00:04.30 greybeards This greyards on storage episode is brought to you today by minao and now it is my pleasure to reintroduce a B Periaai co-founder and Ceo of Minao who's been on multiple great birdss podcasts. So ab why don't you tell us a little bit about yourself and what's new at min io. 00:20.89 AB Um, um, one of the co-founders and Ceo and mean was and object storage and what's new that there's actually quite a bit of activities going on I think the latest news is the it's. Search switching to the next gear. It's all about multi-cloud now the adoption of minow has grown quite a bit even inside the cloud and we are we are targeting basically all all the multiple all the clouds and we are going after it. 00:52.96 greybeards Ah, so is being object storage has been kind of a secondary storage for I would say Decade I first started talking about it back in the 4000 True 2004 timeframe Some discussion in object storage might become more like primary storage where does min Io fit into that framework. 01:14.55 AB So that's something that we have been quite clear from the beginning in terms of where we want it to be but ah, but the market has come a full circle around it now now the market has aligned itself to object storage is the primary storage. But if you looked at it was not like something we predicted it was already there when we started if you looked into the public cloud object storage was the foundation whether you store some static websites container images application artifacts all the way to. Aml right? like look at like snapchat to snowflake. They are built on objects to it from emr to even outside of Aws from bigquery azure ml power bi name it objects storage was always the foundation. Always the primary storage. In fact, aws the whole cloud thing started with. 02:07.90 greybeards The s 3 stuff? Yeah, yeah. 02:07.11 AB Objects to h as a service first that is s 3 didnt came yeah then and then came everything else but outside of the public cloud. The industry was dominated by San And nas vendors they were in dismissal right? They they they thought that the future would be. More like enterprise on the cloud that is file block and vms and a file block and Vm cloud is called managed service like the traditional ah Ms. Piece that market has changed I think now you can clearly see bmware is embracing Kubernetes wholeheartedly the the new model is ah containers objects. 02:36.99 greybeards Oh yeah. 02:42.51 AB Kubernetes is the infrastructure api standard yamalyst the definition of the of their stack right? and it does become today. Object storage is the primary storage and why this change happened even outside of the public cloud there are 3 primary use cases for San And Nas databases right? database right. 02:59.28 greybeards Oh yeah, yeah. 03:02.51 AB That now look at from from kafka elastic clickout name. It alllthough pretty much all the popular open source databases old school ones all the way from http://sqlserververdica teradata all of them they have gone object storage. Then look at vm images snapshots container data images. All of they are not the container. The vm images and the database snapshots. All of them have come to object storage. In fact, the new world. There is no vms is containers container images on artifacts are on object storage always and then the last thing is all the archival data. A actually archival data. Not so much for the for the for the primary storage for the San Nas the a ml data that also came to object to so today I think object storage is accepted as primary storage not just inside private public cloud at private and edge as well. 03:42.88 greybeards So yeah. 03:54.11 greybeards Yeah, it's still a struggle seeing seeing that in the enterprise to some extent given the Enterprise Applications proclivity for block and file but I see object being you know the the major storage play and and any of the Cloud vendors these days. It's like. 04:11.27 AB Yeah, or. 04:13.92 greybeards It's it's like abs and those sorts of things are you know are very rarely used anymore. It's all all the data sitting on object storage and ebs is maybe temporarily used and things of that nature. It's. 04:22.80 AB That is true. 04:22.51 Keith Townsend Yeah, so that that right, you're hitting a really interesting point which is most of like the resistance that I've seen from object storage is from traditional workflows. Let's say the data scientists are still kind of slow the pickup object because. 04:33.63 greybeards Um, you know, um. 04:41.12 Keith Townsend Their tools don't natively speak object a I love to hear what you're seeing from like a end user perspective legacy application I like H Piece E and etc. No. 04:48.18 greybeards You mean something like Splunk doesn't use object I'm sorry I don't know where you want to go. 04:56.62 AB Yeah, yeah, yeah, yeah, actually if yeah. 04:58.87 Keith Townsend So meaning like ah some a lot of the analytic platforms that people use at they at an in client to yeah, not now whatoop that who do hadoop is definitely. You can get block storage I'm talking about like on piece when people connect to fo chairs and run you know traditional. 05:04.60 greybeards Hadoop. 05:12.78 AB Um, yeah, ah yeah, yeah, actually Ray answered that question in just 2 words right? you mean splunk doesn't support object store actually splung smart store is s 3 Api Foundation 05:16.88 Keith Townsend Ah, analytics applications. That's where I'm seeing you know R our our workloads etc. 05:29.63 greybeards Yeah, yeah. 05:32.26 AB Doesn't know how to talk to their san or Nas in factlunk engineer if you talk to they would tell even this splunk hotier. Don't put it on sand and nas the heartier is more of a primary cache with some persistence they make replicated copies of it on local drives. But for all persistence smart Storie is the way to go. And smart store is object store s three api every one of the analytics stack today if you look at from old school established players that all the way to the most modern ones. All of them have gone objects to in fact, the most successful one there. Ah, that replaced sand and Nas was htfs and htfs is now coming to object store for a while they were using s three a adapter to make object store look compatible to to adoop applications right? So it gave file interface its cfs file interface to object store. So. 06:19.10 greybeards The Files and. 06:27.10 AB How do applications hive and everything you didn't have to rewrite. But now nowadays if you look at all these modern a ml like kubeflow which is the data pipeline standard for kubernetes uses mini ways sdk to talk to us 3 compatible objects store. So object store is now the standard for ourl analytics. 06:35.75 greybeards Kubeflow. Yeah, yeah, yeah. 06:41.94 greybeards Maybe this is maybe it's the question of definitions. What do you mean by analytics Keith versus I you know I look at analytics is big data kind of solutions but are which is okay, ah scientific language you know stats language. 06:47.88 Keith Townsend So so if I'm if I'm and if I'm using r if I'm using some type of it so languages like Hpc like Hpc has been extremely resistant to object storage. 07:01.64 greybeards Yeah. 07:06.15 greybeards Um. 07:07.60 Keith Townsend Because those users those data scientists because they're still data scientists are not necessarily using Hadoop They're not using modern applications. They're using the legacy tubes. They've always used. 07:16.53 greybeards But I mean are and and Python all these other solutions that do data Science I mean they have object storage Apis don't they. 07:23.99 AB Yeah, you. 07:25.59 Keith Townsend Yeah, they have object storage apis. But again, it's a question of where does the storage live today if the storage live doesn't live in object storage. You have to either move it from your foul system based it's a workflow question like what comes first the chicken or the egg I love object storage. 07:30.52 greybeards I see. 07:33.68 AB Um, got. Yeah. 07:43.71 Keith Townsend Absolutely because it's cheap deep performant now. But my workloads exist. So my data sets exist on on foul that migration has been a big slow in my experience. 07:48.20 AB Yeah, yeah. 07:54.87 AB So but there are 2 parts to this right? one is a data scientist downloading some Csv like a Json type log data or seen kind of data set. They download it to their laptop and are their workstation and they perform and the local drives. 08:03.95 greybeards Right? right. 08:13.52 AB Sure you can run mean I on those laptops to write the application developers. Do it all the time but they would just store these csv files on a file and if you're running r script rate just a local operating system provided this file system is just fine but where did these datasets come from. They come from a large data repository. That's off an object store. And it used to be htfs that became object to because you can access the data scientists spread across. They actually can download the data set over Http securely because s three api is it is just http is fully api hdfs to Sand Nas accessing across the. Across the cloud forgetboard even van they found that object storage is a lot more convenient and secure to do that. But for local processing sure local file system is just fine and outside of this use case the the hpc market like predominantly is mpa based workload. They have their own highly optimized gmpaio and the luster type systems had native integration with the mpa yeah mpi systems sure those those community hpc community from say 92% to go to 94% efficiency they have to make it twice as complex they would because it's worth. But them. But and that's the that's a market that I until we see a pool. It doesn't make much sense but the commercial and commercial hpc market is quite different from the national labs the commercial hpc market have has moved on. They are all objects to rich nowadays. 09:44.87 greybeards You're talking like bioinformatics and things of that nature. Yeah, and so I I see where you're coming from Keith but yeah, you're right I mean if I'm doing an application on my laptop or something like that or in my test environment. 09:49.57 AB It was moving yeah moving away towards. 10:01.84 greybeards Um, probably want to look at Files but that data is coming off of some object storage someplace in the Cloud or someplace and and you know where it's all being gathered and stuff like that. Don't you think that's the case. Yeah I agree. 10:11.33 Keith Townsend Yeah, well user habits are very hard to change rate I think the if if you're coming out of if you're coming out of environment where you're you know you're at a university et cetera and you're learning this stuff new and your ah your first experience with interfacing with. Your data Sas is from object storage. You're going to keep that you're you're just going to keep that that one remote the same. The opposite is true. Ah, when you've spent years you know connecting to the f drive and running it off. User habits are extremely extremely difficult to change whether or not. 10:32.53 greybeards That mindset. Yeah yeah. 10:42.34 AB You never helped go. 10:49.74 Keith Townsend Underlying technology is better or not is not really relevant as can you get the users to adopt it and my question is more ah about that in adoption and it sounds like Aia what you're telling me is that if you want a monitoring experience when it comes to data analytics object storage is where it at. 11:05.53 greybeards And yeah, but. 11:05.85 AB Yeah, yeah, yeah, and the file is actually even the the data Science Community once we gave them a nice file explorer that they can find their browser and download the data set and search all that if you give them something convenient, not necessarily a better technology or spot on there right. Just give them something even easier. They actually change their habits. But but where the file will continue to play a role is the enterprise community particularly the ones who cannot hire software engineers to modernize infrastructure as a code that's actually still a good part of the industry. It's the the traditional vmware based. 11:43.54 greybeards Yeah, yeah, exactly I mean I know I came from the block world long before Files even existed. So yeah, it's it's it's something that's and and embedded in in what I do but to a large extent. You know if I'm doing ai and all types of things. 11:45.39 AB Like a vdi gift quickly. Ya. Yeah I don't. 12:01.76 greybeards It's all it's all based on objects someplace I mean it's all sucking in objects maybe coming into a file or a Csv file or something like that and and being processed there but in the real world guys that are doing aiml. It's it's all objects. 12:15.43 AB Yeah, exactly. 12:17.38 greybeards It it seems to me I don't know you look at Kubeflow or look at ml apps or something like that it it seems like it's getting all its data from object stuff. 12:22.61 AB Um, yeah, the files will stay by. It's like mainframes is still a profitable business right? Files will be there Files and blogs. 12:28.87 greybeards Yeah, tell me about you know mean io doesn't work on mainframe. Yeah, it probably does with a Z something Kubernetes thing or something I don't know linux. 12:38.88 AB Actually surprisingly there is a there is a native board for for power architecture and there is there is actually a startup now. They I think it's model line or something they use min Io to modernize mainframe applications to become Cloud compatible. They use me Io there. 12:55.20 greybeards There you go. 12:57.74 AB It's an market that I I don't have much expertise I Love rather leave it to partners. 13:00.13 greybeards Right? right? right? right? Well you mentioned the cloud earlier on a b I mean it seems like the yeah the cloud has always been to a large extent object oriented and you and you may just say statement there I have to go back. It says Aws actually started with s 3 alone. 13:16.34 AB Um, yeah, that is actually true Ebs was not yeah when they. 13:17.20 greybeards Is that what you're saying I never I never saw that never realized that that's interesting. 13:22.52 Keith Townsend Yeah, actually the the that's pretty interesting I talk about that a lot you know people think about e c two one of the you know most common if not still the most commonly used service outside of s three I remember when Aws announced their or. Amazon announced their cloud services aws and the service was s 3 and I checked it out and like why would I ever won I said why would I ever use this. So take my advice with a grain of salt I was obviously wrong wrong about that along with you know, probably thousands from others you know, but yeah, yeah, the s 3 is is the. 13:54.32 greybeards A thousand other things I know. Yeah. 14:00.97 AB Yeah, after all the clothes. Yeah. 14:01.14 Keith Townsend Oldest service out there. 14:01.68 greybeards And it's the most popular. Oh. It's interesting. That's interesting. So what about the multi-cloud there's ah you know the problem today with enterprises adoption of cloud is that you know occasionally aws or azure google go down and and yeah, yeah, I need to have services that. Now span clouds like I have services that span data centers in the past in order to keep up where does min io fit in that framework of of you know, multiple cloud operations. 14:28.49 AB Yeah, yeah, so when Arablis goes down when Google all these clouds goes down actually I do see tweets in the community that I am I am safe I'm running minau and I did not get affected. They do talk about all this right? But the reality if you look into. Ah, clouds uptime is definitely higher than most of these data centers. They run themselves I don't think I don't think that uptime is a big deal. All all infrastructures eventually go down here and there. But if I'm an Amazon customer. 14:52.85 greybeards Really okay. 15:03.25 AB I would still feel confident that their engineers are competent enough to bring it back faster than than my engineers of course my engineers I I have confidence but in in general right? But a. But I think the real reason why multi-cloud is happening is a it's not even because they had a clear strategy today. Most. Cvos have a strategy and mandate that they have to be multi-cloud ready. But even though cvos end of the day they will tell we did a large contract with Google and then all of us standardized on Google but and you can see that most organizations have that exclusivity because they have to get that discount and they made the commitment and they. But the real reason why multi-cloud is happening is because the developers started building applications as microservices and they containerized everything when they containerized everything they naturally brought in kubernetes to orchestrate these containers. They. 15:56.60 Keith Townsend Is the. 15:56.92 AB Detachched their application stack from the public cloud and they looked at public cloud as between asking my it to provision de or super micro servers here on a Aml file in um, in a moment I can provision these servers I they left it and went to the public cloud it was more of infrastructure as a service. And then they went there object storage was seen no different from mongodb or elastic or kafka. It's the blob store they adopted they brought in they they brought in their software stack containerized and they pushed to the cloud and overnight they pray my management told I have to go to Google cloud. Redeployed the software stack on boload it happened to multi-cloud happened as ah, as ah, as ah as an after. 16:33.71 greybeards Bingo. Yeah. You think you think multi-cloud is there because Kubernetes and containerization occurred is it is like got I'm not sure which is first here in this in this environment. But even if you're running a cluster. Let's say an Aws and a cluster in Google cloud the data. Is a different question I mean so the data has got to be sitting someplace in this environment. 16:59.15 AB Um, yeah. 17:00.17 Keith Townsend Well right I want to and a dragon. Maybe I love love your feedback on this is I think I have a $2000000000 proof point here just read this morning jp morgan Jamie Diamond was on the investor call I'm assuming yesterday day before of this recording. 17:08.72 AB E. 17:19.41 Keith Townsend Talking about how Jpmorgan has invested $2000000000 in cloud much of that spin going to the data center to enable cloud. So I think to a b's point. The cloud experience is what most businesses are. Ah, driving course I yeah idc stat 88% of enterprises want to have ah be able to repatriate static cloud workloads but have that cloud capability and that starts with the underlying store just is the great barriers and storage we believe in storage and. 17:40.12 greybeards Searching for a. 17:58.88 Keith Townsend You need this, You'd be able to store storage across platforms to be able to do that. 17:59.74 greybeards Intensely I might add So how does how does min Io Facilitate this you know, storage residing multiple platforms and stuff like that. 18:03.70 AB Um, yeah, yes, yeah, yeah, yeah, yeah, so this one and combining the previous question too that did kubernetes on containers drove this strategy or the other way around right? what. Our bet early on was when we when when b came in there was already Google cloud and azure starting their journey and Amazon S 3 was like the standard when we saw they were each of them were incompatible with each other and then outside of the public cloud. 18:30.92 greybeards Right. 18:38.94 AB Htfsan as anything you looked. It was a it was a array of standards every one of them incompatible with each other. We knew that this was not going to be the way in the long run everything will look like a Ws or it is a Ws itself that was that was inevitability. So We knew that. Given enough time the problem will will will be solved right that but then we can Fix. We don't need to fix the compute side compute site when we started from Cloud foundry like a mesos to Docker Swarm compost. There are There are so many standards from plant. Yeah have yeah we saw kubernetes. 18:58.83 greybeards Saw itself? Yeah, and. 19:11.24 greybeards There were plenty of orchestration solutions. Then yeah yeah, yeah. 19:17.33 AB Was was better positioned. It was written in the infrastructure language of choice as go and they understood the community sentiments better. They drove? Well it was a better idea to declarative model. We saw that that compute side will be solved but the data side this industry for switching. From posix to object object api like s 3 itself is a monumental task. It happened finally now at least it's happening and it's happening on all the emerging markets. But for them to go from posix to multiple Api standards. That's not going to happen right? That's where we saw that if if. Instead of releasing yet another open source standard. It's okay for us to stick to s 3 Api Amazon won't be unhappy about it. So we we chose to promote s three api as the standard across all cloud and mina who's position is mino can run inside. Aws. On our post. It can run on Google cloud even rancher ah anthos or bmware run through everywhere s 3 cannot and that was our bet and we knew that in the long run multi-cloud will be inevitability and we would be able to help the the community at um at a giant scale. 20:16.12 greybeards All this stuff. Yeah. 20:29.89 AB If we don't do it. We would fail Anyway, we focused on the application application developers to help build a powerful ecosystem and that paid off. 20:36.78 greybeards So you so you solved the Api problem with with with taking the s 3 bet and it was a successful bet from an io and and you guys benefited from and more power to you but the question still remains a b where's the data. The fact that I use s 3 to access. It is the right thing across every cloud and every on-prem environment in the world. Absolutely. 20:55.97 AB Yeah, yeah, yeah, you know they way it like now I have ah plenty of data points to actually understand what's that What's all what's happening when a customer say comes from awss 3 to minavo or say even they they. They have minive on-prem and they went to cloud and they deployed min iu in the cloud. They actually don't move the data around because the new data that's getting produced is more than all of the historic data combined so they actually don't move the data which is expensive and time consuming they they build. The new infrastructure and the new data goes there and they're all kept in silos a they some organizations choose to centralize some organizations choose to go decentralize that model. But overall they never move data and. 21:43.80 greybeards So yeah, you're saying that the data ends up being distributed or partitioned across these multiple Cloud slash on-prem environments based on what application started in that particular environment and what data needs they had at that point is that how it plays out. 21:57.24 AB Now correct correct because you can move the application code but not the data but you can't move the application code if you're stuck with the aba. 22:05.56 greybeards And the way it supports the multi-cloud is that you could have let's say an application running on-prem and reference a url which happens to be in Aws or Google or whatever and and still access the data right. 22:17.44 AB Yeah, in theory. Yes, right? But then what? what really happens is that see even when but then if they pick one cloud and deploy min io they actually deploy in multiple regions across the world and applications wherever they are. They are also global. They tend to. Pick the one that is in close proximity. Ah, even though s three api of miniai work is Https and you can access from anywhere. The bandwidth costs are not the same right? It's more than the storage infrastructure cost. 22:44.50 greybeards Yeah, yeah. 22:46.69 Keith Townsend So Maybe it sounds like what you're talking about the value necessarily isn't in building Multi-cloud apps but having a Multi-cloud operating model in which ah you can adapt or move your workflows whether you're developing ah developing. A point of cell application in one Cloud or a data analytics platform in another the way that you address your data is not changing across public clouds. 23:15.88 AB That is very accurate. Yeah, that's very accurate. In fact, they all they care is their are softwareers. It's in simple plain terms like in developer terms right? They just want their applications to be containerized and it that it's It's what vmware envis and that software defined Data center. 23:15.93 greybeards Or the on-prem environment. Yeah yeah, yeah. 23:34.65 AB It's more not more than the infrastructure layer. All they care is their application stack in the Ml file if thick if I can take my entire software stack and roll it out to any Cloud on Demand I'm good to go and that's actually how they build it on miniqueue go to Cicd environment and then goes to production environment that Pattern has been Followed. Always. They don't want none of these guys actually are building application connected to EkS or a case and all their data so storing on stored on s 3 They don't build application inside the Cloud They build it elsewhere but even their day one launches on s 3 born in the Cloud applications. Those applications are not built in the Cloud. And if we find that all they care is their software stack to be independent of any Cloud. What made that possible was containers and kubernetes further. Yeah. 24:20.35 greybeards Kubernetes containers and all its stuff you mentioned mini cube I was going to try run mini cube on my max here max cluster. But that's another story. So Kubernetes is the key to to multi-cloud as you see it right. 24:21.62 Keith Townsend So. 24:26.37 AB Yeah, is. Um, absolutely. 24:36.00 Keith Townsend So ab what you're saying is pretty much in line with what I found at Kubecon last year as I was sitting at the launch table. It's one of my favorite spots to sit doing conferences listening to ah this this like modernization of the platform team like. 24:45.30 greybeards Yeah, yeah. 24:52.22 AB Um, yeah. 24:54.29 Keith Townsend Before the platform team might have provided vmware vspeer and they provided vms now the platform team is like this weird mix of these traditional infrastructure operators with developers sitting inside of that team that is is. 24:58.60 AB Um, yeah. 25:13.70 Keith Townsend Consuming services like men I O or deploying services for men I O to enable you know I can call it more of an enablement team that enabling the developers that's solving the business challenge challenge. They're kind of a shim between. 25:24.30 AB Um, yes. 25:28.61 Keith Townsend The developer solving the business challenge and the Cloud provider is is that what you're seeing. 25:30.75 AB Yeah, absolutely, you can see right here the change to go to cloud. It cannot start with it. The problem is that the cloud is incompatible with the enterprise I right? file block Vm if you take that stack and retrofit in the cloud it it doesn't run or it runs very poorly. So you have to involve developers to are. It's not just a matter of automation right? like shefer and puppet tried that it's not that case this time they have to rewrite application to go cloud native often. They're finding that it's cheaper to rewrite than retrofit this time all the organizations that have become cloud ready. They involve developers into the mix where it became opscentric dev opcentric. They worked hand in hand those of organizations succeeded the rest who resisted cloud actually and resisted developed they who are claimed that those are developer tools are complicated what we found was the I there was a wall between it and. The development team. The development team went to cloud nevertheless. 26:30.46 greybeards I Never look Back. So I mean I think that's you know, taking an enterprise application and and and and making a cloud version of it. There's this lift and shift discussion or or refactoring and reimplementation kind of thing or redesign altogether I Guess a solution. So you see does does lift and shift work or is it not. 26:52.63 AB He doesn't actually the the biggest biggest proof is vmware itself tried in their first version of their cloud was vmware as a service right? They took vm I give you same vmware as a hosted offering that looked like just outsourced data center. That's not what cloud is. That's where customers basically say their version of cloud is it has to look like Aws right at the the Aws experience. That's what I think Google and Microsoft understood and they they did not give the same same old software as a service they it's it's not just about automation right? It's fundamentally incompatible meaning. Throw away all the legacy. The biggest advantage of cloud is break the legacy systems throw them away. We can build modern infrastructure like how we built it for ourselves like that was Amazon's message that resonated you you completely took that value away and if you brought back legacy. And it's no better. 27:48.80 greybeards Ah, ah we have to have to have a a lengthing discussion about that off live. But um, so so where does opinionated solutions fit into this Multi-cloud environment. It was a. 27:53.99 AB If. 28:04.11 greybeards It was a word I have said the look up when I was I saw that. 28:05.78 AB I Think the so the opinionated is too broad right? If opinionated is in the form of stack like we talked about La stack in the dot com times. Still people tried to do this I Recently last week I heard came across merk some stack like that. 28:17.19 greybeards Right. 28:25.74 AB I was like what what is that I didn't even know what it was but then it's like mongodb something but what I find that the opinionated stacks in the form of pass. It never worked. It's something that even Amazon and other cloud players gradually increased providing the instead of giving you opinionated stack. 28:28.87 greybeards Yeah, yeah. 28:43.18 AB Like google actually the very first version was based on python based Google app engine type model right? It was more closer to pass. It did not work it in the what we are finding consistently all the time was the the developers don't like opinionated stack. they they they they want 28:47.46 greybeards Right. 29:01.84 AB Building blocks. So they can compose their own application infrastructure stack for for this is something it this this this topic alone requires like a whole day discussion that you can see that the biggest success ever happened in the past world is heroku and it's not a big 1 right? So it. It was a small exit. It. Still there but past never work because developers don't like opinionated taxs but having said that we like being opinionated in the sense when when like when early on when we started minao a community was like hey why not use Swift Api because that's open source. Why are you promoting s 3 apis three api is not even a standard and its proprietary. My point was pick one you want Swift Apa or you want s three api I'm not going to do both and the answer was very clear sometimes you will come and ask actually they still ask why don't you add and fs api give file block an object. 29:46.67 greybeards Yeah. 29:56.71 AB And I still tell them the same thing that if I it's if I added a legacy protocol the all the advantage of s three api is gone the advantage of s three Api. The reason it is incompatible is because Pose X is a legacy if I did Pose X Plus Plus it's actually minus minus I'll end up giving you. A media car objects to age and a terrible file system not because I don't know how to build a file system that was my previous project right? But here is what it is. You. Do you want that? No one will say I want a media card product. 30:21.20 greybeards In that. 30:28.84 Keith Townsend So I Love this opinionated view at the Laur stacks where developers actually don't care about how you implement the specifics of details. They just kind of want to consume storage. How are you seeing that. Being kind of validated in the market with developers. 30:52.64 AB Um, I'm um, okay, can you please repeat Keith I'm not sure if I followed. 30:57.20 Keith Townsend So the the rephrase the question you know developers when they approach storage when they approach kubernetes they want a provision storage like what are some of the pain points. They're realizing like oh the regular kubernetes providers are not enough. 31:02.44 AB Um, yeah, yeah, yeah, yeah, oh yeah, that's actually there is a lot of confusion in this topic right? they they I'll tell you from 2 different angles that. 1 if you talk to the actual consumers of the storage itself. They actually don't even call it storage. They know, um, most of these they are all developers who are dealing with the data they look at it as a data store. They minau for them is no different from Mongodb or elasticsearch is just that. If you are talking about metadata type data that you want powerful query interface you would put in a database if you have blob data and you want lots of persistence you would put in an object store that they look at Miniu as an object store and that is how the consumers of the cloud who are the application builders data engineers data architects a ml data scientists. Located it as just a data store period. But if you talk to the infrastructure people there particularly if you talk to the storage vendors they brought in San And Nas and then wrote a Csi Adapter and they all want to look cool. These are same old appliances. They suddenly become. The Kubernetes is ready and they claim that they are Kubernetes native storage and correct Csi. But that's not what actually they the the kubernetes storage is about right? The application there is a big disconnect there and a lot of con confusioncian too. 32:18.41 greybeards So this is all the persistent volume stuff that was through Csi and all that stuff. 32:34.47 AB Every one of these modern distributed data stores if you look at who wants this the traditional sand nas in the cloud san and Nasa are considered legacy. It's only meant to bring legacy applications that cannot be rewritten as a stop gap by yourself sometime. That's when you will go for Efs or ebs. Otherwise. Imagine like snowflake written in ebs or efs they would not have taken started only right? they that the csa providers are meant to give you legacy compatibility. But if you are talking about a modern application. Even the databases themselves right? that they are stateful sets where would they store. Look at every one of the modern distributed database. They have gone scale out and all they want is a local persistent volume like previously. It started with host path and then came local volumes. But even the local volumes does not have a Csi driver which means it cannot be dynamically poor provisioned and I saw that there was a problem in Kubernetes. They there are no csi drivers to manage local rights local Jbots which is what every modern distributed data store Data Processing frameworks are built on not on sand and Nas we may recognize this un brought in piece and direct for their tonzo environment but outside of it. They this is actually an emerging area of discussion like openabs came with the local Tv and I think ran long har from rancher also has something like this I like I needed this for mean iu whether or not I solved this problem for the rest of the industry so we wrote something called direct pv it's a direct persistent volume. All you want is a bunch of local drives automatically provisioned and managed for Kubernetes through a Csi driver. So. It's just a volume manager not a storage system storage systems are distributed data storage inside. 34:17.65 greybeards And so so drug pv is access through a Csi and plugin or or or what. 34:23.88 AB Yeah, it is actually a Csi driver and you give direct pv all the local drives and when you ask say if I'm running. It's not just menau right? say maybe I'm running Elasticsearch and elasticsearch makes copies of replicated copies of their data sets and on their local drive. But long- term persistence they would put it on object storage. Let's leave that aside just to run elasticsearch if you brought in San Or Nas and put a scale out system on a scale up architecture. It wouldn't scale and it'll be inefficient. All Elasticsearch wants is local Pb but if you use just. Kubernetes provided local pv there is no csi driver so you have to manually precreate these volumes and then provision that's kind of inefficient. It break the automation if you use dere pv when you when you provision your elasticsearch or kafka anything right? anything distributed when you want these volumes when when take when. In elasticsearch says I want 10 tb on on 8 nodes each each each of these 8 nodes 10 Tv local you make a volume claim and direct pv will give will run your elasticsearch containers with the host local. Yeah localfin italy exactly just local right. 35:30.90 greybeards Across all the nodes across all the storage a so I I'm trying to understand here. So let's say I have min Io using my local storage or defined for using Direct Pv for local storage. But if I want to access like object storage sitting out on the web and I'm a container. 35:47.17 AB Um, yeah, you don't need it. Yeah, you just use object store s three api just like Redis or mongodb or anything else when you access data services. Yeah, the actually this is a interesting segue into. 35:49.11 greybeards I still just use a Url I don't have to do a persistent buying plane or anything like that right? exactly exactly? yes. 36:06.53 AB What is disegggregated storage the industry talks about disaggregated storage if you talk to Application developers they will tell you disegggregation is between stateless microservices and stateful data services that is data data stores. That's what they mean by Disegggregation. That's what Cloud thoughts about disegggregation talk to the storage vendors. They talk about disegregation between the drive and the storage systems The data stores. That's a we differently inbm your fabric here. 36:27.79 greybeards Yeah, yeah, and the compute right? and all it stuff. Yes, yes, yes, yeah, it's it's it's it's bizarre it's bizarre. So um, guys where was I going with this It's the database stuff. So if I'm using like mongodb or something like that in a containerized application I'm using a mongodb api I'm not I'm not creating a a persistent volume on a Mongo database anything like that right right. 36:49.29 AB Me. Very There is no the the data be underneath the mongodb mongodb would make a persistent volume claim and then like that's where direct Tv look Lt Yeah yeah, the application won't see it. 37:04.98 greybeards That's where it stores its data and stuff like that and yeah again Kubernetes seems to be the key to all this stuff. 37:12.12 AB Um, it is now the the Api ah of infrastructures. It's the Apa standard for infrastructure. 37:18.89 greybeards Yeah, yeah, declarative and all that stuff it's it's bizarre. 37:24.17 AB Some some it has to happen right? Otherwise like it's very hard to build nowadays like it's the it's not the installation rates. How do you operate at scale operations means every day you are rolling out New updates operations has become the most important problem. 37:36.90 greybeards Um, yeah. 37:41.48 AB And without standardization. It's going to be very hard. 37:41.96 greybeards Yeah, yeah, exactly exactly exactly. So um, recently, there's been some new funding for Min Io or series was closed is that true. 37:54.42 AB Yeah, then I that is actually a the funding is closed I it's a it's a quite large round and it's the announcement is not public yet I did you hear from Jonathan. 38:08.23 greybeards I haven't heard from Jonathan but we can schedule the podcast publication to be whenever you want or we could delete this section if we need to it's up to you. 38:12.96 AB Um, yeah, yeah there I'm not there is an embargo on it I it's in a week or two I think I think if you check with Jonathan then yeah, yeah, we I can I can share the information but there is an embar on it. Yeah. 38:22.66 greybeards I will. Okay, what? So Why don't you go ahead and and and provide the information on on the the round I guess. 38:37.67 AB Yeah, the the the existing investors preempted preempter term with a term sheet that Intel Capital this is Pat just that Pat glsinger himself presented a term sheet that was like a very humbling to receive that e. Ah, a then softbank participated along with the existing investors. It's a 3000000 round at a billion dollar valuation yeah and it's series b and we are still a. 39:03.29 greybeards Um, ah ice. 39:09.53 AB Small team at the time when we when we got the funding it was. We were like 40 member team. 39:13.86 Keith Townsend Oh how does that compare to like the overall market like this this is a crowded space when Ray hit me up to do this sponsored episode with me I'm like oh another kubernetes storage provider like how does this set you apart from your competitors when it comes from a financial. 39:20.42 AB Um, does. 39:31.90 AB I I never saw the amount of money you raised the number of people you hire to be a real measure of success right? It's still the case you can see previously. We raised only 23000000 series a and it didn't it didn't slow us down. We were accelerating like crazy and still. 39:32.40 Keith Townsend And finances. 39:50.77 AB Like today like own one point one million docker pools a day and it's not just you can some might tell yeah, that's it's just from a darker hub alone. It's not including the private repositories and all other reposities right? They be every. 39:54.42 greybeards One point one million did you say one point one million doctor polls a day. Yeah I want yeah of bazar. 40:05.57 Keith Townsend Yeah, I'm not um um um I wasn't supposed to say this but you know my private infrastructure that I give access to to my team I found the men I owe vm I'm like whoa look I don't know if Ray infitrated my team or not. But. 40:11.85 AB Um, and. Um, yeah, it's hard to stop at this point it is. It's just there but we knew that there is only 1 way that this market will consolidate and there will but I actually like to go into a crowded market. Because the market is established the hardest part part is actually to go do concept selling and create a new market. It's easier to go into an established market where there are so many players you create a superior product like fine and superior for me is not more features are more beefy shiny it is fine craftsmanship listen to the users. 40:41.30 greybeards Now you tell me. 40:53.99 AB Them just what they want and that fine touch and finish you get to connect to the users if you put that product first academy attitude. It's hard to go wrong and that was the reason that we were quite confident that. The market is so big 10 years from now. No one will blame me that I picked the wrong market because data is going to be everybody's problem if I built. Yeah, yeah, yeah, it kind of fade off right? It's really simple, simple ideas that Unfocus allowed us to get here. 41:14.49 greybeards It is everybody's problem. It will be everybody's problem for the foreseeable future. 41:25.17 greybeards Yeah, yeah, no yeah I you're in the right place The right time I'll say that much and so I don't so after. 41:29.29 AB Yeah, yeah. 41:36.64 greybeards But what comes next in your world of Min Io I mean you've got you've almost conquered you've pretty much conquered s three api Compatibility you're sitting here with ah with a pot full of money and and who are you going to go after next. 41:42.90 AB Um, yeah, he so it a. From ah from our thinking point of here we never saw anybody as a competition we were eternally unsatisfied with our own creation is like every artist if you look at ask them about their ah past work. They're actually not happy in spite of being a big hit. We are just we know we can do better and good part is software. There is always version 1 version, 2 and version. 3 can keep on improving so that part is all I'm thinking right? But on the other hand now should I should I introduce the next product our skill is to actually we are creators and the product creators we can go go do Mineo of this mino of that. And we can do all that right? But then they on the other hand mean is but while we got the land grab the commercial journey is just starting and if I launch a new product. It will create a the the branding part problem product creation is easy creating a business around it and promoting the brand is. 42:40.25 greybeards Um. 42:51.94 AB Very very hard I am better off like starting like accelerating on the commercial journey right now. The customers are actually coming to us because we are deeply in production across many of these enterprises and when they come in I think it makes more sense for me to help them. Ah, run their infrastructure at scale with more ease and better security and better operational disability. Those are more important to me and that's actually where basically if you look at amazon s 3 versus us 1 thing Amazon themselves admit that s three has become very complicated. And a and mini was case while while the industry talks about how easy miniu is I actually think that it is not right it we can still there is a long way for us. Yeah, make it easier I think I will never go wrong if I do that as compared to launching new products. 43:33.87 greybeards Make it make it easier. All right. 43:42.37 AB But I think one area I would continue to invest on the next big if shift if at all from mean io to the next layer is what you do with the data now that you stored all the data in Miniu we became the data infrastructure for your organization. What you do with the data is more important search type functions and why. Unlocking the value of the data through ai machine learning functions I think that's an area I would definitely invest. 44:05.90 greybeards Ah, interesting. Well this has been great Keith any last questions Ray B before we close. 44:11.99 Keith Townsend You know what? the 1 thing that I had on kind of my list. Ah we talked about vmware awful lot. But we haven't talked about kind of ah the world of Vmware still exists. There's still the you know the £800 gorilla in the market any features integrating with. 44:17.99 AB Um. 44:25.99 AB Um, yeah, ah. 44:31.47 Keith Townsend Existing vmware Landscapes We we should talk about. 44:35.12 greybeards Besides the data persistence platform. 44:35.33 AB Um, yeah I think the so the so vmware actually has has been investing quite a bit in this space. They to make it compatible and well if you look at their strategy right? Let us make vmware compatible. 44:52.66 Keith Townsend Um'm sorry ab I get I got the question question real I wanted to make sure we were talking about the object storage I'm looking at the ah the notes from the what we want to hit on this Os images application artifacts and snapshots and I'm actually working on some vmware stuff in the background and snapshots gave me the shortcuts about. 44:57.77 AB Um. 45:09.88 Keith Townsend Ah, ah vmwas I'm going to restate the question so 1 use case that we haven't talked about has been kind of the standard os images application artifacts snapshots, backups et cetera like the the the mundane task of that we've done in traditional storage arrays. 45:11.44 AB Yeah, yeah, great. Yes. 45:25.36 AB Um, one a. 45:29.88 Keith Townsend Where's min I O in providing the that that type of capability. 45:32.15 AB So that it started out on the artifact side as a common use case the like from from jfrog to but I think this is harbor container major repository like all of them just storing the container images that itself. It is clearly object storage is the backend for that and manyu grew quite popular there and what we started seeing was it's not just about container repository they they are constantly building new container pages every day they make a new patch. how how is how is this new commit tested. They build a new container. And the whole stack. Whatever the code has changed all becomes results in new containers and they get tested in a sed automated framework and that results in a flood of container majors used to be vmm images. That's why you needed all this copy data management secondary data management that has become more of now artifacty store. And that is and that market also came to objects to pretty much all the cicd frameworks underneath. You will find that if it's not min io has to still be an object store. Probably if it's not Minio. It'll be some kind of public cloud even the models there another one that's growing out of that. 46:40.30 greybeards Yeah, yeah, yeah, yeah now. 46:47.85 AB The ah application artifacts are container images but not all but now we are seeing a new class of artifacts that are just machine learning models. 46:54.44 greybeards So the models themselves are becoming not quite containers but container like solution or so artifacts. 46:54.92 Keith Townsend He. 47:00.00 AB Yeah, they they are kind of the data continues right? exactly? Yeah you yeah, that's exactly I think this. 47:02.20 Keith Townsend And then we get the power of the platform with that So versioning E Cetera with that other thing. 47:08.23 greybeards Yeah, yeah, yeah, the whole Ml apps thing is is all about that all maybe anything else. You'd like to say to our listening audience before we close. 47:19.40 AB Like the questions were great I enjoyed the discussion we touched upon everything. 47:22.36 greybeards Okay, good. Well Abe thanks for being on our show today and thank you ah to min io for sponsoring this podcast. That's it for now by keep and by a b until next time. But bye. 47:30.30 AB A thank you gate and. 47:33.90 Keith Townsend By rate. 47:35.30 AB And yeah, thank you they thank you? Keith.