An Introduction to Rubrik with Chris Wahl

so my name is Chris wall I'm the chief technologist at rubric I'll be doing kind of the introductory stuffs you know the boring what's the problem that kind of jazz and then I brought a lot of amazing talents that work at rubric to really dive deep into the product the vision the strategy all that kind of jazz so hopefully that's cool I wanted to start out by outlining our focus because I think that's really important and especially this room I feel like if I put a statement like this on the screen and it will really resonate with the messaging that you're expecting from a top-notch vendor in this space and bonus points if you spot the amazing hashtag that's there so I obviously love to help those that are looking to play our famous game when we put that back on there would that be without my awful did you see it okay like yeah there you go but those that are playing the buzzword bingo game at home you now win right you've got all of them on there for you but seriously though I do want to do a quick introductory type spiel just for those that hadn't seen us before or potentially you know haven't watched an earlier field day video what's wrong with you you should do that here's our four co-founders the company is a little over three years old now the products been out a little over two years right so the whole soham Nitro which is a really cool name and AJ are all from various blends of consumer enterprise companies so there's kind of the background of the founder level obviously they look into a target the data protection market it's actually quite huge from a funding perspective we have had some updates from our previous time on tech field day or field days in general so we did close our series D and I don't even know if we had our series C on a previous one so a C and D have been closed adding IVP as one of our investors for the series d putting a set just under three hundred million dollars invested into the company a lot of people seem to like what we're doing there's some buzz out there not as a delegate as an architect in the in the channel I didn't really care about this stuff in the past apparently it is somewhat important especially the larger enterprise customers that are looking for those fruit points and things like that I won't spend on that too much it's time to get to the most interesting part of the presentation the problem I will be short I will be sweet I will try to get done with it as soon as possible because Theresa has told me I need to get to the bleep in point well she won't be interested so obviously I did help do that so real quick for those that are familiar the problem the problem is the data protection or let's call backup for and simpler word sucks quite quite plainly back in the 90s when I started working in IT which is either a long time or not very much time depending on how long you've been at IT we had to deal with this kind of architecture you know you go out in your shop your storage and your servers and you build proxies and I tell you the first couple times I did it I really sucked at it because it didn't know what I was doing and the vendors are all liked by my particular piece within the stack and put it together there was one kind of major iteration to that architecture that came out in the 2000s where was like Oh instead of going direct to tape you can now do dupe that data put it on disk and replicate it and do all that kind of fun stuff but ultimately you're responsible for doing all these things even if you buy from a vendor that essentially like puts them all under one roof this architecture still exists underneath the covers right and this isn't a very good architecture that's very monolithic in nature doesn't scale very well and back to my first point it just kind of sucks right so there we go that's sad panda we don't like sad panda especially a field day we like happy panda I do not have a picture of a happy panda even though Andrew picked one out for me so what we do at rubric very quickly we take all these things that you need to build back up we distilled them into software we distribute them across a platform which comes in physical cloud or virtual as it's a way to distribute and in this particular case I'm just saying we collapsed down into the software stack the first thing that we put on the market was an appliance to you four nodes you scaled them out to start with you know one appliance add more appliances it logically looks like one thing from a services and address ability perspective it is one thing there's no demarcation points for compression or deduplication anything like that so you just grow as you need and we're trying to leverage things like public cloud or just object and file and nads and things like that for archives right those tend to be very simple use cases I know you heard from some other vendors in kind of that public object kind of space talking about this is a popular use case to put backup data in there yes it is especially when you globally do do put under one name space right so a little bit more techie real quick cover the nerd knobs a bit you've got your primary environment service storage all that kind of jazz physical applications virtual applications whatever you deploy the appliance on Prem or potentially our cloud version for your cloud workloads I'm just going to cover the on Prem for this particular slide we're put that in it inventories your environment either it's physical or virtual we're talking to the api's for something like a manager like be center or Acropolis or hyper-v something like that or physically we're talking to our connector to interface our API to that particular software once that's done all the data is actually flowing parallel to flash within the nodes right so there's a lot of performance the bottlenecks you're typically having to solve with proxies or whatnot go away this is all controlled using policy because having to build backup jobs also sucks I don't know why you'd really want to do that and as an architect to try to be like oh here's a thousand applications what order should I do them in I don't know that's why we have traffic lights for cars there's no one person that can do all that so the policies dictate all of the ingest and the backup things like that all the stuff you don't want have to deal with if you want to recover something which is probably like the point of a backup it's not just a fancy you know thing that you put your data center that blinks and eats power you can do it instantly so we're talking a few seconds in most cases you know up to a minute potentially if it's really huge or weird everything's secure right so all the encryption both internode communication pulling data out of the primary environment putting data into an archive environment etc fully encrypted everywhere using your keys because with ransomware what not really just plain common sense you obviously want to secure your data and then finally if you want to get the data out of the appliance which we suggest because if you were just infinitely by rubric that'd be kind of pricey put it to private or public resources that's object store Bob store you know file whatever it is that you have so it takes this architecture where you're kind of having to get a person or a collection of people to put all these things together the proxies and servers and enterprise search servers only kind of jazz you just kind of throw all that away from an architectural perspective you don't need that anymore simplifies the entire equation to put a certain amount of rubric and appliance form factor in the data center potentially some in the cloud and then just scale it out as you need to you know performance and capacity because I want to deal with that you probably seen that before because that was like our main pitch for a long time but honestly this is cloud filled day right they can't just be like here's your data center back and some stuff and shoe it off to the cloud that's kind of like that's lame right that would be like BS cloud field day so we're in like legit big-boy pants mode a couple differentiators and again they all have really cool demos for all this stuff the rest of my team I'm just the guy that kind of warms you all up make sure that the coffee is hitting the veins things like that so that was the beginning so big differentiators the first is that when you look at other products whether or not they say like oh it's unified or you're actually building it using software and storage and stuff like that yourself it's typically not end-to-end management there's a manager of managers there's very various element managers that are being talked to one of the covers or potentially you're sold a unified thing and then you are the manager of managers which also sucks right so in this case rubric the software the fabric itself controls everything that can be done using kind of we call here set-and-forget simplicity because I want to just build some policy interact with api's and be done right I want to spend a few minutes a week tops dealing with the system I don't want to be in there like I don't want to really I want to forget what the UI looks like entirely that's kind of the goal right I just want to manage things at the policy and API and kind of you know Service Catalog level so that's another major point here everything's API driven even under the covers and then finally it should be you know it's kind of a loaded term there the cloud native design I know there's some apologize for a little buzzy there I already already let you win the buzzword buzzword then go but the idea that this is not a bolt on we did not take an original the idea that was built for on-premises type workloads and then add cloud is kind of an afterthought the cloud technologies that we started with for s3 and object and thing like that we're native to the platform right there intrinsic to how we do business the api's are intrinsic to how we have the models and the modules of software talking to one another so this is not Bolton it's not like oh just cloud things kind of caching out we should chase that it was like from day one this is in the DNA of the product right so what we started with is when you look at tech field a 12 or 10 or virtualization field a 5 which we're all at previously was you've got VMs running and then physical workloads and things like that you've got some the rubric appliances running of the data center what we started adding to this equation is well yeah you can shuffle that data out to s3 or just let's say cloud storage generically right with s3 from Amazon as our original intent we also allow you to run rubric as a cluster in there well let you build it management manage it as a separate standalone cluster running in that environment in fact you could just go totally whole hog into the cloud and not even have a physical deployment if you so choose but if you wants you can tether them together and replicate you know have the have the data move around based on your needs you can start spinning up workloads based on backups that you've archived to the public cloud right we'll go into that and that technology which we call cloud on but then beyond that if you're looking for a multi cloud strategy we also allow you to deploy the software into Azure which was the second cloud that we targeted as well as using their blobstore replication between clouds potentially archiving between clouds you can really choose in some cases to go with one and then have the other big type your backup right here your backup of a backup so the topology is whatever you need it to be and I want to point out because this is important it may be a piece that others have left off what they're talking about this is all generally available right this is GA code our ultra release for Dido is not vaporware slide where you know going to be here in six to nine months this is actually here and we'll be showing to you today if the Wi-Fi gods will be kind of me and the internet guides will be happy so some use cases if you can take data and you can put it into a public cloud from an archive perspective why can't you act on it and start automatically looking at the shape and the format of that workload and handle all the heavy lift to convert it into an ami running an ec2 which is what we introduced with cloud on things like that really empower you to start leveraging that giant mound of ec2 compute that's available beyond that from an app and database perspective we've always allow you to do instant recovery you know the idea bringing back an application beyond file and folder level stuff that's bush-league full application restores within a few seconds we also now offer that with databases our first one is Microsoft sequel right so you can and Michael fall here will actually demonstrate that where we can bring on huge huge databases instantly you know like within seconds or a minute so you can start leveraging that and the locality of that recovery is not limited to the data center right so you start doing some really interesting things when you have a giant pile of compute and resources in a public cloud and the ability to spin up or move data around at will so I was kind of my final homage to to rubric you have a data management platform that spans pretty much any private or public resource that you want right and that's been our vision all along I start with the data center at at the enterprise make sure that we're solving their pain points and then help them get into that cloud model that they're looking for with a bunch of interesting topology options right but underneath there that's where I get excited it's all API driven it's native there's no price for it it's just like someone asked can you turn the API off I'm like well then the product would stop working because we make calls to the API between our software stack you can actually hit f12 on Chrome developer mode you can see all the calls in real time but that means that if you're looking to do anything about obviously you need API so this isn't a heavy lift it's pretty simple a few summary points here I'll only highlight the last one it's an experience that you actually love using right I've never thought in my life I would say that I liked using a backup product but it's fun customers like this is cool someone tweeted kind of shamefully like I really kind of like this rubric thing I like backup and I feel dirty I'm like and you should feel dirty but also awesome is you know you have a cool system in play now so what did I do like like 10 minutes tops 15 minutes I got all the what is the problem statement out of the way and you've won by this buzzword bingo I think we're cool there and this was cool spinny thing that you can just stare at and come bez memorize that I will end it there and we'll switch to the nerd stuff before I kind of walked off any questions like an answer right that's why oh good amounts of money for a backup such data management company yes y'all ready for this Rebecca company that's why they have to be a data manager yeah then they're man yo Yuki there's a question is up yes the question is your investors as obviously am humbled you for the information but I mean there obviously there's a long game here a very must be a very long game with very big plans to support that kind of valuation can you say in what's my roadmap next two years like not not just on the roadmap better people of course a purpose roadmap yeah what's the business road that oh you're going to be grow up what we're gonna be when we grow up the same people we've been competing against I think throughout the whole process you know it's not the we've not really been in the market of you know let's go after the startups like that it's been the incumbent vendors that own the market chair and I don't want to name names but if you look at who owns the market today that's obviously who you have to take it from to someone that's trying to not be hardware storage vendor and become data spat even replacement yeah it's difficult to take a platform that is 10 plus years old and make it do all these things without a complete refactor so I think there's a certain advantage prety chicly at that point but also if you look at that slide that I built with you know the on-prem going to multi-cloud just keep kind of envisioning beyond that point we keep adding more more options to the topology that drive use cases that people have been kind of chasing and trying to you know they've got their hands just barely around it they're trying to grasp it and they kind of keep swishing through the clouds so to speak you know if they can't get an ephemeral grip on it and now they kind of can and an interesting way the vehicle is through backup in recovery is it is a vision also always backup in recovery or is this the first stepping stone to other kind of things well I think from a messaging and a target perspective backup recovery is a problem everyone has so why start with an itch and try to expand it we took something very public and painful which is backup in recovery but if you kind of rewind a little bit to that being a use case backup and recovery there's also copy data management analytics you know potentially touched as that kind of that kind of jazz there's a lot of different use cases I've always viewed back in recovery as just one of them and typically the strongest and most relevant to an architect or engineers mind it's something that you have to have or at least you say you have right you know Eric and I did some consulting you know some people say they have backup they don't but you probably should but it's certainly not not the end at all it was just the start I think I think even backup in recovery it's probably the single most screwed up area within an IT function and has been that way for a very long time they do want ice it screwed up I don't mean to be immediate I ran through they weren't the most horrible clunky but it's not it's not done well yeah you know there's not a lot that's invested in it that really should be and part of it is because it gets incredibly complicated especially if you start thinking about RTO RPO and and all the different application and integrity checks and everything else when you go on and on and on and so one of the things I'm kind of interested in and interested in looking in more detail is just how do you simplify that okay because to me if you simplify that that's a no-brainer for practically every enterprise on the planet right so great here that you're going other places but if you get backup and recovery done well that's a huge huge success I don't know that Isis to speak Tibet what what is the TAM for backup and recovery and are you guys public about your run rates we don't the last public number we say the run run rate was approaching a hundred million annually as far as the Tam it's I think the last published numbers are a couple years old but it was north of 40 billion we'll say 50 billion to make the number easy and that was the copy data management specific piece of the puzzle so it's pretty big yeah I think that's a different way to answer the question is that market is so huge we don't really need to focus on anything else yeah I don't know that I think though increasingly there's kind of become commodifies and so even if it wasn't for the ridiculous valuation that's very impressive valuation this isn't what I want to be looking to the future and it's the smart stuff you can do on it I'll make kind of commoditized storage Beck yeah stuff that's important it's not that I disagree with you I agree with you I'm just saying there's such a need for this across so many IT organization so many enterprises today that if that's all you did that's a huge win that's a huge win for not just for rubric but I mean for the companies now I agree with you I think if you do that well then you have access to an interesting set of data and analytics that you can use some very cool things with but I mean that that is true but it's a heck of a tough sell because backup and recovery is one thing live mounting databases is cool but getting into the data management where you then run analytics on secondary copies you're now selling into your analytics team into your DBA team into your backup and recovery team it becomes a far more far more difficult sell but I do agree that sorting out backup is a a major pain point there's two bits that have made it really cool one is that it's become easier because of the way the deployment technology works like that's just become simpler than what we had with servers because of the changes in technology then we had ransomware come along which suddenly turned what was an insurance problem into something which is like I need it now because everyone decided they need to have insurance when their neighbor's house burns down everyone's neighbor's house is currently burning down so everyone's gone hmm maybe I should do this recovery thing probably is easier to use and there's these are not the only couple I brew brick I'm not the only company to have technology that makes it easier to do stuff like finding recoveries and so on we've got your point about what you do next if everybody needs backup and recovery you're already there so now you start saying well we've got the data anyway let's upsell you onto other stuff yeah that's fair enough I was deliberately being being a sort of devil's advocate because it's a high funding it's a higher bar to to justify that kind of money like nothing lasts doing well yeah I need for the question well that's worth if wanna cry comes in and locks up all of your stuff and you can't and you cannot run your Hospital yeah yoram what's that worth to you yeah but going back to your you bring up a valid point right which is there are a lot of different groups that have to get involved in this right but at the same time how many of those groups really want to have to deal with the nuts and bolts of backup and recovery I'm oversimplifying back covered but that's okay I mean the run of sums the conversation we've had for in terms of sales conversation and you're talking to you know head of storage but can you have a conversation with CIO says hey you and dismantle analytics we can do you know you know your storage do debt she can be the hero of of IT generally or of the business generally because she can enable snapshot and clone and analytics on top of that and that having that conversation at the different layers yeah and throughout silos is difficult one thing you can take you're very traditional sequel server customer database move that to the cloud spin it up and then use Amazon's built-in and a on that is a is a completely different mindset of yes or we can back up in the store your your database and it fails in a positive Delta so we have a lot of awesome demos I want to get as good I don't want to I love the conversation though but thank you very much for this particular segment

Channel: Tech Field Day

Views: 33,856

Rating: 4.6778522 out of 5

Keywords: Tech Field Day, Cloud Field Day, Cloud Field Day 2, CFD2, Rubrik, Chris Wahl

Id: D0gYUYbVDUA

Length: 20min 25sec (1225 seconds)

Published: Sun Jul 30 2017