Enterprise Architect here: in my opinion, hybrid cloud is the best balance. Protecting your intellectual property on prem while leveraging the scale and fast time to production that public cloud offers. The problem with all in public cloud is data protection, especially in regulated spaces. The problem on-prem it’s typically the business views IT as a cost center and constantly pulls funds, additionally, IT departments have remained silo, which does not allow them to be flexible and agile to meet the needs of the developers and the business. Ultimately, we are facing a people process problem .

Thank you for your wonderful effort

Hi I'm Alex 😂 Does the IDP offer only services and tools for Ci/CD or also custom developed APIs that can be shared across multiple dev-teams for their apps?

Cloud isn’t expensive, incompetent engineers and clueless tech architects n managers are. Now that they’ve rode the wave on mouthing cloud buzzwords they’re hitching a ride on the new wave.

Thank you for this video.

If you examine the OS license on AWS, you'll find that for Microsoft Windows, a dedicated host or instance is required, or you can purchase a Microsoft License Mobility through Software Assurance. Microsoft Monopoly :)

Well Done!

Awsome video 🙏

Great video! Would love a similar look at garden and skaffold and maybe some comparisons between the three? The fact that Docker seems to have acquired Tilt would imply that Docker's going to offer that Tilt functionality OOTB at some point not too far down the road.

On premises is much much better, Cloud is costly, runs on code based infrastructure hence has several instabilities, new security vulnerabilities everyday

Grt video🎉

Hello , how the other pods can be scheduled on tainted node which you are saying in video at 2nd min onewards , If the node is already having taint with noschedule I believe no pod which do not have any toleration can not be scheduled on this node. Please clear to me . Or I am misunderstanding this in some way,

Not able to access the labs somehow. Can you check please, not able to enroll to this program?

Based on my experience cloud comes with following issues - 1. performance. 2.Also none of the cx gets rca of any issue, cloud tac just gives some reports and push the cx to reach other tac. 3.this one specific to azure, your networking sucks.

Cloud was never a good idea for most companies and was well over used by companies that could have done the same thing locally. They were sold the snake oil and they swallowed it.

I think, cloud is good for personal projects (small scale but scalable projects) but not for scaled projects. My observation is that, it's actually opposite of what they advertise. Get your own infrastructure as soon as you can.

Cost is factor but not only factor ...if you consider other factors..the answer is clear...

great video, you just got a new subscriber :)

company always make right

Wise decision

All that to get the very secrets imported from secret providers by ESO, stored as base64 smh... safety is a myth lol!!

Hybrid is the future

CLoud is the ans but architecting is very imp and GCP Anthos (Google Distributed Cloud) can solve many of such problems if you want to run your loads on on-prem containers with control at your GCP project end!.

Expecting the cloud to be cheaper is a prime example of “expectations not met”

The promise of AI is the culmination of legal theft of intellectual property. Just like all things people want to spread someone put a blurb into the ears of people and it spread and that is what happened when it was decided nobody wanted to be on board with this blatant robbery of IP. I mean you could see it coming a mile away. 1 take away the ability to support your customer by outsourcing. 2 inject false sense tech advancement with web 2.0 which spurred injection of users not normally on traditional data centers. 3 reduce access to current data centers creating demand by consolidating global data center access. 4 gate keep access through ala carte subscriptions with no benefit over traditional data center value proposition. Do any of you remember the invoice of your data center or virtual data center before cloud? The offerings don't even exist except in very niche b2b situations where the company has a longstanding partnership(and lots of money) with the data center provider. It took 10 years but they found a way to milk even more money of the business world on top of the blantant robbery that was already taking place in the sale of data center access in the first place with hybrid cloud. Only difference is now you lost 2 decades of workers who have the skills to reduce cost which lead to being at the mercy of the very providers who sold you on cost reduction peomise of cloud computing so long ago. They now have the tools to always peek into your data with hybrid and AI and they are making you pay for it.

if cloud really expensive then on prem DC, what if we think this way. In DC according to my organization we have separate team for each domain for example storage , unix , windows , NOC and some others. We have to think about the cost of paying the employess , DC building , Electricity consumption and the list go on. If compared to that we can say that cloud in all in one solution and the price we are paying for that is relatively less. What do you guys thinking about this. Lets discuss !

Cloud only makes sense for startups or when the app has a small customer base

ideally the responsibilities should be spread across a team, not thrown on the of back of this mythical renaissance man (who should also be an expert in the business domain, a mobile developer and have some UX chops and possibly be able to throw together some LLM ;) ) Also, there's a reason companies move to SRE or platform engineering models

Cloud is not for everyone. Its for business who provide external customers services. If the majority of their product is for in house operations need for example pharmaceutical companies then cloud adoption is not right choice as they need to pay high wage for cloud skill. Business has to analyse whwther cloud really value money wise . They should not rush since everyone saying cloud cloud.

Please work on Cloud task in kodekloud engg fast

I completed the course and it was a good learning experience. Thanks Mumshaad.

wtf this is so good

100 VMs = 100$ 90 VMs = 110$ 80 VMs = 120$ Less the number of servers, higher is the cost

Great

As a data eng. I can say it is easy to spend a lot on cloud computing. And it is important to have the right skills otherwise you will spend too much. I saved potentially a lot for my company and it is because I have experience about optimizing things not adding additional resources.

For an Engineer's point ot view..i guess Hardening and staying close to Linux will be the key to survival in any Arena of Computing..be In on-prem, Hybrid or Cloud.

The problem here is you have not onboarded the right architect in your company. He should have well organised design before you onbaord anything before security approval.. if no designs at right place will come up having such titles - people leaving cloud !! Very simple

Hi!. I believe that the most used today and tomorrow is well is hybrid cloud, in this scenario the critical data is into the on premise still

I think Not having FinOps in organization is causing this cost

or it can be complexity. it's true that aws, azure and GCP have the most amount of feature and services but most application needs just compute, db and some networking and in my use case manage k8s, which i can get that simply by using something like digitalocean or linode or even hetzner

Figma use react not wasm 😂

can you make a video on how and best practices for configuration management?

I did watch 2-3 videos from other channels before reaching this video and finally after watching this particular video i understoood 12- factor app so well and in depth with examples. Thank you so much @KodeKloud for making it easy and fun to learn !!

Please add azure devops course with habdson labs

Fully dependent on the the know how if your IT team if you have one. Some can run very effectively on prem. And like you said it's based on work load I think hybrid is the best solution here.

(sorry, long story ahead) Coming from someone who's in the middle of this conversation, I also see (mostly early adopter) companies move back on prem. Mostly because of cost, also because of security rules (banks, manufactoring, etc), and because the savings in terms of staff were absent. The Cost factor actually has multiple layers. First, sizing by hyperscalers tends to be aggressive. A lot of workloads that moved to the cloud, were sized too small and ran better with at least x2 the resources. This means that the initial budget already was too small. Second, a lot of workloads that were moved to a cloud solution actually were not compatible with cloud technologies. If you have an app that can only run inside a VM, you will have to facilitate an environment IN THE CLOUD where this can run. AWS has VMC, Azure has AVS, etcetera, so you can. But there is a cost attached and that usually is not as economical as running it on prem. And third is the added complexity and dependency. If you have an office where people work and your office internet is down, people can't work. That costs money. To prevent that, you could invest in a redundant internet line in case one fails. But that also costs money. And extra hardware like routers and stuff. If you even decided to move your VDI desktops to the cloud (some companies do) then this dependency multiplies. Fourth, generally hyperscalers guarantee the availability of their service, not your stuff on their service. If Azure fails but returns to service after an hour, they have met their SLA. It's not said that YOUR DATA also returned after that hour of downtime. This means that you should set up some form of backup / disaster recovery in case there is a loss. You can have that replicate back to on prem, but that would be counterproductive, right? You wanted to move stuff in the cloud, not back on prem. So you look for a solution in another cloud to backup to. Now you've added another layer of complexity (cloud to cloud connectivity) AND another cost factor: a DR cloud solution. Security is another reason. Many banks and manufacturing companies have quite a lot of intellectual property and personal identifiable information (IP and PII) in their data. At least inside the EU but I assume also elsewhere, there are requirements by national banks and governments that PII is secured in a specific way and companies themselves would rather have IP in their own safe inside their own building. Now, imagine that you have a database with customer information, so PII data, that you have on prem and you have an app in 'the cloud' that needs access. It will be complicated to secure, it will be not as fast as it would be when those systems would be next to each other on the same infrastructure so the result is that the app moves to on premises because cloud was not an option because of the PII data. Now, there are a million examples of this in various scenarios where the outcome is: cloud is not suitable. I'm not going to go into the staff issue but as far as I know, migrating anything to the cloud has not resulted in mass layoffs in IT departments anywhere. It mostly eliminates some problems and replaces them with other challenges that need to be managed by that same staff, most of the time also requiring that staff to be additionally trained and certified before they are fit for their new job, causing additional costs. Conclusion: there is a place for cloud, there definitely is. Especially for services that are used by the masses that are easy to host, like websites, email, telephony, live streaming, etcetera. However, once you get to the primary process of a company, as in; their reason to exist, most of the time the IT that goes along with that, is not cloud compatible for any or all of the above mentioned reasons.

You piece on Pros & Cons @06:06 is accurate. What you say about cost vs value, is a key. If a business that migrates to the cloud, is unable to extract the potential value from the move, it eventually results in the conclusion that it is only costlier. Not all businesses succeed in extracting value from an innovation for a variety of reasons.

I'm just halfway through the video and can vouch that this guy explains precisely in the way how it should be explained, crisp and clear. Great content @KodeKloud 👏👏👏