Michelle Coppens
00:02
Hello everyone and welcome to today's webinar titled the benefits and drawbacks of a multi cloud strategy. Our speaker today is miles brown senior cloud and devOps advisor at Tech data exit certified
00:20
Miles has over 20 years of experience in the IT industry across a variety of platforms recognized as an AWS authorized instructor champion.
00:31
And a Google Cloud Platform professional architect and instructor Miles has delivered award winning authorized it training for the biggest cloud providers. All right, lots to cover today. Before we get started, let's talk about the webinar functionalities.
00:50
As you probably noticed your microphones are disabled. So if you have a question, please enter them in the Q AMP a box at the bottom of your screen will have a dedicated Q AMP. A session at the end of the lecture but post those questions whenever they come up
01:07
Today's webinar is being recorded and we're going to send a copy to everyone who registered
01:13
Will also share to promos at the end of the webinar, so stick around to learn a little bit more. All right, let's get started. Miles, you can take it away.
user avatar
Myles Brown
01:23
Alright, thanks, Michelle.
01:26
Well, as I mentioned, I'm me, you know, AWS and Google Cloud instructor have a pretty good familiarity with Microsoft Azure as well as exit certified it's partnered with all three vendors are the big three.
01:41
Cloud providers and we do authorized training for all of those, as well as, you know, other vendors like IBM and Oracle, who also have cloud.
01:49
We also, you know, our partner with some of the DevOps kind of tooling that we might end up talking about at the end of the quarter the webinar, things like talker and Cooper 90s. So we have
02:00
We have training on all that and we'll talk a little bit about our training right at the very end.
02:04
But for now, let's jump in. We'll talk a little bit about this concept of multi cloud the book, the benefits and the drawbacks. That's actually
02:13
That's probably half of what I'm going to talk about, then when I'm really going to get into is the tools that make multi cloud easier.
02:19
So we'll look at those drawbacks, and see how can we, kind of, you know, less than those drawbacks by by the right choice of tools. So let's talk a little bit about hybrid and multi cloud to begin with.
02:33
But, but first you know really the timeline of the major cloud providers, since the, you know, sort of, early 2000s, people have started building kind of private clouds in their own data centers using things like VMware
02:49
But, but really it was around 2006 when AWS became what it is now.
02:57
Some of the some of the services predate that but it is by far the most popular and mature, you know, public cloud provider.
03:05
Azure is definitely second place, it came up in 2010 and Google Cloud came out in 2011 so they're playing a little bit of catch up.
03:13
Google Cloud is is sort of very distant third but they're really coming on strong over the last little bit. And like I said, you know, IBM and Oracle. They also have public clouds, but
03:25
You know their, their market share is is is diminishing because you know they are very niche, whereas the big three are becoming bigger and bigger every year.
03:36
And I think early on, people would pick one vendor and for most people, they picked AWS, because you know they were the most well established and
03:46
And so you would have sort of one cloud and you would work with it. But then that thing started to seep in where they said wait a minute.
03:52
I'm starting to use a lot of AWS specifics, or if you're a Microsoft shop you start using some of the Azure specific stuff and you say, wait a minute. Am I getting into the same situation. I got into with
04:04
You know enterprise vendors in the past where I got vendor lock in. You know, people are stuck with
04:10
Oracle databases that are very expensive and they push you forward and say hey you know we're no longer supporting this old version, you got to pay it upgrade you know
04:19
And so you know that concept of multi cloud started to pop in
04:23
The other part of multi cloud is the fact that people a lot of organizations already had an investment in a private cloud.
04:31
And they were using that they didn't want to scrap it but then they did want to use some of the features of a public cloud, so that idea of a hybrid one is very popular. And so just last year right scale, which is now flex era.
04:45
They did a poll of, you know, quite a few cloud developers and architects.
04:52
And what they found was that 84% of enterprises with over 1000 employees have a multi cloud strategy now when they say multi cloud. They include hybrid. And that's the bulk of it right
05:03
The idea that you're going to have multiple public clouds already not that many people have that but but when you look at what people are looking into it is very, very popular.
05:14
So, so let's talk a little bit about hybrid and then we'll, we'll concentrate the rest of the webinar on multi cloud. So hybrid is very popular because you've got this existing investment and a lot of times something like VMware
05:26
running your, you know, on your on prem data centers. And then you start to use some public cloud.
05:31
And somebody will start off just using the public cloud as sort of off site backup to say hey you know we've got this data here. Let's also put it into AWS S3. And that's a very durable place to store it, it's offsite backup.
05:45
If something tragic happens in my data centers. I haven't lost that data.
05:50
And they do a good job of securing it and you know and and very durable storing it and
05:57
Some people use a hybrid cloud. And what we call Cloud Bursting where they're providing them primarily running their workloads in there on prem.
06:05
But in their private cloud. But you know when things get busy. They'll use one or maybe even more than one public cloud provider for excess capacity.
06:15
We talked a little bit about this. I did a webinar yesterday, just on kind of comparing the different clouds, and I mentioned, you know, one of the companies that does it sort of Cloud Bursting at zoom
06:25
Which had to do this in the last month. Very much so. Because you know their business went through the roof.
06:31
And you know they they are able to run things normally in their, in their, you know, private cloud in our own data centers but they definitely need to use AWS and Azure when things get really busy.
06:46
So that's hybrid cloud. When we talk about multi cloud, you know, for our purposes today let's let's just concentrate on when we talk about using more than one club public cloud provider.
06:57
And why would we do that. Well, sometimes people want to run workloads in a particular cloud depending on availability or latency or price.
07:06
Sometimes people want to redundantly copy data across multiple providers and sometimes people want to utilize different services for each cloud provider.
07:16
And so that's sort of the three different ways that people end up using multi cloud that idea that
07:23
You know, I want to redundantly use more than one that's that's to say, well, hey, what if you know a particular vendor gets attacked or has some sort of major outage. You know, I want to be able to run things in multiple places.
07:39
But mostly it's, you know, trying to pick where do I want to run this workload, because there are some differences between these different vendors and
07:50
And so that's really, you know, sort of getting us into that talk of
07:53
Why do we want a multi cloud strategy. And so the first one is that idea. Don't put all your eggs in one basket. Right. We've been we've been burned by that before.
08:03
Where we're stuck with one vendor and that means you're at their mercy. When it comes to price changes.
08:08
And have support for certain services, things like that. And so just that having that capability of saying, well, you know, we're primarily using and I can't tell you how many times I've talked to companies that say
08:21
We're multi cloud, but we're we're cloud agnostic, but we're primarily running in AWS and I hear that a lot.
08:30
So that's sort of one thing just just kind of that idea of being
08:36
Able to move between vendors, probably the biggest reason though is what I mentioned earlier that that idea that
08:45
I want to mitigate against risk of some sort of outage.
08:49
Whether it's, you know, natural disaster takes out a bunch of data centers or whether it's, you know, something, something more like they're getting attacked
08:57
You know, whatever it is that ability to move workloads around or have them running already in multiple clouds and being able to switch over. That is interesting. So I got a question already. We're going to get to the drawbacks, and what the prices are and everything else. Yeah.
09:14
So just continue along with the benefits the price flexibility is actually a benefit and if you have a workload, you know, different cloud providers.
09:24
Often have some of the same services, but their pricing models might be more advantageous to your particular workload in one versus another. Now it's tough to be that
09:36
That flexible to run a particular workload, wherever you want. But, but if you get a sense of, for the kinds of things. I'm going to be doing Google Cloud might be cheaper or AWS might be cheaper, you know, depending on
09:49
What you're doing. So that's another thing that comes in, but I would say the biggest reason I I talk to customers that are going to multi cloud now is because they want to take advantage of each providers sort of sweet spot. What did they do well.
10:08
So AWS has a massive market share, because they were in their early and they have the widest breadth of services. They do a great job on, you know, they do a great job on everything.
10:19
But, you know, things like service compute with their AWS lambda and some of the other services around that.
10:25
Make it very, very easy. If you're trying to take a monolith and break it up into little micro services and you want to run those
10:32
Without worrying about virtual machines and operating systems and file systems and all that kind of thing. They do a great job of that. Now, Azure Google Cloud both have that idea of, you know, sort of cloud functions.
10:46
But I think AWS is sort of the most mature around those
10:50
Microsoft Azure. They have their own sort of things that you really well. If you look at their paths capabilities. That's platform as a service.
10:58
They'll take a.net application, you know, if you're a.net developer, you build up a zip file and you say, hey, I want to take this
11:05
And go and make a nice scalable web application. And I don't want to worry about.
11:10
You know, thinking about where where to launch the EMS. How many to launch, making sure that if one dies, it gets replaced. You know, like, all that kind of infrastructure to make it highly available and scalable. They take care of all that
11:24
But of course actually can do that because you know they're more concerned with you know.net and Microsoft specific stuff. They don't have to worry about every single language on Earth.
11:35
Right now they have added support for other kind of platforms in this way, but that's one of their hotspots. They do a few things pretty well. Right.
11:44
Google Cloud. If you look at it sort of Google as a company they developed Cooper Nettie they develop TensorFlow.
11:51
And then the open source those. So they're they're open source, you can use them anywhere, but Google Cloud does a really good job of those, especially around the machine learning.
11:59
And even though they have a small market share, you know, they have a good chunk of organizations that are really heavy into machine learning.
12:09
And that's because they're building on the very, very deep expertise that Google has around this area.
12:16
And so one of the things that I'm seeing now is I'm seeing companies that, you know, primarily are using AWS, they've got all kinds of workloads running in there. And they've got lots of data sitting in S3 buckets.
12:28
And now they want to start doing analysis on it, you know, and they're starting to look at machine learning and they're saying, Well, we could do it using the AWS services which are okay.
12:38
But look at some of the advantages we have in Google Cloud around that. And so they say, Okay, we're going to use Google Cloud just for that analytics part
12:46
And will bite the bullet will move our data will copy our data from S3 into Google Cloud Storage. And in there, we'll go in bang away at it using the machine learning.
12:57
Options that they have because I prefer those right and so we're starting to see that quite a bit. And in fact, that is a major marketing push from Google Cloud. You know, I went to their Google Cloud. Next, which was their big
13:13
Conference, but a year ago they were supposed to have one last week, but it got cancelled, you know, along with everything else.
13:19
And so last year that was there, you know, they basically were pushing this concept of multi cloud.
13:25
And the even Bill, you know, some some applications around and something called an SOS, which was really based on top of Cooper daddy's to say, hey, you could build this application, you could run it on prem, you could run it in any cloud.
13:38
But you start to run it all out of Google Cloud as far as the
13:44
You know, running the application and
13:48
That is there being strategy. Now, we talked about all the great things about multi cloud, what's the downside and there's a few. And so we'll look at a few drawbacks.
13:59
First off, as soon as you start splitting your spend across more than one vendor that means you're not spending as much on a single vendor. And what that does is it, you end up
14:11
missing out on some of the volume discounts and and enterprise agreements, you know,
14:17
You know, most of these cloud providers have sort of a tiered pricing like like in AWS just around storage in S3.
14:24
You know, the first 50 terabytes that you store costs about 2.3 cents per gig per month. And then the next 450 terabytes. Well, then, that goes down, all of a sudden, now I'm only paying 2.2 cents per gig. And then, you know,
14:38
And then beyond that. If you're doing massive amounts of storage. If your Netflix or somebody like that.
14:43
Well, then you're not even going to pay their the retail rates, you're going to make an enterprise agreement and say, Hey, I'm going to spend a million dollars a year on this so you better give me a very good price. Right.
14:55
And so, you lose some of that bargaining power when you start splitting things around.
15:00
Also, this comes back to the question, well now that I'm using several cloud providers. Now I have to skill people up on multiple cloud providers. Right. And there's going to be more cost associated with just that complexity for sure.
15:18
So that's one downside. The next downside is security.
15:21
And I think, think about that, you have to now understand how does this, how does the security controls work. What are the security controls in each of these cloud providers.
15:30
And if I'm doing multi cloud stuff that just means more surface area to to secure. So not only do I have to learn it. Now I have to, in practice, make sure that all my security controls are set properly in all these different places. So security is definitely a concern.
15:47
You know there are tools that they have for security moving data in and out of the clouds and certainly between the clouds, you know,
15:55
So it's not really that it's it's less secure inherently but it's more surface area means less secure because there's more points of
16:06
ingress and more more places to have to secure and then just that complexity in general that comes from having multiple management interfaces multiple configuration kind of techniques.
16:18
Multiple deployment tools, all of that makes makes this multi cloud thing you know quite a bit more difficult. And if you build an application.
16:28
Is that application, easy to deploy to all three public clouds, probably not. Not known as I get to the tooling to help with this complexity and
16:38
And that's going to be what we try to concentrate on for the next 20 minutes is, what are some of the tools to help make this multi cloud easier and so
16:50
What we'll see is that there's many tools that help these these a lot of these tools are not just for multi cloud, but also hybrid cloud, right.
16:58
And I would suggest that these tools kind of fall into one of four categories.
17:04
And these categories. I mean, I started this categorization myself. I don't think you'll see anybody you know say these are the four, you know, multi
17:13
Cloud tool categories. But let's start just with the idea of provisioning resources in each of the each of the providers.
17:23
Well, what you'll find is that each one has its own unique set of tools and
17:27
So in AWS, they have a tool called cloud formation, where you build templates which are really just text files and either JSON or gamble.
17:35
And you build it out and you say okay, I need to set up a network of the PC that looks like this. I want to launch a bunch of virtual machines which are called easy to, you know, and so it's it's a huge file or you break it up into many files to be more modular
17:50
But if you want to do the exact same thing and Azure will they've got arm which is as your resource manager similar kind of concept in Google Cloud. They have Deployment Manager very similar kind of concept but
18:02
You know, they're similar concepts, but you can reuse the same script and all of them, right. So, so now you got three different tools just to get your stuff going right.
18:13
Now there are some more cross platform tools. So if you take a configuration management tool, you know, traditionally, you had puppet and chef and things like that.
18:24
Well answerable has become sort of a very popular cloud.
18:28
Configuration management tool because they embrace this. And they said, hey, if you want playbooks. That's what you call the sort of scripts and answerable
18:36
For deploying infrastructure in AWS. We got playbooks for that we got playbooks for Azure. We got playbooks in. So that's sort of one option Tara form is a very popular.
18:50
infrastructure as code tool from a company called hashi core and Tara form, it's, it's sort of the idea that
19:00
You're not going to use the same script and say now deploy with this setup in AWS or in Google Cloud, you know,
19:08
It's not like that. It's that you only have to learn one style of scripting and then you could build scripts for each of the
19:15
Cloud providers that you're using, you know, so, so they don't really provide our write once, run anywhere experience, but they are saying started saying you don't have to learn three different styles of scripting at least
19:28
If you want more of a write once, run anywhere kind of experience. That's where containerization and orchestration starts to come in. So if you build your application into a series of Docker containers.
19:41
And then maybe use Cooper Nettie to do the sort of orchestration.
19:46
Well, then you can launch your, you know, set up Cooper daddy's in any of those clouds and then go and launch your application into that.
19:53
And so the application code doesn't have to change when you move from one place to another, the deployment is very similar, but you do have to worry about managing saying Cooper Nettie is in each of the clouds.
20:04
Now, there is some help there because the public cloud providers all provide a managed service for Cooper 90s.
20:11
So I would say that Google Cloud does the best job because they invented Cooper 90s. So they have something called GK he the guru Cooper Nettie engine.
20:20
And so basically the idea if you know anything about Cooper days is that they're taking care of the control plane for you.
20:26
And so they're going to manage that. You just have to think about, you know, hey, what are the nodes that are going to make up your, your Cooper natives cluster. And then you go and start to plug into them.
20:36
Very similar Azure has a KS advertiser Cooper Nettie service and in AWS, we have ek is that the the elastic Cooper Nettie service. I guess is what they're calling it now.
20:49
So those sort of provided, but there is still that thing where you're going to have to learn how to manage, you know, use those managed services in each right
20:59
There are some nice tools for just simply doing the deployment.
21:04
That are that are sort of cloud agnostic as Spinnaker comes to mind this was something built at Netflix.
21:10
And then they open sourced it and so a lot of people are using it now and it says, hey, I want to launch this virtual machine in AWS or in or in or in and it can include you know on prem kind of into VMware setups.
21:24
Or it also incorporates Cooper daddy's and stuff and says, hey, I want to deploy these pods to Cooper Nettie and then you just point to which one
21:33
So that's sort of one of the things we can do. But we kind of we kind of lose some functionality. When we want to be vendor agnostic, right. The idea is
21:46
If, if, if I want to be able to move this workload around then it can't really take advantage of some of the extra bells and whistles that that cloud provider has for me. Right. And so that's sort of
22:01
You know, a lot of times it's kind of platform as a service capabilities that we're losing out on and
22:08
And that's a bit of a bummer. Right. So what we see is that some frameworks have stepped up to provide sort of pass features.
22:18
To your application in a way that I can deploy the application and that passes framework into you know my own private cloud on prem or an NT into any public cloud.
22:30
And so the two sort of main ones that come to mind are open shift and Cloud Foundry and Cloud Foundry is a little bit older open shift has really become a lot more popular lately.
22:43
It sort of is owned by Red Hat now and and it's based on Cooper Nettie but what it does is it says, hey, not only are you
22:52
You know, building pods and the basic deployment unit in in Cooper daddy's but we're going to provide all kinds of services on top. And we're going to do it the exact same way no matter which cloud, you're in
23:05
But that comes with its own costs, you have to now learn open shift and how to how to work in that environment. How to administrate it and then you have to pay for it, too. And so, so there's always trade offs with every one of these options right
23:22
Now at at exit certified we do Red Hat training as well, where were authorized Red Hat training partner training partner of the year.
23:30
They have open shift classes. They also cancel classes because they bought the company and simple as well. So, so we do all those when it comes to Cloud Foundry.
23:39
You know, probably the most popular implementation of that is Pivotal Cloud Foundry and we are partnered with Pivotal. We have those classes as well.
23:47
Now the problem with everything we've mentioned so far, you know, the tools here, they're all sort of piecemeal, you know, okay, I need to
23:55
Deploy some infrastructure. Now I want to build my application up in a you know a vendor agnostic way and deploy it. Well, then you have to start thinking about what, what about monitoring.
24:06
Right, if you look at monitoring each public cloud provider has its own unique set of tools for that. And in AWS, we have cloud watch
24:14
That's a service that gathers metrics from all the other services. It also has a logging component. You can put your logs into there.
24:21
And it's very specific and how it works as your has as your monitor, same kind of idea. So they got eyes, your monitor. They got metrics, they have logs, things like that.
24:30
And then Google Cloud, they had, they bought a company called stack driver and so they called it stack driver for a long time now. They kind of
24:38
Tweak thing they they're getting rid of the names that driver. It's called Google operations platform. And there are a bunch of elements to it, but one of them is the Cloud Monitoring cloud logging. You know, so on, so on and so
24:52
If you're going to launch things into multiple clouds, then you're probably gonna have to learn how each of those works and use different interfaces to look at those or use some third party tool that can connect to any of them.
25:05
Right. And so we see some cloud monitoring tools that are just simply like
25:11
network monitoring. You know, they'll, they'll gather the log files from you say your AWS VPC, your virtual private cloud. And they'll get the flow logs and you'll see all the traffic.
25:23
And so there's some tools to do that. That's where data dog started but then they added in more application performance monitoring and so most of the traditional AP and vendors. So like New Relic and
25:37
App dynamics, you know, some of these tools.
25:41
They added in cloud support. So most of them first started with like VMware and popular private cloud options.
25:50
Maybe, maybe OpenStack, and then they added in AWS and then they looked and said, oh, as pretty popular at that, you know, and so they keep adding. So if you look at something like New Relic
26:02
Is a very popular kind of application performance monitoring tool.
26:08
You know, a couple years ago, February 2018 they added, you know, multi cloud monitoring support and so you can get sort of one pane of glass and to have an example here. So one of its quick to enlarge
26:20
You know, one pane of glass, where you can see what's happening in AWS and Azure Google Cloud and, you know, so the idea is that they have kind of
26:30
Hooks in. So what are they using under the covers. Well, they're using exactly the same tools from AWS or Azure or from Google Cloud
26:40
But they're interpreting those and building nice, you know, kind of canned analysis for you. And then the ability to data in your own analysis on top.
26:50
And so, so there's a lot of those kinds of tools data dog has become a pretty popular cloud monitoring tool. Now these tools are not cheap. By the way,
26:59
Right. So like I said, you know, once you get into this multi cloud adding the tools in is going to increase your costs something like Cooper daddy's
27:08
It's really just, you know, learning it and stuff like that but but when we get into the real tool tools, you know, commercial products, then you're starting to cost money to add in this
27:23
Yeah, so I think
27:27
That's Cloud Monitoring now the next couple categories of products are things like a cloud service broker.
27:36
So Cloud Service broker is usually concerned with, you know, like let's track our spending on each of these clouds. Right.
27:44
And so, so they're going to provide some governance self service so that you know your your
27:50
Your it people and your finance people are kind of happy, right, so that we can
27:56
We can say, well, you need to be able to launch things and you want to launch them wherever. But let's find the cheapest place to run that kind of workload.
28:03
And then let's let's find out you know through one interface.
28:06
You know who's allowed to launch things where, and let's figure out what what the costs are going to be. And then, you know, sort that out over our company and and what groups are in that company, so it's
28:18
It's a, it's that kind of an idea, but it's it's a little smaller in focus. It's really around the
28:25
You know who's allowed to launch where and what are the costs involved and getting a good idea of those costs. What would start to see now is kind of bigger, more encompassing
28:37
Tools called Cloud management platforms and they'll typically provide those same functions as well as real management and monitoring options so you don't have to go and get 50 different tools, you might get one tool to rule them all.
28:51
Now a lot of times these tools they might have hooks in, they say, Well, you know, we've got a little bit of monitoring, but if you want to use some third we want to use New Relic
29:00
Because you're used to using that already for your on prem stuff you can, you know, you can hook into that from our cloud management platform.
29:08
But they're going to add in, you know, automation orchestration, both for individual virtual machines like launching these workloads, you know, these, these VM, wherever, or maybe at the Cooper Nettie level launching applications.
29:22
They might even include service request cloud inventory multi cloud migration multi cloud backup kind of stuff.
29:29
And so that's that's what these sort of tools do now. What are those tools. Well, some of them are the more traditional I Ts.
29:40
It SM tools like like somebody like service now right
29:45
So if you're big enough organization, just trying to track spending on on it is you know everything.
29:53
And so they might use some sort of IT service management tool and a lot of those have added some cloud capabilities to them. But if you look at what are the real top cloud management platforms.
30:05
This was from earlier this year. This was the
30:08
Gartner magic quadrant, you know, somebody like VMware, they have a cloud management platform, but it's it's a little bit smaller and focus it saying
30:17
We're assuming you're using VMware v. You know, for your private cloud. And then VMware on AWS or VMware can also run in in Azure or Google Cloud
30:27
You know they have ways to make those things work. So you know if you want sort of one pane of glass to manage all these things, you know, they have the cloud management platform.
30:38
If you look at the Gartner magic quadrant, that sort of the two biggest ones traditionally have been Morpheus, and right scale now right scale is now called flex era. Right.
30:48
And so they've got some pretty nice tools for managing and governing across public and private clouds.
30:55
And, you know, here's some of the things you can do with their multi cloud management, you know, unified the visibility of public and private cloud resources and
31:05
Leverage reusable blueprints to orchestrate infrastructure and service across all of your clouds. So that's what they're trying to provide is
31:13
It. We're going to give you a way to say, you know, what does this infrastructure look like
31:18
And then you should be able to say now deploy that saying here or here or here now under the covers, you're going to have to do some extra scripting to make that happen. Right, but that's that's the idea behind these tools.
31:32
And biotics got bought by so software. So it's called snow software.
31:37
Cloud bold scholar and then it's 40 more right i mean that's the thing. Some of them, you know, you might consider more of a, you know, just a management tool or or or a full cloud management platform or some of them are just brokers.
31:55
And so, you know, there's a lot of options out there. Ted asks, well, didn't see cloud checker in here. Yeah. Cloud checker if we just go look at that one.
32:09
There's no II in it. Right. Yeah. So here's cloud checker they consider themselves a cloud management platform as well and and they are definitely in that in that running
32:21
They didn't. They didn't make the the Gartner magic quadrant. Now there's lots of reasons why they might, you know, they might not want to participate or something. But again, it's the same kind of concept. So they've got the constant management side.
32:37
The resource utilization and then they've got the automation component and the security, the compliance. So once they get into the automation part that's usually the difference between just a broker that's around, you know,
32:50
Making making sure we could track your costs versus really managing things, you know, and so that's that's really where they come in.
32:59
So like I said, there's about 40 more and they all have sort of what they do. Well, what they don't do well I'm not telling you which one's best
33:08
I don't know, you know. All I know is that I know who spends the most money when they when we go to like
33:16
An AWS event or something and they have the biggest booths and I talked to them. I do go around those trade floors. I mean, for me that's the best thing about going to those events, you know, they've all been canceled this year.
33:29
Or they go to like more of a you know a virtual conference and and what you miss out on that and is that ability to just walk through the expo floor.
33:41
You gotta be careful. Like you go to something like AWS reinvent there's people just running from booth to booth, trying to get swag.
33:47
Right. But that's usually the opening part, maybe the next morning, but then after that later in the week I walk around and I just talked to people and get a sense of what's your tool like how's it different from this tool.
33:59
And so I would really suggest that that that's that's the best thing I think it comes out of those those conferences, because typically those conferences, all the all the sessions are online.
34:11
You know, fairly quickly after that so you, you don't have to go and sit through the sessions, necessarily, you can watch those on your own time. So that's something that I would consider
34:23
We got a couple other calls questions. Can you see MP tools use be used for traditional hybrid cloud management tool. Yeah, I would say definitely
34:33
cloud forms is an interesting one. So, so there's a couple of products like a category of products that we didn't really talk about, you know, when we sorry let me get back to my slide earlier.
34:46
When we talked about the pads frameworks we talked about some of the tools around this Cooper Nettie isn't, things like that, you know, I was sort of concentrating on the
34:55
The frameworks that are really for multi cloud you know you're using some cloud provider. There's a couple of tools.
35:03
That people use just to build, not so much pass but but the infrastructure service like if you want to build your own private cloud.
35:12
And so OpenStack, it's probably the biggest one, and then cloud. Cloud stack is an Apache product. And so there's a bunch of those kind of tools to that are sort of for for building your private cloud. And if you're going to build a private cloud. You know how to kind of do that.
35:34
And then, you know, when it comes to sort of
35:40
Dealing with both the private cloud, whether it's a commercial one like VMware or something more open source.
35:48
You know, a lot of these CMT tools can talk to all of those right and so if you if you go back and we look at some of those
35:56
CMT tools. Let me grab one like scheduler. I think they've got the list of multi cloud support. How do you know what this one doesn't have the list that I was looking for.
36:11
Maybe one of these other ones global might have the list.
36:15
Just want to see the logos. Yeah. So here yeah they talked about being able to talk to OpenStack as your VMware AWS Google Cloud mechanics.
36:27
You know, so all kinds of both public and private cloud stuff as well as container stuff like Uber nannies and infrastructure as code like answerable puppet chef Tara form.
36:37
And and you want to be able to kind of manage all of that from one place. And so that's what these tools are typically do it.
36:46
So they're sort of sitting in between the business and it and the tools that we're using, and
36:55
Let's see.
36:57
Yeah, I think that's all I wanted to say it was just, just to show that. Yeah, they can certainly talk to a variety of, of, you know, both public and private
37:08
Right, so let's get rid of that one.
37:11
But he's got a good job. No, he says, it's going to be difficult to convince the management as well to several tools and approaches. I think that's a good point.
37:22
The organizations that I've seen this happen in it's usually from the top down.
37:28
Right, so, so typically your, your individual like it practitioner, they learn one cloud, they're happy with it. Then they don't see a reason to go beyond it. Right.
37:38
Usually comes from the top down the CIO talks to some other CIO, who says, oh, yeah, we knew we had to go multi cloud.
37:45
I mean, sometimes the multi cloud thing happens organically just in big organizations, you've got, you know, one hand doesn't know what the other hand is doing right
37:53
And so I've definitely talk to organizations where, you know, one part of them is an AWS shop. The other parts and as your shop. Right. And eventually you know it trickles up where somebody says, Hey, we
38:08
We need to figure out our cloud strategy, you know, we are spending money on different clouds. Does that make sense.
38:13
I like the idea that we have expertise in both, but let's let's really try and formalize that and make up multi cloud strategy right or somebody
38:23
Somebody's not happy. They have some huge outage with Azure and their companies, kind of in trouble for a while and they say,
38:30
We can't have this again, you know, if you look across the three major vendors, you know, as your seems to have the most outages.
38:36
And AWS, probably the least right but, you know, sometimes it's an event that causes us to think, okay, I mean, very often, early on in the cloud. People were running just ignoring region.
38:48
And then every once in a while, we would have an outage of a region. And that's when a bunch of companies would say, hey, my company has to work even if you know
38:57
Part of AWS is down, so I need to use multiple regions and then eventually they get to a point where they say maybe it's bigger than that. Maybe it's multiple cloud vendors, but it's generally coming from the top down.
39:09
And so if you're trying to make a case to your boss, I don't know, it is it is a hard pill to swallow. Because we said, you know, if you're going to have the tools in at expense.
39:20
Just having to learn multiple clouds. That's an expense trying to secure them right so there's there's definitely some some things in there.
39:32
A catch asks, How does open shift provide pass in a way that gets the subscriber of open shift the extra bits and pieces of the infrastructure services from each of the public cloud service providers. I mean, it's not so much.
39:45
That they're tapping into the things that the cloud providers have it's that it provides
39:51
Some of the same kinds of things that are common amongst all of them, but they're doing it in a way that is, you know, I only have to learn open shift. I don't really have to learn anything about you know somebody in my company has to be able to figure out how to how to, you know,
40:11
Provision the basics of AWS or Azure or Google Cloud and yet open shift setup in it. But beyond that,
40:19
Most of my practitioners just have to learn open shift the way of doing things, which is really Cooper daddy's and then we get some extra pieces. And so that's sort of
40:29
That's the idea. It's not that it's hooking in and taking advantage of the extra things of the cloud. It's providing some of the similar kinds of ideas.
40:39
We have another question.
40:42
As far as I know, the management or focus on company's core business. The rest should be based on simplicity and application balance between costs and security when it comes to information, I'm sure. I mean, I think if your management doesn't think about
40:57
Their reliance on one supplier, they're doing you a disservice. Right. So I think, I think it depends on the management I've definitely seen multi cloud in many, many organizations and many more considering right so and the places where I've seen it, it has been sort of a top down.
41:22
Okay, Michael Johnson says, sometimes the business requirements dictate MC example where data sovereignty come into play and or latency and availability.
41:32
MC. Not sure what MC stands for
41:40
I'm assuming it's maybe a particular cloud. So, so maybe Microsoft Azure or something.
41:46
Yeah, I think that that's an interesting point. Right. A lot of times, oh multi cloud having sense
41:56
Yeah, so I think that, you know, one of the first things you have to look at is, you know, where he does these cloud lives right
42:06
And if you're you're mostly in AWS shop and you say okay now we're doing business in this new country and they have data sovereignty laws that say their data cannot be stored outside of that country.
42:16
Well then often you'll have to go to Azure because they have
42:20
Regions in more countries than than any of the other cloud providers. Now it's easier for them to launch a region somewhere because their concept of region is much smaller than, than the others.
42:31
They don't necessarily have multiple Availability Zones in a region. They have certain local regions.
42:36
Might be just one data center right but it's secured and everything else so so yeah you're right about that. Sometimes I might say, hey, I need to use cloud. But in that particular country.
42:47
I'm going to have to use a different cloud provider right so that that could be a good point. Thank you Michael for that.
42:55
I'm Todd says with era open shift. Same with Tara form have to learn how to write infrastructures code in the tool and then you can leverage cloud and on prem. Yeah. It is definitely
43:09
You know, another hurdle, right, because you have to learn, you know, the functionality of your cloud. And then you have to learn the functionality of
43:19
Like a tool like Tara form on top of that. So that's sort of the, you know, that the the crux of the problem is you know what we're doing with these tools to fix our drawbacks. They're also introducing you know more cost more time, more complexity themselves, right.
43:38
And I would say that, you know, these cloud management platforms they try and be everything for everybody, but they're also introducing a layer.
43:47
Which is not cheap. Right. But also, you know, now you're saying, well, okay, I've got these workloads and I want you to go and run them somewhere. So there's a middleman in there.
43:58
And that middleman. We have to say, well hey now they've got access to all our data. Well, now we got to think about how secure is that, you know, where we got a valuable that as well. So
44:12
You know, there is no silver bullet here right multi cloud is complex and expensive.
44:20
And there are certainly drawbacks. But there are some benefits as well. And what we're seeing is that
44:26
You know, through the use of the right kinds of things and and I'd say, you know, most of them had some low level, it's something they start Cooper daddy's that says, at least my application can be easily moved around.
44:40
You know that's that's our first step to really making something multi cloud is to build an application that can easily live in any of those places.
44:49
So training is definitely going to be a big part of it. Right. And that's, that's where we come in. We have training on all three of the major cloud providers and some of the others as well.
45:00
I mentioned that we do red hat. So we have Hansel and open shift training and we got all these links, I think, Michelle will probably drop those into the into the chat.
45:13
And
45:15
As well. We have a promo going on until May 1 called spring 20 it will give you 20% off instructor led training. We got a link for that. I think she'll put that in the in there as well.
45:30
We run these webinars fairly often we, I guess, the ones that would be pertinent to this are coming up probably early May. We've got VMware one because VMware VMware sevens come out so you know that's a big deal. So we've got a webinar on that.
45:48
I think may 5 oh yeah they're actually put it in V sphere seven
45:54
And then probably after that, we have one on Cooper 90s. Right. So if Cooper daddy's something you're interested in. We have training on Cooper daddy's
46:02
It's a little different, you know, like exit certified typically partners with the vendor of record, and we have their course where right and you have to register sort of a certifier instructors through their programs.
46:15
But when it comes to something open source like Cooper 90s, there isn't just one vendor of record.
46:19
You know, anybody can hang their sign and say, hey, we do Cooper Nettie streaming. So what we do in that case is we kind of
46:24
Look for the most official one, the one that we think is the best course where and for us. That was the Linux Foundation, because they had partnered with the
46:32
The cloud native computing foundation was sort of in charge of Cooper Nettie and they build some
46:38
To four day classes, one for sort of app developers. One more for Cooper 90s administrators and they also prep you for the CNC F certification exams. If that's you know have interest. So, so we have all that kind of training as well. We're also a Docker.
46:54
trading partner so Dr glasses. Oh, we got a couple more questions. Do we have a strong business case for is in into cloud providers. Yeah. Yeah. Like earlier on. Yes, he joined me
47:09
Yeah, I think probably the biggest ones are, you know, sort of the four big draw of the benefits, you know, just to recap are
47:19
You know, sort of that idea that I don't want vendor lock in. You know, because then I'm at the mercy of them for pricing or end up lining, you know, some sort of service that I was really important to me, but also that idea that
47:35
You can run workloads where their cheapest for you. You have a little bit of risk mitigation against the outage of one provider versus another.
47:46
And then really just that idea of taking advantage of anything beyond the infrastructure as a service, you know,
47:54
It got says, you know,
47:57
Oh yeah around Ted's question sink sink kind of idea.
48:01
You know, something like Tara form, it's really letting you deploy multiple clouds. It's not really managing
48:07
And so that's why I had those sort of four categories. The first one was just tools to help you, you know, deploy to multiple clouds are building applications that can live in multiple clouds, but they're not true.
48:18
You know cloud management platforms, but that's a whole other level. But if you pair that with, you know, some good policies internally and a cloud.
48:28
monitoring tool that monitors across all of them. And you've got good cost allocations. I mean, you can kind of build up the same same set of tools that cloud management platform has
48:41
I think that's probably everything. I'll just take more questions. We got about eight minutes left, so if there's any other questions thrown in the Q AMP. A. Did I miss anything from the chat. Some people were throwing things in the chat here.
48:58
Oh, people were asking about the recording. Yeah. Later this week, you should get an email with the recording.
49:04
So you'll have access to that. And like I said, we've got, we got a few more webinars coming up. They seem to be popular right now because, you know, I think people are working from home. It's a little easier to jump on a webinar.
49:22
Oh, something else in the Q AMP. A Oh, thank you. You're welcome.
49:27
Alright, so we'll, we'll leave it there. If you have any other questions will stick around for a few minutes.
user avatar
Michelle Coppens
49:33
Thank you so much. Miles and thank you everyone for your time today. As we said we're going to be sending out a recording of the webinar to everyone that registered
49:42
Check out that spring 20 promo registered a class before May 1 and you'll be eligible for 20% off and it looks like I have a couple more questions now.
user avatar
Myles Brown
49:54
Oh, yeah.
49:55
Oh, just, just to know any of those links in the chat. As soon as you leave the tax gone. It doesn't save so you might want to go and copy and paste some of those
50:04
Dad says, Would it make sense to leverage multi cloud in a distributed fashion have one application that one part of it is better for one vendor and another parts better with another a web front end with the database back end somewhere else. I think
50:21
I mean, I've seen a situation where people actually distributed even just the web apps across multiple vendors, but I think that's a real rarity. Right. I think what you're doing is you're introducing
50:36
I guess.
50:38
Now if any one of those providers is down your whole applications down, you know. So I think it introduces some other problems, but I think it's a natural consequence if people really embrace that concept of microservices.
50:53
And, and, you know, the way I've seen microservices done in some places, you know, you've probably heard of the term to pizza teams.
51:01
Where you get some developers testers some operations people together and there instead of they're in charge of this set of microservices from end and you know from from roadmap to gather requirements building it testing at deployment to production managing it in production. Right.
51:18
And typically, when we embrace microservices, you know that microservices is sort of a small thing for this business, and then anything kind of major in businesses is, you know,
51:30
Is really utilizing a bunch of the micro services put together in to make some sort of application. And if those micro services are in different places. That's fine. The problem comes when we're moving data between them, you know, so
51:45
You know, most of the cloud vendors actually all the big ones have the exact same model when it comes to data transfer.
51:51
Right. It doesn't cost me money to put data into AWS. So if I upload a bunch of files from my own data center in AWS.
51:58
I don't pay any data transfer costs, I pay for storage, you know, once it gets there.
52:03
And I don't pay anything for data transfer. If I'm accessing that data from another like a virtual machine inside that same region.
52:13
But if I'm passing data from one region to another. I pay a little bit of money when I take data out of that cloud vendor. That's when I pay a lot
52:21
And so if you're moving data from one cloud to another. That's where it gets costly and and so if you're just doing an one time, you know, like I mentioned that idea where people have
52:32
Primarily AWS and all the data is in there and then they pull a subset of that data, move it over to Google Cloud and do analytics there.
52:40
Right. Ideally what they're not doing is grabbing data again and again and pulling it over. They pull it over ones that work on it. And then, you know,
52:48
New data will also get duplicated. But then you're only paying that data transfer fee once if I've got little applications running in every cloud and we're moving data back and forth all the time.
52:59
I think you're going to see some real costs come up with that.
53:02
Right, so it doesn't mean that it doesn't happen. And I think naturally as a result microservices and trying to compose microservices from different
53:11
Parts of your organization that might be on different clouds. I think you're going to see some places have that. But I don't see the big benefit of it. I see a lot of drawbacks. Yeah.
53:22
So that's a, I guess I've answered that as much as I can.
53:28
Oh, I don't think I can send the slides along, but they'll be in the recording so you can kind of see them there. Yeah.
53:41
Thanks.
53:44
Yeah, just imagine the security issues if something happened. Exactly. All right. Well, I'm like I said I'm sticking around a little bit longer.
53:52
If you need to grab any of those links from the chat. You're going to want to grab them before. Before we shut down the call. Also, I have my email address here. Let me throw that in the chat to everybody, it's Miles Doc Brown at Tech data.com
54:10
I think that, you know, if you have sort of technical questions we can get into discussions otherwise.
54:16
You know, if it's more of a sales kind of question you're looking for training, I'll probably 1400 one of our sales reps and you know we do, we do a lot of virtual training, we've been doing virtual training since 2012
54:29
And zoom is sort of part of the, the overall what we call I MVP.
54:36
And, you know, we've been doing them for a long time. We have people like Michelle who kind of make sure at the beginning of class. Everybody said they have their materials their ad is working.
54:47
We really encourage people to turn on their, their cameras, unlike you know a webinar where I'm
54:51
Talking 200 people, but generally in a class, you've got, you know, maybe, maybe 1215 people everyone gets their camera on, even though we can be in a physical training center we've closed all our trainings and currently
55:04
You know we can do this. And so we've been doing this for many years. We do a lot of hybrid where we have when our classrooms are open, we might have five people in class in a particular city.
55:13
And then another 10 people coming in remotely. So all of our instructors are very used to doing this kind of virtual training.
55:19
But you know when things get back to normal will have classroom training and then a big part of it is you know companies want to bring somebody in. And so we'll send somebody to us. That's easier. So we have all those different training modalities.
55:34
And like I said, you know, we had a lot of different vendors that we have authorized training for and we are the training partner of the year for many of them in AWS Red Hat VMware IBM. So that's, that's some of our big ones right there.
55:50
Alright, I think, the questions are mostly shutting down. So I'm going to stop sharing
55:57
And thanks so much miles. I'm going to stop the