How can you save money and improve performance with cloud computing?

It's the wrong question. Moving to cloud is best thought of as increasing value, not decreasing cost. It's never cheaper to lease time on someone else's computer- it's much smarter to think of it as a way to reduce your overall costs (simpler development, fewer system administrators), and increase value (easier to reach scale, and advanced functionality / reliability is easier to use). Going into cloud with the intention of saving money by lifting and shifting your datacenter doesn't work in real life. Going into cloud to transform how you design, build, and deliver digital services does- but you don't save money, you create additional value.

Moving to the cloud, as the comment from Tim Howland mentioned, does not decrease cost. In fact, in most cases, it will be more expensive, particularly without solid planning, and a clear understanding of the purpose, expectations, and what it entails. A good strategy, beginning with a solid exit plan as well, beforehand, can help eliminate vendor lock-in. The cloud does not eliminate the tech debt either as applications that are refactored and not rearchitected merely transfer the debt to the new cloud environment. The skillsets required to move to a CSP differ as well, so the elimination of the legacy headcount can decrease, but is replaced with the same or more headcount, with contrasting skills at much higher compensatory levels.

Some of the factors to think about while choosing a Cloud provider and it's service: 1. The ease of use and the support offered by the cloud provider. 2. The type of services that fits the solution you are building. 3. Understand which offering suits your budget. 4. The security and compliance requirements that your Organisation needs.

Rightsize Resources: One of the primary benefits of cloud computing is scalability. Optimize Storage: Implement data lifecycle policies to automatically move data to lower-cost storage tiers when it's not frequently accessed. Serverless Computing: Consider serverless computing platforms like AWS Lambda, Azure Functions, or Google Cloud Functions. Containerization: Use container orchestration platforms like Kubernetes to efficiently manage and scale your applications. Reserved Instances (AWS) or Committed Use (Google Cloud): Use Auto-scaling: Monitor and Analyze: Multi-Cloud Strategy: Implement Content Delivery Networks (CDNs): Use Cost Management Tools: Reserved Bandwidth:

I agree, but it’s important to realize that the cloud is rarely a cost-saving solution, and while moving *to* the cloud is fairly straightforward, migrating *from* the cloud is fraught with complexity and complication. I’ve heard of several organizations that are staring at huge bills every month and have destroyed their capital position because everything is an expense with no depreciation.

Some of the best practices to optimise cloud cost are: 1. Set Budgets and Alerts 2. Utilise cost recommendations that each cloud provider offers and act on them 3. Right size the services, remove unutilised, Idle resources 4. Leverage Spot Instances, Reserved Instances, Savings plans 5. Snooze the resources in lower environment during non business hours Moreover, the cloud native solution that you are building shall be cost optimised by design.

First, depending on your traffic, data size etc, many find it surprising how costly some cloud computing based solutions are. Some of them have a very big price tag and it caughts you by a surprise, even when you have set price alerts. Refactoring out later from already vested in AWS solution is not ideal. Hence, it is often paramount that your AWS Architect truly takes time to understand how pricing really works, many skip this phase. Secondly, do not go serverless just because you can. Do it only if you really can justify why you need it. E.g. many use lambda's for a stable non bursting tasks where a traditional service running in a container would make much more sense and be cheaper.

Scalability at no extra or upfront cost - you pay for what you use mostly. Possibility to destroy any prepared infrastructure before using it and replace it with ease and at minimal (if any) cost in case of last minute architecture changes or new circumstances. Managed services that make a cloud provider take care of safety, patching, monitoring and other duties you would use manpower for in an on-premise environment. Cheaper, safe online storage for short- and long-term periods. And on top of all that you get to decide which new services you want to take advantage of while keeping a possibility to have full control over legacy solutions just like before migration or even utilizing a hybrid architecture during transition or permanently.

Optimize resource uses Keep the cpu n memory footprint as low as possible. Use event based apps whenever continuous app/api availability not required. We have saved 80% cloud cost by using graalvm.

one of my favourite benefit of cloud computing is scalability. Gone are those days when we need to wait a lot for the hardware availability, provisioning and then reprovisioning (due to pack of performance testing). Peak seasons, performance requirements are no more the real pain points and mainly this is one of the reasons, organizations are moving to cloud IaaS and PaaS

The recently available sophisticated solutions are keeping the data in transit and data at rest secured. I would never imagine that a workload on the cloud is any less secured than a workload on-Premises. The security posture and technologies need to be well thought through, the intrusion prevention, Identity management, Siam solutions are available on-cloud to make sure end to end security of client data. 'Zero trust' is the word complimented with the cloud security solutions.

Additionally, 1. Automate - leverage Infrastructure as code automation for provisioning your cloud resources with right tags on right time. 2. Use tags - A good tagging policy helps you to get the granular view of your cloud spend. 3. Zombie hunt - regularly check for unused/idle/unclaimed resources and delete them

Budgeting and forecasting are two important aspects of any IT service. This is also applicable to cloud computing. Organizations need to assess the Business case and then have a strategy to maintain the budgets. This can be done easily with some tools which continuously look at the consumption, suggest over provisioning (by performance analysis), trigger alerts for forecasting that could hit the budget and help in creating automation scripts to balance out the capacity vs budget alignment.

Train your cloud team: Preparing your team for cloud needs systematic approach considering short term and long term goals for preparing the team for cloud. 1.Identify training mode: Cloud learning would be combination of different methods including self-learning, hands on labs, classroom training and on the job training. Choose combination of the approach so that you can provide choice to your team. 2. Leverage available training sources: There are multiple training sources available from partners, internet, online training providers, certification, classroom, labs, guidance driven approach from your own team. Again, go for wider sources to create choices.

As-is vs to-be state: Build clarity on how you will do things differently than the way things are done currently. This exercise will help understand if cloud computing is making any difference to way things done currently. Is it bringing mere resource driven cost reduction or process driven cost reductions or result driven cost reduction. This clarity will help understand ROI on the cloud computing. E.g. In Product development, does DevOps on cloud helping you reduce, engineering cycle time, quality, scale, availability etc. or it is just mere moving current development labs to cloud.