AWS re:Invent 2022 - Achieve enterprise greatness with Amazon FSx for NetApp ONTAP (PRT230)

Aug 16, 2023

AWS re:Invent 2022 - Achieve enterprise greatness with Amazon FSx for NetApp ONTAP (PRT230)

Amazon FSx for NetApp ONTAP is a fully managed AWS service that brings the enterprise-grade capabilities of the NetApp ONTAP storage operating system to the AWS Cloud. In this session, learn about recent enhancements and integrations since launch, including SAP HANA certification and VMware Cloud on AWS integration, and find out how NetApp’s unified management platform and unique capabilities can enhance your FSx for ONTAP experience. Finally, see a demo of how to build, protect, manage, and govern your data estate in a hybrid cloud environment using FSx for ONTAP and NetApp’s unified control plane. This presentation is brought to you by NetApp, an AWS Partner.

Learn more about AWS re:Invent at https://go.aws/3ikK4dD.

Subscribe:
More AWS videos http://bit.ly/2O3zS75
More AWS events videos http://bit.ly/316g9t4

ABOUT AWS
Amazon Web Services (AWS) hosts events, both online and in-person, bringing the cloud computing community together to connect, collaborate, and learn from AWS experts.

AWS is the world’s most comprehensive and broadly adopted cloud platform, offering over 200 fully featured services from data centers globally. Millions of customers—including the fastest-growing startups, largest enterprises, and leading government agencies—are using AWS to lower costs, become more agile, and innovate faster.

#reInvent2022 #AWSreInvent2022 #AWSEvents

Content

1.02 -> - Hi, so thank you everyone for joining us in this session.

5.76 -> My name is Gadi Oren, I'm with a product team at NetApp.

10.68 -> I have with me also Yuval Kalderon,

13.71 -> that works as a Cloud Solution Architect, also with NetApp

17.518 -> and Andrew Crudge from AWS with the product team of FSx.

26.04 -> So today, (indistinct)

32.73 -> so our agenda for today is to cover really

38.01 -> top interesting use cases for FSx.

40.44 -> We'll start by speaking a little bit about the partnership

43.74 -> between AWS and NetApp.

46.26 -> We will cover some of the integrations

48.45 -> and how they came to be

50.73 -> and the majority of the time we'll spend talking about

54.9 -> top use cases and we'll touch also on the new BlueXP

61.17 -> that's launching and have even a little demo for you.

67.29 -> So a few words about NetApp.

70.59 -> NetApp is a fortune 500 company.

74.82 -> It's a global cloud-centric data software,

79.56 -> data-centric software company.

83.01 -> We have about 11,000 employees

89.73 -> and founded literally 30 years ago.

93.72 -> And the interesting point to mention here is that

98.303 -> we have a really thriving, thriving network

101.85 -> of four, over 4,000 partners that's really

106.62 -> helping our customers to be successful.

110.22 -> The company started from the software domain

115.29 -> and we had leading from the storage domain

121.38 -> and we had leading hardware for many years.

124.8 -> In the recent years,

126.6 -> we have re implemented and reinvented the company

131.88 -> as a leading cloud data services company.

137.43 -> And we actually expanding from just a storage company

144.33 -> to also doing other things like computer,

150 -> computer services and other areas through organic growth

155.19 -> and through mergers and acquisitions.

158.25 -> Some of you may have heard about the spot acquisition

161.4 -> and other companies that we are purchasing.

167.19 -> So let's talk a little bit about the journey.

171.69 -> The journey that of NetApp and AWS as a partnership.

177.54 -> So the first time that we partnered with AWS was around 2012

182.34 -> where we started working in the cloud.

185.04 -> It's worth mentioning that in 2014

188.58 -> we had the first cloud volume,

192.78 -> the cloud volume offering that came out to the cloud.

196.59 -> At that time it was more of a self-manage storage,

200.64 -> which was completely detached from the hardware

204.54 -> where it started.

206.58 -> Fast forward around the, we had many, many notes

212.25 -> and innovations that were taken to the market

216.39 -> around the end of 2020.

218.64 -> We have signed an agreement with AWS to take

222.48 -> our technologies and package them into a service

227.73 -> that will become later FSx for on up.

232.83 -> We launched a product in September of 2021

237.75 -> and it's really taking off very well.

243.96 -> At that point, it would be really interesting to hand off

246.69 -> to Andy that will talk a little bit about, you know,

249.84 -> what was the point of view of AWS

253.59 -> going into this partnership.

256.98 -> - Thank you.

259.08 -> Everyone, as Scott mentioned, my name's Andy Crudge.

261.66 -> I lead the product management team for the FSx

263.92 -> ONTAP service on the AWS side.

266.562 -> And, you know, wanna give you some background about

269.37 -> why we launched the service.

270.81 -> And you know, what it comes down to at the end of the day is

274.08 -> we see file workloads on structured data pervasively

277.444 -> whenever we talk to customers.

279.63 -> It's a lot of data out there, a lot of workloads out there

282.36 -> that rely on shared file storage.

284.58 -> And not only is this market huge

286.71 -> and it's growing over time as well

289.11 -> and we're seeing customers commonly use file storage

291.75 -> to power some of the most business critical applications.

294.93 -> And as part of, you know,

296.88 -> a lot of customers that we talk to historically

299.37 -> have been on prem

300.203 -> and are looking to migrate or extend

302.16 -> or burst their workloads, their applications,

304.68 -> their data to the cloud.

306.57 -> And you know, what we've seen is that if you look at,

308.94 -> you know, on premises,

309.96 -> there's a lot of different file storage offerings out there,

313.02 -> options that customers choose.

314.34 -> And one of the most popular ones is ONTAP.

317.28 -> And you know, many of you here I'm sure

319.08 -> are familiar with ONTAP

320.88 -> and the rich data management capabilities that it provides.

324.84 -> You know, all the SnapMirrors, snapshots,

326.52 -> cloning capabilities that you've all got,

328.62 -> you're gonna talk about a little bit later.

330.72 -> The performance profiles, you know, the low latency,

332.67 -> the high IOPS that it delivers, as well as the reliability,

335.73 -> the high availability, durability, security that it provides

339.3 -> and for customers to migrate over to the cloud.

342.81 -> For a lot of customers,

343.83 -> they don't wanna have to re-architect completely upend their

346.71 -> storage solution as part of making that jump to the cloud.

349.98 -> And so with FSx ONTAP,

351.63 -> what our goal is, is to provide customers

354.173 -> with the capabilities, you know,

356.4 -> if you're an existing existing NetApp customer

357.81 -> to provide a familiar experience where you have

360.99 -> a stored solution that offers all the best that ONTAP

363.69 -> has to offer in terms of data management features

366.12 -> and capabilities.

367.86 -> And also for customers who aren't using ONTAP today,

370.53 -> again to provide those same rich features

374.22 -> in a simple agile cloud consumption model

378.12 -> as an Amazon service.

379.5 -> So our goal really with this service is

381.27 -> to provide the best of both worlds,

383.4 -> the capabilities that ONTAP provides with the simplicity,

387.03 -> agility, peace of mind, low TCO that customers have come

391.53 -> to love about AWS native services

393.99 -> and really to bridge these two worlds together

396.3 -> to give customers the value of these two capabilities.

401.79 -> So you know, I'm excited to share, we actually,

404.61 -> we've been iterating on the service since we launched it.

407.07 -> As Gadi mentioned,

407.903 -> we launched the service in September, 2021.

410.46 -> It's just over a year old

412.02 -> and we're excited to have launched, I think it's, you know,

416.37 -> nearly 20 capabilities between now, between launch and now.

420.81 -> I'm excited to announce a few capabilities that we launch

423.581 -> just this week at re:Invent.

426.09 -> Some of these are new performance enhancements

427.92 -> for the service.

428.91 -> So one capability is just this week we've doubled

433.68 -> the maximum performance that FSx for ONTAP delivers.

437.31 -> Previously you could provision a file system with up to two

439.89 -> gigabytes per second of throughput capacity

442.02 -> and this week we've just doubled that

443.46 -> to four gigabytes per second of throughput capacity

445.92 -> and correspondingly have also doubled

447.63 -> the amount of IOPS that you can provision.

450.18 -> Similarly, when we first launched the service,

452.4 -> we offered a multi AZ deployment option.

454.68 -> This is a deployment option where your data is replicated

457.68 -> and highly available across multiple AWS availability zones.

461.34 -> And with that we offered an NVMe read cache

463.83 -> or flash cache that delivers even lower latencies

466.83 -> for actively red data and higher IOPS.

470.16 -> We're excited this week to also be announcing that we're

473.16 -> extending this NVMe read cache capability

476.16 -> to our single easy offering as well.

478.38 -> And this allows customers,

479.76 -> especially with IOPS heavy and read heavy workloads

483.24 -> like databases and eltron design automation

485.44 -> to drive up to six gigabytes per second of throughput

489 -> and hundreds of thousands of IOPS, you know,

492.03 -> to really exciting capabilities that are gonna significantly

494.79 -> raise the bar for customers in terms of how much performance

498.09 -> they can drive

499.11 -> and how much work and how much value

501.15 -> they can drive from their data.

504.24 -> We've also announced a few capabilities

506.46 -> to make it even easier to use the service.

508.77 -> I won't go into too many details with these,

510.63 -> but first is we've actually launched a new capability with

513.928 -> our multi easy deployment option where you can now access

517.83 -> new file systems you create from on-prem

520.62 -> without needing to configure any additional routing.

524.1 -> You know, long story short,

525.12 -> when we launched the service,

526.59 -> when you created a multi easy file system,

528.69 -> you'd need to go into your route tables

530.22 -> and trans the gateways and add one extra routing rule

532.77 -> to access your data and that's no longer, longer required.

535.53 -> We've heard customer feedback and you know,

537.36 -> customers are always looking for ways to simplify their

540.03 -> deployments and so we launched a new capability

542.67 -> just this past week that makes it even easier for you

544.86 -> to access your data using a transit gateway

547.29 -> from on-prem without needing to actually configure

549.614 -> any additional routing or any additional firewall rules.

553.92 -> We also now offer for new file systems nitro based

557.13 -> encryption and transit.

558.18 -> This is a capability AWS launched a few years back

561.39 -> that basically offers encryption between

563.55 -> supported EC2 instance types and your file systems

567.332 -> without needing to set up or manage encryption.

570.87 -> And we've always supported KBOs based encryption

573.06 -> for NFS and SMB traffic.

575.37 -> And this option delivers two real value propositions.

578.7 -> One is you get encryption automatically,

580.74 -> there's no setup required,

582.09 -> you don't need to join an identity store to get encryption.

585.45 -> And the second is it comes with zero performance impact.

588.24 -> It's built in to the hardware

590.79 -> so that you can get encryption,

592.23 -> meet your compliance needs without really needing

594.63 -> to do any additional work.

597.51 -> Lastly, we've launched a whole suite

599.85 -> of additional capabilities in our API and console.

602.88 -> For example, we now you can create data protection volumes

606.09 -> through the FSx console.

607.53 -> You can configure volume snapshot policies.

610.77 -> You know, there's a long list of capabilities we've launched

613.38 -> and, but our goal is with this week's launches

616.44 -> as well as over time,

617.97 -> we're always looking to raise the bar on several dimensions

620.73 -> including performance, including usability and simplicity,

623.82 -> just making it even easier for customers

626.13 -> to manage their data in the cloud.

629.79 -> So with that, again, I wanted to give a quick introduction

632.19 -> to the FSx ONTAP service,

633.6 -> share kind of the AWS perspective

635.19 -> that led to us launching the service.

636.93 -> And at this point I want to pass the mic

640.5 -> along to Gadi and Yuval who are gonna be double clicking on

643.17 -> some of the top use cases we're seeing with the service

645.9 -> and some of the ways in which customers are deriving

648.038 -> a lot of value and are delivering even more

650.79 -> performance using the service.

655.65 -> - Thank you Andy. That was insightful.

659.76 -> So I wanted to go over a few of the use cases that will

663.63 -> demonstrate trying to make it a bit easier,

667.26 -> 'cause there's a lot of data.

669.33 -> We also mark the slide so you can can follow along.

673.47 -> Essentially the use cases that we wanted to highlight is

677.64 -> first of all our VMware announcement and how you can

681.39 -> essentially create flexibility

684.84 -> and reduce your TCO with VMC and FSx.

690.06 -> How you can use FSx to extend or migrate data

694.56 -> from on premise into AWS.

698.16 -> What does FSx mean for SAP HANA and how it can help you

703.53 -> achieve a much better operating model

707.61 -> and help your organization work a lot better.

712.65 -> Similarly, how it can also optimize MSSQL

716.07 -> and Oracle deployment and what we can do in that area.

719.55 -> And finally going through how FSx has a unique offering

724.44 -> regarding Kubernetes, whether it's ROSA or EKS,

727.977 -> and we can help you guys achieve a lot more

730.92 -> with those workload.

732.99 -> And finishing off with a demo around BlueXP experience

736.797 -> and the capabilities we offer from the control plane.

741.96 -> So I'll start with VMware.

743.79 -> So essentially it's always good

746.52 -> to start with the best thing.

749.19 -> If you need a shared storage for VMware cloud,

751.59 -> essentially FSxn is the storage to help you achieve that.

758.4 -> It's, we've worked quite hard with VMware

761.04 -> to certify the solution

762.36 -> and offer a data store that can help you

766.35 -> decouple the storage from the compute.

769.5 -> But why would you want to do that?

772.448 -> Essentially when you look at the VMware environment,

777.3 -> there's three factors that essentially size your environment

781.59 -> without FSx.

782.94 -> So you have the course, the memory and the storage.

786.66 -> So if your storage requirement is a lot higher,

790.44 -> you do before FSxn you needed to make it equal essentially

796.68 -> on the computer memory,

798 -> making the TCO of such a solution

800.97 -> not really viable to migrate.

804.33 -> Essentially with FSxn,

806.13 -> you can take the storage aspect of the sizing

812.04 -> and decouple it living you with sizing the environment

816.18 -> according to the computer memory and allowing us

819.21 -> to optimize the storage for you

820.86 -> and make it a lot more palatable in terms of TCO.

825.72 -> And when we look at the solution itself and what we offer,

830.04 -> so essentially you still get a vSAN

834.18 -> within your VMC environment,

837.06 -> but you are able to add NFS data store that comes directly

841.53 -> from FSxn essentially as I mentioned,

845.04 -> it helps you to right size the solution.

847.77 -> The other thing which is really cool is that you can use

851.1 -> the different data management that we offer like snapshot

853.89 -> and other capabilities that Gadi will elaborate on

857.34 -> to further optimize the environment.

860.16 -> So one example would be using Snap Mirror

862.65 -> and protecting your data somewhere else.

865.47 -> So moving it from on-prem.

869.04 -> The other important aspect of VMC environment is obviously

874.71 -> the VMs himself have a shared storage requirement.

879.57 -> So it's really important to note that all of those

881.82 -> are also supported with VMware.

883.62 -> So if you are deploying, for example,

886.8 -> an SQL environment on VMware and you need shared storage

890.79 -> in order to realize the architecture,

893.61 -> you can definitely couple that with FSx

896.79 -> and gain the different capabilities

898.69 -> and have Ingress amount that supports any of the protocol,

902.91 -> NFS, SMB and iSCSI and obviously all the different benefit

907.17 -> in terms of data management that we provide.

913.23 -> So going back to TCO,

915.12 -> I had two slides that I wanted to show.

917.97 -> One slide is essentially giving you guys a quick

921.477 -> understanding of the amount of saving that you can reach.

925.32 -> So this example has 400 VMs

928.92 -> with around 150 terabyte of storage.

932.82 -> So essentially with decoupling the storage requirement

936.3 -> from the computer,

937.133 -> you're able to save around a million and a half

940.08 -> in terms of TCO over three years.

944.34 -> And obviously everyone has

946.23 -> a slightly different environment, right?

948.78 -> This doesn't fit everybody.

951.15 -> So what we did,

952.26 -> and I'll show you guys how to get to it in the demo section,

955.35 -> we created a really nice calculator that helps you plug in

958.83 -> what you need and it will size the environment

962.16 -> and also will show you what's the TCO that you're,

965.88 -> that you will save with connecting FSxn

970.17 -> into the environment.

973.17 -> With that, I'll pass it to Gadi to show you guys

975.54 -> how you can migrate and extend from.

978.45 -> - Excellent. - On-prem.

979.89 -> - Thank you. Well, yeah, so definitely VMware

983.16 -> is a very exciting use case and seeing how free companies

987.48 -> bring their best to the market and it's very advanced.

990.687 -> And I'm going to cover actually the fact that, you know,

996.42 -> using FSx is really the best way to migrate and extend from,

1001.67 -> we think from any storage,

1002.78 -> but specifically with ONTAP into the cloud.

1006.74 -> And in contrast to VMware, which is very advanced,

1011.21 -> this one is really important because

1013.34 -> it's really where most companies would start

1015.68 -> as they're moving to the cloud.

1016.79 -> So you see both companies in the beginning of the journey

1021.02 -> as well as companies that are well into the journey,

1024.17 -> they have different flavors of that.

1026.93 -> So when you're looking at extending or creating some sort of

1031.7 -> a backup ODI, you can categorize it roughly

1036.284 -> into three groups with a classic trade off.

1040.94 -> On the left you will find the most simple,

1043.722 -> arguably the cheapest one situation where

1050.33 -> you get to pay less but the RPO and LTO are more,

1057.35 -> are less favorable.

1058.183 -> But that's something that many companies start with,

1063.47 -> which is you can replicate the data, you can back it up,

1068.66 -> up to the cloud and you have redundancy.

1071.72 -> Now if at some point you actually wanna use that data,

1075.05 -> you want to go into production

1077.15 -> or you want to maybe do data mining

1079.04 -> or maybe you want to do a ODI,

1080.81 -> you'd have to restore the data,

1083.51 -> you'd have to install applications.

1085.37 -> So it will take some time, right?

1087.23 -> But as I said, it costs less.

1090.29 -> In the center you'll find a use case that is

1093.86 -> sort of a compromise in that situation.

1097.55 -> What you'll do is not only do you back up the data,

1100.43 -> but also you moving application active data, right?

1104.48 -> So the data that is currently being used by applications

1111.38 -> will move as is and will be ready to use.

1113.93 -> Applications are still not deployed on the other side.

1118.88 -> And in the event that you want to start using that,

1122.96 -> you still have to either deploy

1126.44 -> or restore the actual applications.

1130.37 -> It will take short, it will take less time,

1132.5 -> you will be more ready and cost a little bit more.

1135.38 -> And eventually on the right hand side, sort of a warm DR,

1139.4 -> you basically not only replicate all the data,

1141.83 -> both the active as well as backup,

1144.56 -> but you also have the applications install and ready to go.

1149.57 -> And in the event that you want to kickstart that

1153.569 -> you just have to reroute the communication

1157.04 -> and point the applications at the data.

1160.16 -> So there's really many advantages for all those use cases

1165.35 -> as far as ONTAP goes.

1169.07 -> ONTAP has very strong security hardening

1174.17 -> built in all those tools to back up

1176.32 -> and they're all built into the ONTAP operating system.

1182.15 -> You have very strong storage efficiency including tiering.

1186.41 -> So automatically data that is not being used or is cold

1191.15 -> is going to be tailed out to much cheaper medium

1195.372 -> and basically give you automatically the savings

1199.67 -> and resiliency, durability is at an enterprise level

1206.39 -> and those storage efficiencies that I mentioned,

1209.84 -> they exist and the source and they move with you

1213.14 -> into the cloud so you're not losing anything.

1217.07 -> Finally, you can enjoy what's the cloud capabilities

1223.01 -> by being able to scale up,

1224.81 -> scale down and basically leverage the existing resources

1231.89 -> and you can access the information for multiple protocols.

1235.34 -> So whether its SMB or NFS

1239.75 -> or even object store type of protocols.

1245.48 -> So I don't want to get too much into the weeds

1248.42 -> of the technology that enables those advantages.

1252.32 -> You can read through that,

1253.34 -> but I will mention two or three that's worth mentioning.

1257.24 -> So snapshot, snapshot is a very basic capability of ONTAP.

1263.27 -> Basically this is point driven copy

1268.43 -> of the data as it is now.

1270.74 -> And because ONTAP works with as is pointer based

1275.3 -> and block based, it is very, very fast to take a snapshot

1279.59 -> and it's very cost efficient to have snapshots.

1283.25 -> So a lot of the technology is based on that.

1286.1 -> Also worth mentioning SnapMirror.

1288.62 -> So SnapMirror is an extremely efficient protocol

1292.07 -> to move data between systems.

1296.39 -> The data will move,

1297.74 -> but only the Delta only what changed at the block level

1301.25 -> will be taken, packaged efficiently, compressed and moved.

1306.95 -> And so you can really deploy any configuration,

1311.36 -> any geography configuration

1313.37 -> that you need to achieve your goals.

1317.45 -> I think the last one I'll touch on is Snap Center.

1319.82 -> Snap Center is really an umbrella of services

1325.67 -> and agents that will allow you to take application

1332.12 -> consistent snapshot and backup, right?

1334.52 -> So many advanced applications have many of

1339.47 -> the advanced applications are stateful, you know,

1341.93 -> databases and similar and taking a snapshot of the data is

1346.49 -> not very meaningful if it's not done within the context

1352.16 -> of the application.

1353.6 -> And SnapVault will be able to do that

1357.26 -> in coordination with a specific application.

1360.89 -> So if we were to go and look at how each one of those

1365.39 -> three cases, three configuration would look like,

1367.97 -> this would be the first one where you running on site,

1372.26 -> you're taking a backup and that backup basically

1376.61 -> ends on the other side within FSx,

1382.88 -> some of that data will be over time cold and will

1385.61 -> go into the capacity too, which is cheaper in terms of cost.

1393.02 -> And you have the data ready right there

1395.99 -> for restoring if you want,

1399.485 -> you can tweak that into more advanced configurations

1402.65 -> where not only does the backup wait there,

1406.01 -> but you can also back it up in addition to AWS backup.

1412.1 -> And so if you want to operationalize that and go live,

1418.04 -> you have to install the applications,

1420.29 -> you have to install the data, right?

1422.3 -> The second configuration that we can see here,

1425.84 -> we're actually copying data that includes both backups

1429.08 -> as well as live data.

1430.76 -> And as I mentioned in that situation,

1433.37 -> if you want to go live,

1434.99 -> you have to deploy the applications

1438.23 -> and point them at the data.

1440.63 -> It will take you some time,

1441.95 -> but will be faster than the first one.

1444.92 -> Lastly, in that third situation, we're copying the data,

1450.47 -> we actually have the applications deployed ready to go.

1454.64 -> When you want to turn off the switch,

1457.76 -> you point applications to the data

1460.85 -> and you wheel out the communication and you're good to go.

1464.99 -> So I wanted to cover a use case of a real customer

1470.81 -> that took all those best practices and put them to use.

1476 -> So this is a company called Pearsons.

1479.96 -> The challenge that they have. that they had was that,

1483.44 -> they had a set going for critical applications on finance

1487.82 -> and HR and it was not working well for them.

1491.84 -> The RPO and RTO were really not great

1496.16 -> and the solution that they chose is

1499.07 -> they chose to utilize FSx for ONTAP.

1503.99 -> They used the snap mail, snapshots and flex clones

1509.81 -> to implement variation of the solutions I just covered.

1517.1 -> The outcome was really very successful.

1521.54 -> They migrated millions of files

1524.69 -> and they didn't have to re architect anything,

1527.3 -> they didn't have to re-platform,

1530.57 -> it was just to activate the existing capabilities

1535.1 -> and files started to run.

1537.92 -> The actual data refresh went 10 x in terms of speed because

1544.85 -> everything is much more efficient, more compressed

1547.94 -> and they dramatically reduced the RPO and RTO

1552.95 -> because of the previous points.

1557.022 -> And eventually what was, I think some of the more

1563.21 -> exciting thing was that they had a lot of people

1565.7 -> working on that solution. It was big, it was heavy.

1568.55 -> These people are now free to do higher value business,

1576.77 -> create better business value for the company.

1581 -> So as I said, this is a fairly basic capability.

1585.56 -> We see it with beginning,

1587.27 -> companies beginning their cloud journey

1589.22 -> and as well as companies that are well in.

1592.73 -> It's very good for migration and it's very good for backup

1596.96 -> and disaster recovery situations.

1601.22 -> Yuval, maybe you wanna take it from here

1603.59 -> and tell us a little bit more about what's next,

1606.17 -> what does customer can do

1608.45 -> after they have those capabilities?

1610.1 -> - For sure.

1610.94 -> So you can definitely use Snap Mirror for migration,

1614.09 -> disaster recovery and third site copy, like Gadi said,

1619.19 -> other capabilities that we have around

1622.01 -> not just replicating the data,

1623.6 -> we can also extend the data and create the cache out of it.

1628.04 -> So with Flex cache and global file cache,

1630.71 -> you are able to take that data

1632.57 -> and essentially create a cache of the data,

1635.33 -> whether it will be on premise to cloud

1637.55 -> or cloud to on premise depending on the use case.

1640.82 -> I had a really interesting use case of a customer that had

1644.27 -> 77 different applications that he wanted to migrate,

1647.72 -> but he wanted to start gradually.

1650.66 -> So he used Flex cache to move one of his application to AWS

1655.67 -> while the other six are still on-prem and he can gradually

1659.09 -> move them and achieve that in a great way.

1663.02 -> Essentially you can really create a mesh of your data with

1667.4 -> those token technologies and I think it's pretty awesome.

1673.01 -> With that, I'll go on to speak a bit about

1676.55 -> what we've done with SAP HANA.

1679.49 -> So obviously we can run SAP HANA

1683.897 -> and your SAP platform using FSxn,

1687.62 -> that was also a lot of heavy lifting,

1689.66 -> a lot of good people working on that to make that happen.

1693.59 -> When you look at SAP,

1695.62 -> it's a very demanding certification that the storage needs

1700.79 -> to go through in order to be certified.

1704.09 -> So I think it says a lot about our performance

1707.03 -> and our enterprise capabilities that we're able to provide.

1711.17 -> I wanted to go over a few of the dry ingredients

1715.79 -> that if you wanna make that's kind of environment.

1718.52 -> So we certified the single AZ deployment

1722.57 -> and we're doing it over NFS 4.1 and in terms of the host,

1728.27 -> it's between 512 and 12 terabyte of RAM EC2 instances

1735.05 -> that you can use in this deployment.

1737.06 -> Also on the right hand side you can go look a bit deeper

1741.56 -> in the SAP side for the specific certification,

1744.59 -> what you should deploy and how

1747.231 -> and I think the how is nice,

1749.81 -> but the why is really important.

1752.54 -> So when we look at what does a shared fire system

1756.62 -> or the capabilities that Gadi mentioned can do

1760.01 -> for your SAP environment,

1762.59 -> I chose free use cases that are pretty common

1766.49 -> and really are a no brainer.

1768.5 -> So when you look at the backup capability,

1773.12 -> if you try to do that directly from the SAP database,

1778.01 -> it will involve essentially a dump of the backup

1782.66 -> and then moving that specific file away.

1787.76 -> And if we put essentially FSx under the hood,

1792.86 -> we're able to do that with snapshots that are pretty

1797.78 -> immediate and then snap mirror.

1799.91 -> So essentially the two minutes represent the block changes

1803.48 -> that we had that we needed to move away rather than

1807.32 -> a full copy that needed to be written

1810.2 -> to an offsite location.

1813.02 -> Similarly, when we look at the restore operation,

1816.11 -> it's the same thing, just backward.

1818.93 -> So if you had a backup of the database,

1822.358 -> you need to restore that file to the original location

1825.093 -> and then from there start up the system.

1829.52 -> Essentially by using the snapshot technology we're able to

1832.34 -> reduce that recovery time significantly and then being able

1836.429 -> to reduce your recovery operation in a very significant way.

1843.221 -> And lastly, when we look at system refresh,

1846.35 -> so updating our dev and test use cases

1850.955 -> under the hood for FSx,

1853.22 -> there's a lot of different things that take advantage

1856.22 -> of our snapshot technology.

1859.37 -> One of them is called cloning.

1862.49 -> So you are able to take an operational system

1865.97 -> and clone the real data and drive your system refresh test

1872.09 -> and your dev operation through it.

1875.42 -> So it helps you make, helps make all those operation

1880.04 -> a lot more efficient

1881.36 -> and they come with a significant right because of it.

1887.671 -> And Gadi mentioned Snap Center.

1890.03 -> So Snap Center comes in clutching that use case.

1894.5 -> Obviously in order to take a snapshot that actually

1897.68 -> is meaningful for SAP,

1899.6 -> you need to be able to communicate with the SAP HANA node.

1905.39 -> So Snap Center essentially allows you

1907.67 -> to plug into the SAP HANA node

1911.09 -> and do all of those operations that I mentioned

1914.27 -> in an application consistent manner.

1916.52 -> So you're able to clone something that is meaningful

1920.06 -> and being able to do it together with the database.

1925.19 -> We obviously can do that

1926.66 -> with a bunch of different application.

1929.93 -> Snap Center specifically does it also with Oracle and SQL

1933.38 -> and Gadi will explain that a bit further.

1938.93 -> When we look at the architecture

1942.68 -> that we propose for SAP deployment,

1946.19 -> obviously you have two components, a key component.

1949.52 -> One of them will be the database.

1952.46 -> The other thing that we can really help with

1954.59 -> is the shared file environment.

1957.89 -> So essentially you'll deploy SAP HANA,

1962 -> you'll attach it to single AZ FSx ONTAP,

1966.47 -> and what you can see here is we're using the SnapVault

1969.381 -> to SnapMirror technology in order to create

1971.69 -> free other environment

1972.98 -> that can benefit from the SAP HANA deployed on FSx.

1978.02 -> So you can take the same data and use it for a backup

1981.35 -> or a secondary copy of that

1984.17 -> and you can use it for a QA testing.

1986.93 -> Obviously if you have the production data,

1989.15 -> there's nothing better to test your new software

1992.48 -> on than the production data itself.

1996.8 -> So you're able to clone that data both to the QA and test

2000.64 -> and for the dev and test environment as well.

2005.38 -> So you're able to either fix bugs that you have

2008.41 -> or develop new software and all using the same data.

2011.8 -> And obviously when we talk about cloning,

2014.23 -> it's also hyper efficient.

2017.11 -> So you can do that many times over

2019.48 -> without recurring any cost and in an immediate fashion,

2023.5 -> which is pretty cool.

2026.35 -> Cool.

2027.183 -> And Gadi will elaborate a bit how you can use that

2030.25 -> same technology with SQL and Oracle.

2032.53 -> - Yep. Thank you.

2034.007 -> So I'm going to cover what you can do with SQL with Oracle

2040.59 -> is really, you'll see that there's a lot of similarities

2043.27 -> to SAP HANA really this is more of databases,

2046.75 -> but there's some small differences

2050.92 -> that was mentioning our experience.

2053.77 -> And you know, part of why we're excited about the area

2056.86 -> of databases in general is that with FSx

2059.32 -> you are actually getting better performance and low TCO

2063.25 -> and for some of us it's exciting and so why really use that?

2070.87 -> So there's a number of situations,

2072.79 -> a number of use cases where this becomes very useful

2077.71 -> using ONTAP, using FSx.

2080.68 -> So this will accelerate migration.

2083.44 -> And really this is sort of a continuation

2086.32 -> of the section I covered before.

2088.93 -> There's many different tools that will help you with that.

2092.8 -> You can do clones and snapshots and backups

2097.99 -> that are database well and very efficient.

2101.71 -> So you can streamline your workloads

2105.01 -> and that will really help you accelerate innovation.

2110.02 -> The backup and restore that are consistent

2113.62 -> can land into the AWS backup framework.

2118.3 -> You can use those tools also to create a setup that will

2122.95 -> allow you to do bidirectional disaster recovery.

2125.59 -> So you can have multiple sites

2127.48 -> and make sure that you can go out,

2130.12 -> you can go back again and maintain the data

2137.02 -> at whatever state that you choose.

2139 -> And I'll talk about a couple situations

2141.34 -> and you can create a very effective deployment

2147.76 -> that is cost region.

2149.77 -> Some companies feel that's very important,

2151.57 -> whether it's regulatory or for just for availability

2156.37 -> and doability.

2159.01 -> And so if we look at an example

2161.465 -> this example right here is a sequel,

2164.41 -> but really they're almost interchangeable.

2167.26 -> It's very similar if you use Oracle,

2171.43 -> you'll see that you have Snap Center running on the original

2175.93 -> site and allows you to basically take snapshots

2179.769 -> periodically.

2181.03 -> So it works in coordination with the database.

2184.18 -> It does an operation that's called quiets down

2188.53 -> and allows us to take the snapshot,

2190.24 -> the snapshots will continually,

2192.07 -> continuously move to the target site with,

2196 -> where we have FSx.

2197.037 -> And so you'll have a live,

2198.94 -> almost live copy that is up to, you know,

2202.87 -> if you set the interval to five minutes, then it's gonna be,

2205.54 -> you know, up to at most five minutes

2209.86 -> from taking the snapshot.

2211.06 -> And it will always be consistent,

2212.65 -> will always be complete

2216.43 -> and that allows you to really transition

2218.83 -> over when needed.

2220.66 -> You can just turn on the database,

2222.82 -> make it live and make that side the active site.

2228.34 -> If we look at the different use case where in that example

2232.54 -> we used an Oracle, but as I said,

2234.43 -> it really works for both of them.

2236.41 -> In that example, we're actually doing,

2238.45 -> we're actually covering a very different use case.

2242.08 -> We in the source,

2244.03 -> in the source site where the data is coming from,

2247.97 -> we're actually taking both periodic clones as well as

2252.22 -> different clones for the purpose of running

2255.28 -> a dev and test operation, right?

2257.47 -> So we're gonna take different slices of the live data that

2261.69 -> will help us do our test and dev situations.

2265.99 -> And you can see on, this will be taken

2269.11 -> and continuously transition to the other side.

2272.65 -> And you see that we have a heap of sort of volumes

2276.04 -> on the left and we can mount each one of them upon,

2280.12 -> you know, whenever we need that to represent a certain case.

2283.78 -> We can even spawn multiple databases

2286.51 -> and just mount the different use cases

2289.84 -> and then run our tests, right?

2291.7 -> And when we're done with our tests,

2293.23 -> we'll just flush everything down.

2295 -> So it's very, very efficient way

2297.13 -> of dealing with that use case.

2302.83 -> So I wanted to talk, to switch gear a little

2307.54 -> and talk about Kubernetes solutions for multi AZ

2314.23 -> and single Z and again, we

2317.23 -> we feel that there's a great advantage

2319.27 -> in using FSx ONTAP for that.

2324.82 -> First of all, why don't you talk about the journey.

2328.18 -> So we started very early after containers

2334.96 -> and containerized solutions became a thing

2336.85 -> and became a reality.

2338.83 -> We started working and creating solutions around that.

2343.36 -> I think an important point was when we launched Trident

2348.94 -> and that was 2016.

2355.18 -> I can't see from me and target is really important.

2360.2 -> It's a open source CSI diver and you can use that to

2366.4 -> basically abstract and use any ONTAP and it's very,

2372.52 -> it has a thriving community of developers and users.

2377.92 -> Fast forward, we basically provide,

2385.63 -> provide the Trident for a number of solutions.

2391.24 -> We are launching a more advanced product called Astra,

2397.48 -> which will use the CSI data but will be a lot more.

2402.67 -> Astra actually allows you to capture not just the data but

2406.63 -> actually the entire application with all its dependencies

2409.99 -> and intel relationship between the objects,

2413.86 -> you can back up the entire application

2416.38 -> and you can restore it somewhere else.

2418.63 -> And that's really tricky because the target might actually

2423.07 -> be of different configuration.

2424.9 -> Astra will figure this out and deploy it in such a way where

2429.61 -> it's going to work as intended.

2434.44 -> The Trident CSI driver is also available for FSx

2440.65 -> and it really helps providing a very unique solution because

2448.69 -> I think we mentioned that, but I will remind you again,

2453.19 -> FSx comes in two flavors.

2456.31 -> It can be deployed over single AZ,

2458.8 -> it can be deployed over multi AZ.

2462.25 -> When you deploy FSx on multi AZ,

2464.62 -> it means that you have a fully synchronized copies,

2467.32 -> each one is sitting in a different availability zone.

2470.17 -> Now, when it comes to the solution that I'm presenting,

2476.77 -> I believe that there is no other storage solution that will

2479.83 -> give you in an easy way out of the box the ability to run

2485.255 -> Kubernetes across multiple variability zones.

2490.15 -> And so the solution would look like that.

2492.16 -> You will have your persistent claim,

2494.29 -> persistent volume claim.

2496.69 -> You can in fact run containers on both AZs

2501.79 -> and they will both see the same data.

2504.43 -> The storage layer below is active passive,

2508.12 -> which means that the containers from both side will actually

2511.09 -> communicate with one of the AZs.

2513.7 -> And if that AZ has a problem or the node fail,

2518.62 -> it will seamlessly non-disruptively fail over

2521.92 -> to the other side application will never notice.

2526.51 -> It's worth mentioning here that this will work

2531.46 -> whether using SMB or NFS or Block.

2537.73 -> And so you can really,

2540.64 -> it gives you a lot of flexibility

2542.38 -> and because there's no cost when you're using FSx,

2546.543 -> there's no cost of communication between the AZs,

2551.44 -> it makes it a very interesting solution.

2554.11 -> We had the case with the customer recently

2557.242 -> that used a different solution

2559.03 -> that when that solution was really small,

2561.37 -> those communication costs was insignificant,

2563.92 -> but over time they got to a point where the communication

2567.79 -> cost between the AZ one and AZ two were the biggest single

2572.033 -> line item on the bill

2573.708 -> and they were very frustrated because of that

2576.79 -> and that actually made them replace the old solution

2580 -> and move to FSx.

2584.17 -> Okay, so I'd like to hand off to Yuval,

2590.35 -> that's going to tell you exciting

2592.81 -> news about BlueXP.

2595.39 -> - Cool. Yeah, thank you Gadi.

2597.52 -> So for those of you who don't know what BlueXP is,

2602.74 -> is we used to have a cloud manager

2606.34 -> as the data management plan for CBO and FSx.

2611.318 -> And essentially in the last insight,

2614.29 -> which is the NetApp conference,

2615.73 -> we up our game and we released BlueXP

2619.81 -> in order to connect a lot of our capabilities

2623.47 -> and give you guys one experience

2626.708 -> to essentially enhance your data management

2631.15 -> and the things you need

2632.29 -> in order to be successful with storage.

2636.49 -> So what are we looking to achieve

2638.8 -> and what we're looking to enhance

2640.27 -> in terms of the BlueXP experience.

2643.406 -> So first and foremost you're managing storage.

2648.01 -> Obviously Andy is doing a great job with enhancing

2651.07 -> the AWS console with different capabilities,

2655.09 -> but we wanna make sure that if you use BlueXP

2658.09 -> to manage the storage, you get a great capabilities

2662.47 -> and features as well.

2666.73 -> So that's the first pillar that we're focusing on.

2670.33 -> The second one is around mobility.

2673.09 -> So being able to help you move the data to where you need

2677.47 -> and being able to achieve that with the platform.

2683.32 -> The third one is around data protection.

2686.68 -> So if you for example need to manage Snap Mirror

2691.84 -> and your application and know exactly where your data reside

2697.467 -> and how does it protect you are able to manage all of that

2702.61 -> through the BlueXP console.

2707.08 -> And lastly, being able to also understand your data, right?

2712.72 -> That's something that I hear from essentially every customer

2716.2 -> that I talk to is that there's data that they essentially

2720.01 -> don't have any idea what's there.

2722.74 -> So we have a lot of tools within our arsenal

2725.89 -> to help you governance the data,

2728.17 -> both from understanding if data is old

2731.5 -> and shouldn't be there or if data from a privacy perspective

2736.51 -> shouldn't be in the location that it's in

2739.48 -> and allow you to put a lot of guardrails around your data

2744.64 -> and optimize it obviously from a TCO perspective as well.

2750.46 -> When we look at the different integration,

2752.26 -> I'm not gonna on each one of those integration in details,

2758.68 -> but essentially blue speak ups with a wide array

2761.92 -> of technologies and services that helps you

2766.24 -> achieve what you need.

2768.1 -> On the left hand side you can see obviously Snap Mirror,

2770.77 -> which is essentially bread and butter for us.

2773.32 -> It's something that you can manage very easily

2778.135 -> in the in the BlueXP console.

2781.6 -> And obviously if you're not using ONTAP

2784.83 -> or you wanna do data,

2788.14 -> moving data between NFS to S3 for example,

2790.93 -> we also have cloud sync within the platform

2794.41 -> that helps you do that as well.

2797.69 -> And lastly, on the right hand side

2799.75 -> as an example for governance and compliance,

2801.88 -> we have the data sense capability that essentially

2805.321 -> allows you to really understand

2807.34 -> what you have within your data.

2809.2 -> So all of that is under essentially one umbrella

2812.92 -> and it allows you to manage your data from data protection

2817.24 -> and also manage it from all the different aspects

2820.66 -> that you need in order to be successful.

2824.23 -> I'll try to demonstrate it also live,

2826.489 -> hopefully live then work is expected.

2831.7 -> - [Gadi] Log in. - [Yuval] Yeah.

2834.85 -> And if it want to let me log in, that will be great.

2844.78 -> Cool. - [Gadi] Excellent.

2846.25 -> - [Yuval] Awesome.

2847.228 -> So I wanted to start off with the, with the website itself.

2851.02 -> So, and similar to how AWS managed the console,

2856 -> you get the information you need and you get the technology

2858.91 -> in the same place.

2859.99 -> So you're able to go to BlueXP.netapp.com

2863.527 -> and you're able to get all the information you need

2866.83 -> in order to be successful.

2868.48 -> So in this experience you essentially get all the different

2873.67 -> capabilities that we provide.

2875.62 -> So we cut them off according to what I explained before,

2881.77 -> storage, mobility, protection and analysis and control.

2885.88 -> And you are able to understand better what we can do

2889.6 -> in order to assist you.

2892.15 -> If you wanna focus specifically on FSx,

2895.57 -> we have a a specific page that helps you understand

2899.38 -> what can be achieved with FSx with our console.

2903.76 -> So if you go over to the FSx page,

2907.12 -> you get the different resources that will help you succeed

2910.81 -> and you'll get the different capabilities

2913.6 -> that will allow you to achieve more with your environment.

2918.97 -> One of the things that I wanted to highlight

2921.07 -> and then we talked about it in the presentation is

2923.5 -> we also have a lot of resources

2925.257 -> that will help you make decisions.

2928.63 -> One of them was the VMC and FSx TCO.

2932.83 -> So within the console experience you can go and experience

2937.965 -> the TCO calculator and understand exactly

2942.19 -> what is the cost of your environment.

2945.28 -> So it's a pretty easy calculator.

2947.38 -> You put in your environment details and you are able

2951.85 -> in the bottom to get a result and explain on the TCO.

2957.13 -> So you can see here for example,

2958.66 -> that in the environment that I choose, I choose wisely.

2962.2 -> I can do 70% TCO saving,

2966.67 -> but essentially any heavy storage VMware environment

2971.8 -> will enjoy a similar TCO savings.

2977.2 -> So if I wanna go and actually use the console

2981.31 -> and apply what I learned.

2984.37 -> So within the experience you can go

2986.83 -> into the BlueXP console itself.

2988.96 -> If I go back to the website,

2991.42 -> you have a quick get started button

2993.52 -> that will get you into the console in an easy way.

2998.5 -> Within the console I'm able to manage my FSx

3002.55 -> and also manage my on-prem environment as well.

3006.42 -> You are able to replicate data not just from ONTAP.

3010.14 -> So if I take an on premise ONTAP,

3013.44 -> I can replicate it right here,

3015.9 -> but I can also take storage that are not ONTAP

3019.35 -> and use the clouds in capability as I demonstrate.

3023.88 -> So if I drag and drop a non ONTAP storage,

3027.99 -> I could still move the data in a nice and efficient way.

3033.6 -> So I'll run you through a replication

3035.73 -> to show you how it works.

3037.08 -> So if I drag and drop the system,

3042.84 -> I'm able to replicate in a pretty easy way the environment.

3047.34 -> If my screen will allow me.

3053.55 -> It's in the button. Wonderful, okay.

3060.54 -> It's okay.

3065.31 -> Oh, Didn't allow me,

3067.38 -> but it does work not in the demo environment

3070.92 -> and I'm able to replicate and essentially manage

3075.39 -> that replication directly from here.

3077.19 -> So if once you replicate,

3078.72 -> you are able to do all the different capabilities like

3082.95 -> editing the schedule, breaking it, and using the data.

3087.27 -> If I want to use the FSx and specifically I can

3094.23 -> go back to the environment and I can go into my FSx system.

3102.609 -> And essentially manage it from here I get

3105.21 -> the information about the system where it's deployed

3108.57 -> and I'm also able to manage the different volumes

3111.12 -> that I have.

3111.99 -> So in my case I have two volumes and I can from here

3115.41 -> directly get the mount command,

3117.99 -> clone the volume or add the different settings that it has.

3124.2 -> It's important to note that everything I'm doing

3126.862 -> can be automated as well.

3128.73 -> So we have a Terraform model and we also have Ansible models

3133.29 -> and everything can be plugged in into automation

3135.9 -> and be part of your CICD pipeline as well.

3142.35 -> And the other key capability that I wanted to highlight

3145.71 -> and it's thing that is upcoming is we're adding,

3150.6 -> as I mentioned with blue screen, we're up in our game,

3152.79 -> we're adding a lot more services and one of the services

3156.18 -> that we're gonna offer is essentially

3157.83 -> the ability to use system manager.

3159.9 -> So you'll get a capability to double click

3163.11 -> and go even deeper into FSx.

3165.75 -> So I have a quick demo to show you

3167.61 -> and demonstrate how it works.

3169.68 -> And so essentially this is an FSx system

3172.98 -> and you are able to use all the different capabilities

3176.34 -> that you used to form a system manager

3179.79 -> directly on the FSx system.

3182.4 -> So for example,

3183.27 -> I'm able to set quotas and Q3 is directly from here

3189.54 -> on the different volumes demonstrated quickly

3193.71 -> if it would allow me.

3196.77 -> Let's go to Qtrees,

3198.6 -> so Qtrees I can go and add a different Qtrees

3202.67 -> on the specific volume that I'm managing.

3209.64 -> Cool.

3212.01 -> Awesome.

3214.71 -> Let's do it back.

3216.18 -> - Thank you.

3218.1 -> So, very exciting, especially the system manager.

3223.5 -> This is the first time that we have system manager

3225.33 -> working for FSx.

3227.85 -> So I want to, we are nearly out of time.

3229.53 -> Wanted to recap.

3231.03 -> So Amazon FSx for NetApp ONTAP,

3235.41 -> it's the only complete fully managed ONTAP file system

3239.64 -> in the cloud.

3241.71 -> We're very excited about it.

3243.12 -> You should definitely check that out.

3246.99 -> It is the easiest path to migrate

3249.81 -> or extend any storage to AWS

3254.85 -> and it enhances your FSx for ONTAP experience with BlueXP.

3263.07 -> You can always definitely use the console, the APIs

3267.45 -> and you also have the BlueXP experience,

3270.51 -> now also with system manager

3275.07 -> and I think we'll cut it short in favor of having questions.

3281.04 -> So maybe Andy can join us.

3284.04 -> Anyone with a question that we can take?

Source: https://www.youtube.com/watch?v=HR8TwDMTB5k