion Mazor: Yeah. Yeah.
Justin Parisi: Yeah. All right. So it isn’t that much different than normal cloud speak then. So it’s managed versus self managed. Same thing.
Semion Mazor: I think it’s also a matter of the psychology aspects of how much control do you want to have over your environment and from one hand, deal with that, and how much you have the expertise to do that, versus how much do you want to getting the result of it without caring about what’s happening behind the scenes.
Justin Parisi: So with databases being on-prem most of the time, there are some challenges that take place when you’re trying to move to the cloud. So let’s talk about those challenges and what sort of things people face when they’re trying to move their databases.
Semion Mazor: So I think there’s the aspect of how you move from on-premises. And one of the important aspects of that is how you build it on the cloud. So there’s a team of 80 percent expert that used to have it in specific way, and then they need to architect it.
In a new environment, there’s several challenges that’s coming with that. And first for databases, maybe the most important aspect is performance. How can I get the low latencies that I need for the application that the database serves? So this is first question. How I architect it together the highest performance with the minimal latencies. Another challenge is availability. How can one ensure that the data is always available, even in case of failure or even disaster? The third challenge is data protection. How do I make sure that now the data doesn’t get lost?
So aspects of snapshots and backup and also disaster recovery can be considered here. The other aspect is creation. Databases in many cases require to copy them and create new environment. So in many cases raised question about how long it takes and how costly it is and how complex it is.
So all the challenges together comes with the bottom line. Also, How much does it going to cost? What is the bottom line for that? And how complex the architecture to manage all those together? So I think those are the challenges that usually comes when architecting databases on the cloud.
It’s performance, availability, data protection, cloning and cost. This is the main challenges that will face from customers.
Justin Parisi: Sounds like a lot of those challenges are the ones you run into on-prem as well. I mean, I’m not seeing anything specific there.
Semion Mazor: I think that the challenge here is, it’s nothing new, just different. And the challenge is how you architect it now to get the same performance you used to have on-premise now when it’s not fully managed by yourself and you have the full control, but in a new environment, like a new ecosystem.
Pedro Fernandes: Yeah, added to that, I think the challenge or the hurdles are mostly like the same that we faced on on-prem. But now you have a bunch of storage services that you need to discover that you need to know in order to select the most appropriate one. So in that case, I think it’s another task that organizations need to take care of.
Justin Parisi: So we have a set of challenges here. There’s always gotta be a solution right? So, tell me how FSxN solves these challenges. What does it do to help the performance, help the availability, help the data protection, help the cloning?
Pedro Fernandes: The name of the service is, Amazon FSx for NetApp ONTAP, right? So basically what we are bringing to the cloud environments in specific AWS, in this case, it’s our well beloved ONTAP with their enterprise great features, right?
With thin provisioning, cloning, snapshots, block level replication with our well known SnapMirror. So those technologies or those features are now available for customers that are asking for the same features that they have in on-prem systems.
Now with FSxN, they can leverage the same type of architectures that they are running now on on-prem. So this will facilitate a lot on the DBAs, on the infra guys, the choice.
Justin Parisi: So let’s address each topic individually. Let’s talk about the performance first. That’s top of mind when you deal with databases. So how does FSxN handle the performance, handle the latency requirements for these databases?
Pedro Fernandes: In FSxN it’s a file system that has multi protocol.
And in this case, we are leveraging mostly block storage protocol, iSCSI, in order to connect the FSxN to the EC2 SQL agents. But we also leverage NFS for Oracle databases. So we can provide sub millisecond latencies, and high performance on this type of workload. So, it’s a really good choice for these workloads.
Semion Mazor: Yeah, I would like to add that with FSx for ONTAP, it can ensure low latency. So latency is dependent on many parameters, but for the aspects of the storage, it will make sure that there’s low latency from this end with millions of IOPS and dozens of gigabytes of throughput and SSD based storage.
So all of that will guarantee that the databases get the performance that it needs. With also great capabilities that even enhance it more for the frequently accessed data with in memory cache and NVMe cache that boost the performance further.
Justin Parisi: So what about availability? Cause that’s important for databases as well. We want to make sure they stay online. So how does FSxN for ONTAP deal with availability?
Pedro Fernandes: So FSxN has two options – scale up or scale out. Here we are mainly speaking about scale up. When you initiate the service, you will have one preferred FSxN, which is called the active one, and you have the passive one, right?
So in that way we can have a single AZ or a multi AZ. It’s just a matter of where are the nodes of the service. So in that way we can provide high availability on single AZ or multi AZ.
Semion Mazor: It’s important to mention that FSx for ONTAP is highly available in any case There’s no other option to have it. So the choice is between having that in the same Availability zone or multi availability zones.
Justin Parisi: Yeah, and that’s good because it protects you against cloud outages if a zone goes down, you have the ability to failover to another zone.
Semion Mazor: Yeah. Exactly.
Justin Parisi: So, if a zone does go down, let’s say it happens on the AWS side, does the storage automatically failover on its own? Like, if it’s a network issue or something that’s not storage related?
Pedro Fernandes: We have our own health systems to double check that. And if that happens, the passive one will become the active one without any disruption on the database side. So in that way, Customers are leveraging an enterprise service that will enable them to create some high levels of resiliency.
Justin Parisi: Okay. So, it’ll automatically do that for you there. I guess you said there’s some monitoring in place. Is it kind of like a watchdog that keeps track of connectivity?
Pedro Fernandes: Yes, because this is a cluster, right? So a health check is made by AWS. This service is AWS service, right? They monitor this and you can also leverage the Cloud Watch logs or alarms in order to see performance, in order to see capacity. So we shouldn’t expect any outage on this.
Justin Parisi: All right. So failover is one part of availability. Another part of availability is disaster recovery and that ties into data protection. So what does FSxN for ONTAP offer for the data protection side of things?
Semion Mazor: Yeah, so there’s several layers of data protection that FSxN for ONTAP provides. So first is snapshot, and we’re talking about NetApp snapshots, which are very lightweight and very efficient because the way they were built. So this is the first layer of protection. Those snapshots are also application aware, so they know how to interact with the database and make it to where there is now a snapshot and therefore they will be recovered if there is a need for that.
So it’s instant restoration of the database data. And this is the first layer. The second layer is having disaster recovery create another site for the storage in which the data will be replicated and synchronized and in case of disaster, the database will get access to the other site. And the third layer of protection coming with backup, which is again, utilizing the efficient snapshot and therefore both the backup and restoration time is very short and very quick and also saves money.
So with those three layers of protection FSx for ONTAP can provide up to one minute of RPO and minimal RTO. And again, it’s as Pedro said at the beginning, it utilizes unique capability within AWS, which is called cross region replication, which is utilizing SnapMirror under the hood. So all this together provides great protection for the databases.
Justin Parisi: So with disaster recovery strategies, there’s this concept of 3-2-1 for backups. How does 3-2-1 fit into a cloud infrastructure? Because there’s that whole concept of the offsite backup. So what are we doing there?
Semion Mazor: I think for databases, it’s less common. There are different architecture that you get for that.
For example, if we’re talking about SQL, it’s more the availability group and FCI architecture. So this is some way to make sure the database is always available. And then the backup is just that you can have another copy of the database if you need to restore that.
So just to explain what I said, the 3-2-1 is three copies of the data in two different formats. And one of them at least is remote. So it can also help with that.
Justin Parisi: And I know that with a lot of database products, especially with ONTAP, there’s external software to help with the transaction log replays and If you’re using SAN to negotiate the whole file system consistency piece.
So what do we have in AWS FSxN for NetApp to do that sort of thing to be able to negotiate those challenging pieces of database infrastructure?
Semion Mazor: Yeah, that’s a great point, because usually when you will talk with the DBA, database administrator, he will not think about the storage as a DR site.
He will do it on the database level which makes sense, right? They’re managing the databases and there’s different tools, both native for the database itself or third party tools to make sure you have several copies of the data and that they are synchronized and that you have backup for that. And that’s what’s usually done. It provides great capabilities. I think what’s nice with using FSx for ONTAP for this level of data protection. So what we’re talking about is actually moving the data production from the database layer to the storage layer is that it becomes seamless for them.
So they have storage for that. And the storage takes care that it’s always available. And even in different region when it’s synchronized. So for them, it’s become from one hand seamless. From other hand, it can be much, much more cost efficient because this way you can use the lower licenses of the database.
So it’s also a very nice advantage. It kind of a hack when you are architecting on on AWS.
Justin Parisi: Don’t say it too loud. They might catch on, the database guys.
Semion Mazor: I think that they are in favor of that because as I know, AWS they’re looking for what is better for the customer.
Justin Parisi: Yeah, yeah, absolutely.
Pedro Fernandes: So in this case, we have also the step center, the NetApp SnapCenter, which is, I call it a Snapshot Orchestrator, which is our backup tool for this kind of workloads. It’s free, it’s available for FSxN. And what it does is like Semion explained, it interacts with the SQL engine, right? And it says just, okay, what do you have in memory or in transit? Please flush it down to the storage. And then the SnapCenter will take an application consistent snapshot.
And in that way, it can also interact with SnapMirror. That’s why I call it a snap orchestrator, right? Because it can orchestrate SnapMirror to trigger a SnapMirror update replicating the last backup into another region or into another AZ where you have an FSxN running in a cost effective way, leverage, for example the data tier option here.
Justin Parisi: Are you talking about FabricPool?
Pedro Fernandes: Actually, it’s a subset of the FabricPool. Data tiering is the capability that we have to flush the old blocks into a capacity pool. And those blocks should be cold, right? So that’s the capability of data tiering.
Justin Parisi: I was thinking there was like service levels where you could replicate, say, let’s see how a performance level versus like a data level or a standard level, right?
Pedro Fernandes: We have performance levels. We have three or four different performance levels when it comes to scale up or even more when we speak about scale out, but this feature, this data tiering option is all about moving blocks, old blocks that will reside on SSD tier onto the capacity tier.
FSxN has two tiers. The performance tier and the capacity pool. And we can move blocks under the hoods between these two tiers. Okay. This is the data tiering option.
Semion Mazor: You touch an interesting point, which we’re planned to discuss when we’re talking about cost performance, but you’re already ask about it.
I think the most amazing thing about FSx for ONTAP when we’re talking about databases, the cost performance it provides in the ability to reduce dramatically the total cost of ownership. In the context of performance, you can decide on what level of throughput and IOPS and decide it per environment. So they’re able to configure those performance can save costs. And the other thing is that comparing to alternative and using other block services, you would need much more instances and compute power to get the same performance.
And because of how FSx for ONTAP is built, It enables to use much less compute power to get the same performance because of the efficiencies and the caching and the different things it has inherently. So the price performance that FSx for ONTAP provide for databases is just amazing.
Justin Parisi: So with the performance levels, if let’s say I set up a performance level at the highest level and I find that’s probably too much. I want to decrease it or I find later on that I need to increase it. Do you offer that in real time? Is that something you can do on the fly? Or do you have to take an outage to increase the service levels?
Or do you have to change the volumes entirely? How does that work?
Pedro Fernandes: You can change it. You just go on the AWS console and change it to the throughput that you want on the fly, of course, without any disruption and you will have the performance that you want. You don’t need to change anything from the volumes perspective.
If you want, you can also apply QoS policies on the volumes for a specific database. Let’s say that the database is a bully database that is consuming so much IOs that you need to cut it out. You can create a QoS policy, apply to that volume, and then The other databases or the other volumes won’t be affected by that bully one.
Justin Parisi: So that’s QoS ONTAP of changing the service level. Is that correct?
Pedro Fernandes: Sure. Yeah, you can change it. You have that capability.
Justin Parisi: It’s interesting because I thought the service levels were going to be QoS based, but I guess they’re not.
Pedro Fernandes: They are not. You have the service levels on QoS.
You still have it as you have on ONTAP on-prem systems, right? But you still have the possibility to scale up via AWS. You can define your own IOP level. If you want, you can change the throughput up or down, doesn’t matter. The only thing that you need to be aware is that the SSD performance tier, because that only increases.
Justin Parisi: So if you’re on the SSD performance tier, you can’t decrease it, you can only go up.
Pedro Fernandes: You can only go up.
Justin Parisi: Okay, gotcha. So as far as performance for databases go, sometimes end users don’t know how much they’re going to need before they set it up. Do you have any sort of calculators or any sort of aids to help you figure out the performance that you’re going to need? Or you just got to try to wing it and figure it out yourself?
Pedro Fernandes: I mean if we are talking to a DBA or a storage administrator, probably they know they have the historic results, they have the historic reports, right? So probably they know what they need, but the cool thing about this service is all the key points on the CloudWatch.
So you can monitor what you are getting from the FSxN. And also you can scale up when you want. So probably you will start smaller, and again back into the topic of the migration, right? Every data that you insert on FSxN will be deduplicated. We will have compression, we will have compaction. So probably if you have 10 terabytes on on-prem systems, probably you don’t need 10 terabytes on FSxN because the data will suffer those transformations. So you can always monitor the consumption, the capacity via CloudWatch or via AWS console, or even NetApp Harvest, which is another monitoring tool that you can use to do this. And you can adapt the service. It’s flexible. It’s agile. That’s the key point here on this type of services.
Justin Parisi: So, we touched on the cost and how FSxN for ONTAP can help with that. One of the biggest, I think, advantages to ONTAP and databases, and this has been for years, traditionally, is the FlexClone, right? The ability to clone the databases without taking any space. And this is done for a variety of reasons.
Sometimes it’s done for verification of the databases without having to actually run it on the database itself. So tell me more about the FlexClone advantage that we have that other companies and competitors might not have.
Semion Mazor: Yeah, so if you’re looking at it in a perspective of on-prem to cloud, I would just say you have what you used to have.
Like if you have ONTAP and you’re already utilizing FlexClone to clone that data, you have it all on FSx for ONTAP. But what does it mean? So usually in order to create new environments and any organization needs several environments of database, just as you said, Justin.
So whether it’s for dev/test environments, schema validation, QA, whatever it is. They need several environments, and each environment needs to have access to data, usually the same data that it has on the primary on the production database, but without access to the production databases.
So with a FlexClone, you can copy the data, give the databases access to the relevant data. Without actually copying the entire data, just use a snapshot, which is done instantly without almost zero effect to the environment and cloned instantly to give access to the data to a new environment.
And this also applies when you need to refresh data when data is changes. And those clones are also writable. So the database can write for them and act as a full copy of the data. The cool things when we’re talking about cloud is besides that it shortens the development cycle because those copy now can be created momentarily is the huge savings that it’s done. So one is the time that it consumes. But the other benefit is the cost reduction because you don’t need to pay for capacity because it takes almost zero capacity to create new environment and you don’t need to copy the data itself. It’s done instantly. So the capacity of the storage doesn’t cost almost at all. And also in some cases there’s other resource that waiting for those environments. Think about EC2 instances. They’re waiting for data refresh to run their operation for that. So you need to pay for those instances.
They’re now just waiting for data to be cloned, to be replicated. With FSx for ONTAP and the thin provision capability, they don’t need to wait for that because it’s done instantly.
Justin Parisi: You mentioned it’s read writable. That’s when the space starts to take shape. When you take the initial clone, it’s virtually no space taken up.
But as you write to the clone, if you decide you want to make changes to the clone, that’ll actually add some capacity reflected in the changes in the blocks. What’s also great about FlexClone and I don’t think you touched on this is, let’s say you want to keep those changes. Well, guess what?
You can split that clone off and it’s a brand new file system separate from the original database.
Pedro Fernandes: Yep, that’s correct. That’s correct. So we can leverage FlexClone, which is basically taking advantage of the parent volume, right? We are using the parent blocks to read, but when it comes to write, there is some space consumption.
So in that case, if you want to test something, a FlexClone is a good idea because then you have two options. In the end of the test, you can just drop the clone. Or you can just split. Split, it’s a under the hoods operation that you are copying the entire parent volume, the blocks from the entire parent volume into this new volume.
So in that way, we can put this volume into the production because it has all the blocks, it has all the information that you need.
Justin Parisi: It’s actually a bit faster than trying to do a protocol copy as well. So if you’re trying to copy it over like SMB or NFS, it’s way faster just to split a clone.
Pedro Fernandes: Way faster.
And it’s always available. You don’t need to wait. It’s already on production. You just go there. You can go on the SnapCenter, right? And you just split the volume. And the operation will be working behind the scenes, but you already have your database working. So in that way, it’s a really efficient way to do the test, the clone, and put the things on production.
Semion Mazor: And in terms of capacity, usually in some cases, you need dozens or even hundreds of different environments. So let’s say I have one terabyte of data and then it’s multiplied by five, ten or hundreds of environment. But with FSx for ONTAP, the first initial copy is smaller because we talked about storage efficiencies and tiering and this kind of stuff. So from the first place, it’s not the same a hundred percent. And this is a game changer.
Justin Parisi: Yeah. FlexClones are definitely the bread and butter, I think, when you talk about databases and ONTAP. I think that’s really kind of what sells people on it.
Pedro Fernandes: Snapshots, reducing also the SLAs on the backups. It’s also another thing. Because DBAs are usually getting with the traditional backup systems that takes hours. And here we are leveraging those backups into snapshots. So the snapshot is really quick, really fast. You still have the application consistency. And you have block level replication into another region or to another FSxN in another AZ.
So it’s another feature that we like it and customers also love it.
Justin Parisi: All right. So, we’ve got a lot of things to work with here when we’re talking about databases and the cloud FSxN sounds like a good opportunity for people to use what they are used to on-prem and bringing it into that cloud. Now, when we talk about the cost savings, we talk about a lot of these features.
Can you tell me a little bit more about what we’re actually going to see in terms of cost efficiency?
Semion Mazor: Yeah, so it’s summarizes the thing we discussed so far. So we can get up to 50 percent of cost reduction when we’re talking about databases. And. It’s due to the all things we mentioned.
So it’s starting from shrinking the data with thin clones and thin provisioning and storage efficiencies and the optimized snapshot. Then you can use fewer EC2 cores when you’re using FSx for ONTAP, comparing to the other alternatives, as a result, optimizing both compute databases and licensing expenses.
So all that together getting to huge savings and summarizing the things we discussed so far and the different features. But in many cases, this is the first thing that organizations, when they’re looking to optimize the operation in the cloud and trying to understand how they’re getting better performance without doubling the expenses on their compute cores.
And then somebody says, Hey, you should check FSx for ONTAP. So in many cases, it’s kind of the foot in the door to check what is FSx for ONTAP for those who is not familiar with ONTAP, and then they figure out there’s a bunch of other great features that can be provided. So by itself, I think it’s a good reason to check it out to see the cost reduction, the cost optimization that it can provide for databases.
Pedro Fernandes: Just want to talk about the cost optimization, because when we are using FSxN you are getting not only cost optimization, but also operational and performance optimization, right? Because we are leveraging the enterprise great features that we already talked about. So the snapshotting, the FlexClone, also the SnapCenter, a tool, which is free, you can use it on top of FSxN, right?
So I think you need to look at the landscape, and you will find so much benefits using FSxN because enterprise grade features, like FlexClones, block level replication, asynchronous replication to other regions. Those are some capabilities that you won’t find in any other service.
Justin Parisi: All right. So as far as more information, like if I wanted to read about this or figure out where to find how to use FSxN or actually how to get to FSxN or any of that, where would I go for that?
Semion Mazor: Yeah, so FSx for ONTAP is an AWS service, so the easiest way to do that is just Google. FSx for ONTAP and get to the AWS page for that and try it out.
I think this is the easiest way. There’s also a bunch of documentation and explanation.
Justin Parisi: Alright, so again, Semion, if we wanted to reach you, how do we do that?
Semion Mazor: Yeah, so you can reach me by my email semion@netapp.com, S-E-M-I-O-N, at netapp.com or on LinkedIn, Semion Mazor.
Justin Parisi: All right. And, Pedro.
Pedro Fernandes: Same for me. Pedro.Fernandes@netapp.com, or you can search me out on LinkedIn, Pedro Fernandes, cloud Solution Architect.
Justin Parisi: Again, thank you for joining us and talking to us all about FSxN for ONTAP and how it works with databases.
All right, that music tells me it’s time to go. If you’d like to get in touch with us, send us an email to podcast@netapp.com, or send us a tweet @NetApp.
As always, if you’d like to subscribe, find us on iTunes, Spotify, Google Play, SoundCloud, Stitcher, or via techontappodcast.com. If you liked the show today, leave us a review. On behalf of the entire Tech ONTAP Podcast team, I’d like to thank Semion Mazor and Pedro Fernandes for joining us today.
As always, thanks for listening.
Podcast Intro/Outro: [Outro]
Why Is The Internet Broken? 