Track 3 Session 3_如何妥善運用雲端優勢遷移上雲.pptx
•Télécharger en tant que PPTX, PDF•
1 j'aime•317 vues
Recommandé
Recommandé
Contenu connexe
Tendances
Tendances (20)
Similaire à Track 3 Session 3_如何妥善運用雲端優勢遷移上雲.pptx
Similaire à Track 3 Session 3_如何妥善運用雲端優勢遷移上雲.pptx (20)
Plus de Amazon Web Services
Plus de Amazon Web Services (20)
Track 3 Session 3_如何妥善運用雲端優勢遷移上雲.pptx
- 1. © 2020, Amazon Web Services, Inc. or its affiliates. All rights reserved. 如何妥善運用雲端優勢遷移上雲 T r a c k 3 | S e s s i o n 3 Vautier, Nicolas Leader, Taiwan Solutions Architecture, AWS
- 2. Two main approaches for cloud value analysis Top-down KPI benchmarking data Bottom-up Quick analysis Value estimate Detailed data from your organization Greater investment of time Higher level of accuracy
- 3. Quantifying business value Cloud Value Framework Cost savings (TCO) Example 50%+ reduction in TCO (GE) What is it? Infrastructure cost savings or avoidance from moving to the cloud Cost impact Staff productivity Example More than 500 hours per year of server configuration time saved (Sage) What is it? Efficiency improvement by function on a task-by-task basis Operational resilience Example Critical workloads run in multiple Availability Zones and Regions for robust DR (Expedia) What is it? Benefit of improving SLAs and reducing unplanned outage Business agility Example Launch of new products 75% faster (Unilever) What is it? Deploying new features or applications faster and reducing errors Value impact
- 4. Cloud Value Framework Cost savings Estimated 18% savings, realized 40% savings 1 year after migration, and realized 58% savings after 18 months Cost savings (TCO) What is it? Infrastructure cost savings or avoidance from moving to the cloud Staff productivity What is it? Efficiency improvement by function on a task-by-task basis Operational resilience What is it? Benefit of improving SLAs and reducing unplanned outage Business agility What is it? Deploying new features or applications faster and reducing errors
- 5. Cost savings Economics of the cloud Infrastructure cost ($) Time Large fixed spend Lost opportunity Traditional hardware Actual demand AWS Predicted demand Key Opportunity cost
- 6. How do customers lower their costs with AWS? 84% of on-premises workloads are over-provisioned Server sizing based on compute needs 78 price reductions* Economies of scale allow AWS to continually lower costs On-Demand Reserved Spot Pricing model choice to support variable and stable workloads *As of December 4, 2019
- 7. Cost savings Modeling on premises Facilities cost Space Power Cooling Facilities cost Space Power Cooling Storage software costs (+ maintenance) Hardware: storage disks, SAN/FC switches2 Storage costs Recurring ISP and bandwidth costs Network hardware: LAN switches, load balancer3 Network costs Facilities cost Space Power Cooling 1 Server costs Software: OS, virtualization licenses (+ maintenance) Hardware: server, rack chassis PDUs, ToR switches (+ maintenance) Diagram doesn’t include every cost item. For example, software costs can include database, management, and middle-tier software costs. Facilities costs can include costs associated with upgrades, maintenance, building security, taxes, and other items. Illustrative
- 8. Cost savings: Server cost is just the beginning For an individual server, the 5-year total cost of ownership (TCO) is 3.5x the cost of the hardware purchased 36.4 7.3 4.9 22.9 28.5 Percentage of total cost by category Facilities OS licenses Rack costs Hardware overhead Hardware
- 9. Cost savings Example 1 non-virtualized physical server, 16 cores, and 64 GB RAM at 50% utilization 1 c5.4xl (3-year TCO) Server hardware Maintenance Rack Facilities Labor Software $8K $3.6K $2K $10.8K $4K $0 On premises Total (3-year) $28.4K Amazon EC2 3-year RI Labor Support $6.5K $1K $1K AWS Total (3-year) $8.5K
- 10. Cost savings: AWS benchmarking insights Source: AWS Cloud Economics Benchmarking Reduction in overall spend per user 27.4% Reduction in overall spend per user 42.4% Lower overall spend per user vs. multi- cloud customer 12.3% AWS reduces costs Cost reduction grows as customers mature and scale on AWS Multi-cloud AWS exclusiveOn-Prem On AWS All customers Multi-cloud vs. AWS exclusive customers On-Prem On AWS Applications with 1,000+ users
- 11. Cloud Value Framework Cost savings (TCO) What is it? Infrastructure cost savings or avoidance from moving to the cloud Staff productivity What is it? Efficiency improvement by function on a task-by-task basis Operational resilience What is it? Benefit of improving SLAs and reducing unplanned outage Business agility What is it? Deploying new features or applications faster and reducing errors “Building a Hadoop cluster in-house would have taken us several months, but we were using this solution much faster because of the AWS Cloud. That was key for us, and it validated our decision to move to the cloud.” —Pascal Bergeron, Director of Algorithmic Trading
- 12. Staff productivity Focus on value-added work Tactical (undifferentiated) Strategic (differentiated) Current state Lift and shift to AWS High-level AWS services Activities Customer maturity on AWS
- 13. Task Typical reduction Description Server budgeting and planning 90% There is no capital server budget or plan in the AWS Cloud Server purchasing process 75% Instance purchasing requires minimal effort in comparison to server purchasing Long-term capacity planning 75% Capacity planning is simply a matter of initiating new instances based on thresholds, and much of this can be automated Project budgeting and planning 75% Project budgeting and planning effort should be significantly reduced Prepare detailed implementation plans 75% Implementation plans will be reduced since instance initiation is straightforward Arrange repair for hardware on occasion of hardware failure 100% Not necessary with AWS Installing, upgrading & removing software 50% Simplify and automate OS patching and updating Staff productivity Example Server administrator Illustrative
- 14. Task Typical reduction Server budgeting and planning 90% Server purchasing process 75% Long-term capacity planning 75% Project budgeting and planning 75% Prepare detailed implementation plans 75% Arrange repair for hardware on occasion of hardware failure 100% Installing, upgrading & removing software 50% Staff productivity Example x 75% reduction with AWS 18.8% efficiency gain 25% of time Server administrator
- 15. Staff productivity: AWS benchmarking insights Source: AWS Cloud Economics Benchmarking Cloud improves efficiency Increase in VMs managed per admin 57.9% Increase in TBs managed per admin 67.7% Increase in VMs managed per admin 147.7% Increase in TBs managed per admin 153.5% With larger gains for rearchitected applications On premises On AWSOn premises On AWSOn premises On AWSOn premises On AWS VMs managed per admin TBs managed per admin VMs managed per admin TBs managed per admin
- 16. Cloud Value Framework Operational resilience Cost savings (TCO) What is it? Infrastructure cost savings or avoidance from moving to the cloud Staff productivity What is it? Efficiency improvement by function on a task-by-task basis Operational resilience What is it? Benefit of improving SLAs and reducing unplanned outage Business agility What is it? Deploying new features or applications faster and reducing errors Condé Nast Russia now experiences uptimes of 99.9%, a 15% increase from the availability in the data center – and if the site does go down, it can be restored in a matter of minutes from a backup system
- 17. Operational resilience Downtime costs $1,034,640$474,000$55,560 High-impact applications Low-impact applications Source: Ponemon Institute
- 18. Operational resilience Categories of downtime cost
- 19. 1. Business user productivity # of users = 1,500 Average FTE cost = $125K/year Operational resilience Example - Media Company FTE cost = $1,500/minute FTE cost = $750/minute 2. Other categories ~50% of cost 0.1% improvement = 525.6 minutes x $2,250/minute ~$1.2M Calculate cost per minute of unplanned downtime Illustrative
- 20. Operational resilience: AWS benchmarking insights Source: AWS Cloud Economics Benchmarking Decrease in downtime 56.7% Decrease in critical (P1/P0) incidents 31.8% Decrease in total monthly incidents 24.1% On premises On AWS On premises On AWS On premises On AWS On premises On AWS Downtime in minutes # of total monthly incidents Mean time to resolution (MTTR) in minutes # of critical incidents Decrease in MTTR 39.3%
- 21. Cloud Value Framework Business agility Cost savings (TCO) What is it? Infrastructure cost savings or avoidance from moving to the cloud Staff productivity What is it? Efficiency improvement by function on a task-by-task basis Operational resilience What is it? Benefit of improving SLAs and reducing unplanned outage Business agility What is it? Deploying new features or applications faster and reducing errors “Using AWS helps Dow Jones to be more agile in developing revenue-generating products. Thanks to AWS, we now build more products and spend less time running a data center. Our overall product development velocity has increased by at least 30 percent.” —Stephen Orban, Chief Information Officer & Global Head of Technology
- 22. Keys to business agility See change as a positive; empower all employees to identify opportunities for improvement Culture of innovation Lowering the cost of failure results in more experiments and more opportunities for success Cost and speed of experimentation Growth of third-party innovations has been dramatic over the past 5–10 years; those able to incorporate these capabilities quickly have an advantage Take advantage of external opportunities
- 23. Business agility Innovate faster while reducing failure costs Simplify requirement design by leveraging built-in operational solutions (e.g., scaling, security) and hundreds of AWS and third-party services and solutions to deliver cloud-native products Increase the speed of build and test through continuous integration and delivery pipelines, and lower the cost of spinning up experimentation by shutting down environments or services quickly Reduce maintenance cost and simplify operations with AWS providing IT infrastructure services, and with AWS or third-party-managed service options Traditional software delivery lifecycle and average % of time spent at each step* Cloud benefits Eliminate deployment wait time with automated application development and smaller deployment batches *META Group/Gartner research 22% 15% 23% 17% 12% Requirements & analysis Design Build Integration & test Deploy 11%Maintain
- 24. Business agility Track and measure KPIs KPI New applications launched per year Mean time to resolution (MTTR) in hours Time to market for new applications Response time to defects (hours) Time to provision new environments (days) Customer retention (%) Deployment frequency (revs/year) Adoption of new features (%) Time to deploy to production (weeks) Value per release ($ revenue potential) Time to deploy to test (days) Employee retention (%) Features per release Employee absenteeism (%) Total # of incidents or defects Employee NPS or satisfaction Percentage of total defects found in test Customer NPS or satisfaction
- 25. -1 0 1 2 3 4 5 6 7 8 9 On-Prem AWS Calculated benefit Conversion rate by month Business agility Example of the value of time to market (TTM) 1. TTM 4. Conversion uplift 3. LTV of uplift 2. Clicks per month 6 months 1 month 10% $1K 10K Total clicks Clicks per month # of months Total clicks 10,000 x 5 50,000 Conversion uplift 10% x x $LTV $1,000 x x $5,000,000 Value= = Customer-facing application for a subscriber-based company
- 26. Business agility: AWS benchmarking insights Source: AWS Cloud Economics Benchmarking Decrease in TTM for new features and applications 37.7% Decrease in time to deploy to production 39.3% Increase in code deployment frequency 341.5% On premises On AWSOn premises On AWS TTM for new applications in days Time to deploy codes to production in days Code deployment frequency On-Prem On AWSOn premises
- 27. Live Nation: Realized value with AWS 18% initial TCO savings 40% reduction in TCO through cost optimization after first year 58% reduction in TCO to date Cost savings (TCO) 50% reduction in traditional IT tasks 10x improvement in # of new projects Improved automation and backup processes with managed services Staff productivity 99.9% vs. 99.999% availability Improved security posture Near-zero performance complaints Operational resilience 10x increase in innovation pipeline Rapid experimentation 90%+ business user satisfaction with cloud services Business agility Progress Internal migration Focus 17 months 118 applications in the cloud & 58% TCO savings
- 28. Best practices for communicating the cloud value Start the business case process early in the decision-making process Involve the right stakeholders (Finance, Procurement, IT, Engineering, Business), and build the case in multiple iterations with them Assign value to areas that are hard to quantify, like business agility Present the overall value, not just TCO Cloud value
- 29. Learn to build cloud fluency in your enterprise Resources created by the experts at AWS to help you build the skills you need Cost savings (TCO) Staff productivity Operational resilience Business agility Digital training Classroom training AWS Certification Enterprise resources Talent pipeline Be ready for global cloud transformation with a custom training strategy Visit our resources for enterprises page at https://aws.training/enterprise
- 30. Thank you! © 2020, Amazon Web Services, Inc. or its affiliates. All rights reserved.
Notes de l'éditeur
- Quantifying and realizing the value of AWS for your business
- In this session want to focus on how to “identify and quantify the business value with AWS” There are two main approaches Top-down approach Estimation of Value realized by other customers who have migrated to the cloud You can use these insights to build or supplement your business case or compare your organization to industry standard and best practices Bottom-up approach Today’s focus Data is the actual data from your company. Various business units, could be time consuming. However Get a specific business case at a higher accuracy level. You can pick and choose one approach, or a combination of both based on your needs.
- AWS Cloud Value Framework is a way for us to think about the value of the cloud. Cost savings Avoiding supplying the on-premises infrastructures with large fixed spend Reducing the ongoing variable spend through the AWS’ economies of scale Staff productivity Increased output by the same size team Many of their previous tactical work are no longer needed Operational resilience Enhanced availability Enhanced Security Less downtime Business agility, New products, feature or new geo location Initial focus is cost savings. That’s what I see my customer focus on at first However, when we go back to customers and ask them where they realize value: Customers who we spoke to believed that those things in the right were really the more compelling benefits of cloud. According to a 2018 IDC report: Average 51% reduction in the total cost of ownership 62% increase in IT staff’s efficiency 94% less unplanned down time, 3x more new application features delivered to the market
- The first pillar of our Cloud Value Framework is cost savings, which is realized by avoiding supplying the on-premises infrastructures with large fixed cost and reducing the variable cost through the AWS’ economics of scale. When working with our customer, Live Nation, the world’s leading live entertainment company, who produces live concerts, sells event tickets, and connects sponsorships. Live Nation has a very short timeline to migration 668 servers onto AWS. Business case, we estimated the team could realize 18% of cost savings. Right after their full migration, they realized close to 20% savings in TCO. As they got better at operating in the cloud 40% reduction within 12 months 58% reduction in 18 months.
- Now when you think about the basic economics of the cloud, let's look at infrastructure costs on the Y axis and time on the X axis. [CLICK] Green line: Predicted demand will go up into the right. [CLICK] Large fixed infrastructure spend is added in a step function basis, so capacity is added in a way that follows this dotted line, above that predicted demand [CLICK] Now actual demand is volatile, and it'll look something like this curvy blue line. [CLICK] Essentially everything between those two blue lines are opportunity cost due to overprovision. That is where cost savings comes. The National Resources Defense Council finds that average server utilizations ranges between 12 to 18%. [CLICK] This here is when your actual demand exceeds your capacity, well, that's slow servers, that's unresponsive applications, that's down time. Possibly that's unsatisfied customers. [CLICK] With a pay-as-you-go on-demand model, you can avoid opportunity cost and lost business opportunity
- As mentioned earlier, the avoidance of a server malfunction, the compromised customer experience, or lost revenue, made many organizations overprovision their on-prem infrastructure. To be exact, 84% of the on-prem workloads are over-provisioned. With elasticity being one of the key advantages of moving to the cloud, we identify this as the first area for cost saving, as you can now spin up and down resources as you need. Second, in AWS, we offer a variety of pricing models, where you can reserve resources in advance for steady workload and achieve (up to 70%) of discount with reserved instance; or bid for resources with Spot (with up to 90% discount) for workloads that are fault tolerant. The third main area is that because of the economics of scale and the continuous innovation, AWS constantly cuts price for services. 78 times of price reductions at this point. This is automatically applied to your bills.
- When we build a total cost of ownership analysis, it's important to have a total cost of ownership view. This means you need to look at the server, storage and networking costs, and you have to look to the right. So, for example, on server costs. It's not just the raw procurement cost of the server. It's all the hardware that comes with it. It's the server. It's the rack. It's the chassis. It's the PDU. It's the top of rack switch and maintenance on that hardware. It's also any software that you may have to install on those servers, operating systems, virtualization licenses and maintenance on that software and lastly facilities costs.[CLICK] You have to consider space rent, power, cooling. In our experience, customers often struggle with quantifying the right most column, which is the facilities cost. It's because those costs typically are shared across business units and products and lines of business. And so they could be difficult not only to quantify but also to allocate properly.
- To put things in perspective here is a summary based on our analysis on over 700k physical servers in both Linux and Windows operations environment over a time period of 5 years. It shows that the total cost of ownership is about 3.5 times the cost of the hardware. Please note that depending on the operations environment, the ratio varies. Linux: 2.7x the cost of the hardware Windows: 5.5x the cost of the hardware --- N = over 700k physical servers (400k Linux, 350k Windows) T = 5-year view Facilities is power, cooling, space Rack is physical rack and 1-time setup HW overhead is maintenance and provision for spares Multiplier is 2.7x for Linux and 5.5x for windows (Gartner uses a 4x multiplier). As time increases, the multiplier goes up.
- Here is an example of TCO analysis where we compare the on-prem cost of a non-virtualized physical server, 16 cores and 64 GB RAM at 50% utilization to its equivalent in the AWS environment. Update the AWS for C5 4XL --list price for the hardware is ~$13k from industry data. Or $8k with a 40% discount. --HW maintenance @ 15% per year. $1.2K (per year) – Gartner (IDEAS company, pricing tool) --Rack Infrastructure including switches & maint = $2K for 2U server. (?? Need explanation) Assume out of 42 U’s 28 are utilized. – BCT --Facilities = $10.8k: (TOTAL $300/month. $200 in power/cooling for a .75 kW server @ $0.18 per kWH + and $100 for space) --Labor: 1 admin manages 100 pServers @ $135k comes to $1,350 year or $4k over 3 years. --Software = assume Linux and no virtualization so $0 for SW. (windows list is ~$3K per license). (VMWare $3.5K per processor) AWS COSTS 3yr RI costs $6,500. If elastic workload like dev/test then running OD would cost $4.8k (??need to update the #). But let’s run with the RI example, which is a good model for production workloads. Business Support $650: 10% of AWS spend Labor cost 1/4 the physical server The 3 yr cost of a c5.4xl is $8.5K.
- Now I am going to quickly go through the insights that we learned from our benchmarking studies. Lets start with the first and left pillar of cost savings. Just a refresher that these are from over 1500 customers surveys who have migrated to AWS for more than a year and here is what we learned. Overall, all customers reported a 19% reduction in IT infrastructure spend on AWS vs on-premises. Again these are all customers that have many other factors causing total costs to go up or down, like some of them double the size of their business or significantly increases users. So when we normalized it by looking at spend per user, the cost per app user dropped at a higher rate than the total infrastructure spend ,to 27.4%. We then slice the data to see if these cost reductions grows as customers mature and scale on AWS. What we learned, as you see in the third chart here is that for apps with greater than 1000 users, spend per user dropped 42.4% after migrating to AWS. These first 3 charts have the same horizontal or x-axis in that they depict relative spend on-prem and on-AWS. Now, the last chart here only shows the on-AWS portion or after migration portion, but these are cut into 2 groups: those that are multi-cloud and those that are AWS-exclusive. These show that there’s a 12.3% lower cost per user for AWS exclusive customers. Again these are the benchmark average of over 1500 customers while individual company’s changes vary.
- Customer example Next we'd like to move on to cover staff productivity. As a reminder, staff productivity essentially is the benefit of eliminating or reducing time spent on tasks that are no longer needed with the cloud. Essentially, what happens is that tactical work drops and that tactical work is able to be replaced with more strategic work. And typically, what we see with customers is that that differential increases less time is spent on tactical tasks and more time to spend on strategic tasks as higher level AWS Service's and manage service's are adopted.
- Staff productivity is all about making your staff more productive by reducing or eliminating time spent on tasks no longer needed in the cloud We have the employee activity on the Y axis and their organization’s maturity on AWS on the X axis. We’ve noticed that as the organization get more mature on AWS, --Tactical work drops as you adopt AWS and even further as you mature in your adoption of the cloud --Enable your teams to do more strategic work which is more effective and aligns more closely to your business goals
- There are 6 roles that we typically consider in relation to staff productivity: Server, Network, Storage, Application, Facilities, Security. In our models, each of the 6 roles have about15-30 typical tasks. In this particular example, we’re looking at a subset of tasks that a Server Administrator would be performing. What a server administrator does will vary by organization. Along with the tasks, we indicate the typical reduction in time spent on the task, based on what other AWS customers have achieved. Certain tasks (*Click*) like installing, upgrading and removing software are significantly reduced with AWS --Where as other tasks (*Click*) like arranging for hardware repair in the event of failure are completely eliminated with AWS
- Typically, these tactical work take up a certain percentage of the FTE’s time. In our server administrator example, these tactical tasks, e.g. server budgeting, HW repairmen take up 25% of his/her time [CLICK]. And on average, by moving to AWS, we see an average of 75% reduction of time spent on these tasks[CLICK]. Therefore, [CLICK] you can expect to see 18.8% efficiency gain per FTE.
- Now moving on to the second pillar of staff productivity and we measured these with VMs managed per admin and TB managed per admin, both during on-prem and after migrating to AWS. What we found from benchmarking study was that overall, there was a 57.9% increase in VM managed per admin after migrating and 67.7% increased in TB managed per admin after migrating to AWS. These are the left 2 charts of the slide here. We also had a lot of profiling questions in the study and what was really significant was looking at those customers that rearchitected, refactored or replatformed to fully leverage the functionalities and benefits of the cloud. [CLICK] Their staff productivities metrics saw more than double the improvement seen in the overall populations and those customers that just simply lift and shifted or still co-hosting on the cloud and on-prem.
- Let's move on to the next area of a cloud value framework. Operational resilience And remember, operational resilience is the benefit of improved availability and security that comes with AWS
- What you see on screen here are figures from industry analysts that speak to the cost of outages and the cost of downtime. In terms of the impact of these applications, we have estimated the downtime cost to be $50K to over a million. Now these are large numbers, and the cost of downtime and security risks are extremely high, and the potential impact your business is significant. While these events may not be frequent. When they do happen, the cost is very high, and so it's important to at least include this component into your business case. We have seen some customers who are uncomfortable with dollarizing these benefits, and so they leave these in terms of KPI benefits. For example, availability. And that's the recommended approach, if your organization is uncomfortable, dollarizing these benefits. However, I think the takeaway here is that we should set a list of KPIs and track their progress, because the impact when these events do happen is very real and significant. Share some examples of the low impact and high impact applications. High impact: manufacturing, SAP goes down, no purchase/billing/delay in manufacturing Medium impact: internal troubleshooting system, customer complaint, important, not as urgent, Low impact: printing system, small team impact
- There are several key drivers that impact the cost of downtime and we bucket them into several categories. You can use this as a rough guide to think about the cost of outages of your applications. Next, let’s look at an example of how to use this as a guide to quantify the downtime cost.
- In this example, we will put into perspectives the cost of downtime, and what gains look like. A media company, who is looking to improve the SLAs of their advertisement deliverable. From the 9 categories of cost drivers on our last slide, we started with the business user productivity. [CLICK] Let’s say there are 1,500 employees [CLICK] in the company with an average annual salary of $125K [CLICK]. We use the standard 170 work hours per month and get to the per minute cost per FTE is close to $1 . [CLICK] Therefore, for this media company, the cost of downtime is $1,500 per minute. [CLICK]For the rest of the cost, e.g. equipment, detection, the after-the-fact disruption, [CLICK] amounts to about half of the cost associated with the disruption of business user productivity, [CLICK]which gives us $750 per minute. The total downtime cost totals up to $2,250 per minute. The current SLA of their advertisement deliverable is 99.4%. [CLICK] With AWS they were looking to improve SLAs by 0.1%, which is 525.6 minutes, out of the 525,600 minutes a year. This will give us the uptime gain of [CLICK] $2,250 / min * 525.6 min = $1,182,600. Part Art, part Science
- Now moving on to our 3rd pillar of operational resilience and here are some of the KPIs we measured and these show that customers have reported 56.7% decrease in downtime after migrating to aWS. Further, they reported 24.1% decrease in total monthly incidents and 31.8% decrease in critical incidents after migration. So we have lower down time, lower incidents and lastly, it also showed 39.3% decrease in mean time to resolution for those fewer incidents that emerged.
- Next, we’d like to talk about the last pillar of the business agility. Share the example of Dow Jones.
- The traditional IT model places constraints on innovation. If a team wants to develop a new application or service, it needs to provision resources to do so. This requires additional time and money, slows time-to-market and cuts into innovation budgets. The cost of failure is another potential roadblock. For example, if a company using an on-premises infrastructure tries and fails to launch a new product, it still must support and maintain the hardware it purchased for the project. With the cloud, developers can instantly provision resources and begin writing code. Everyone will feel involved in identifying opportunities of improvement. They don’t have to wait to get the budget to deploy a full infrastructure stack just to try something new because the cloud supports a rapid, agile approach to IT that drives competitive advantage. They can avoid costs because they can shut down instances when they are no longer needed. In the same time, you can choose software solutions from the AWS Marketplace, which is our managed and curated software catalog with an ecommerce storefront and features over 4,000 listings of software in over 30 categories, security, networking, BI, storage, database, operating systems, and business software. You can quickly acquire capabilities in certain areas by deploying these software on AWS without investing development resources in reinventing the wheels. And you can control the cost as soon as you decide to discontinue by stopping the subscription right away.
- Traditionally the software delivery lifecycle goes through six major stages, from requirement & analysis all the way to maintain. On the chart, we also labeled the typical percentage of time spent at each step. By taking into account the AWS Capabilities such as scalability and security, requirement and design are greatly simplified. You don’t need to re-invent the wheel The continuous integration and continuous delivery (CI/CD) pipelines on AWS helps you automate steps in your software delivery process, such as initiating automatic builds and deployment. The wait time can also be reduced or eliminated through automated application development and smaller deployment batches. With the AWS’ and many other third-party managed service options, you can also reduce maintenance costs and simplify operations.
- Now as we’ve mentioned earlier, the sooner we can set up KPIs, the better we can track and understand how we are progressing in different phases of the delivery lifecycle. Depending on what key milestones your teams are tracking, you might have different sets of KPIs you want to track and set targets around. We recommend setting a few KPIs around # of applications, # of features launched, time to market / deploy, response time, customer retention, etc. [CLICK] In our next example, we will walk you through how we think about a specific business agility question.
- One of our media subscriber based company had the choice of upgrading an application that affected their conversion rates on upgrade. So their consent messaging application often times forced users to wait in excess of several minutes. Now, if they were to have upgraded this on a traditional infrastructure, it would have taken six months. They chose instead to upgrade this application in a hybrid manner, upgrading that specific component with AWS, and they were able to do that in one month. So essentially they had gained five months where their conversion rates were much higher because the consent messaging delays went from several minutes down the five seconds. And as you guys can imagine, when your customers who have chosen to upgrade have to wait for five minutes, there's a significant drop off in those conversion rates. So we're able to look at how many clicks were happening on a historical basis. Look at the LTV or lifetime value of that upgrade. Look at the five month period of additional time they had with that application and quantify a very specific dollar benefit.
- Lastly, the benchmarks that cover the last pillar here on business agility shows that we have 37.7% decrease in time to market for new features, 39.9% decrease in time to depoy codes to productions and almost 3 and a half times increase in code deployment frequency. Hopefully, you can use these statistics to ignite the way you can reimagine and transform your business to better serve your customers. I always ask my customers to think about what you can do, and the impact you can have to your customers to be able to deliver enhancements to them 3x fasters. We also have customers where these benchmarking insights are compelling enough to support a cloud migration because the framework with these 4 pillars, the metrics under each and improvements we report are enough for them to think about what they want to measure, track and set improvement goals for their organization.
- Today, I shared with everyone the AWS Cloud Value Framework. The left most value pillar, which is also the conversation starter and the most common reason organizations migrate to the cloud. There are multiple levers that contribute to the reduction of the total cost of ownership. You can avoid making capital investment by paying for what you use at a pay-as-you-go basis. By selecting the right service, right size, and enabling the elastic design in your infrastructure, you will be able to achieve the cost optimization which in turn future drive down your cloud spend. Now, move to the staff productivity, because your staff can now focus on the more strategic workloads, and less on the previous tactical troubleshooting and maintenance work, you will expect to see increased productivity in your infrastructure staff and application developers. The high availability offered by AWS can greatly reduce the unplanned downtime and business disruption for your employees as well as customers. And the money saved from the infrastructure cost, the released staff resources, and resilient system will all reflect in a more agile and innovative company culture, where you will witness faster time to market for your launches, more time and market reach, that will ultimately impact your business bottom line.
- Now you understand the key components in building a business case and their benefits. To wrap up, it is critical to start building your business case early in your decision making process, invite the right stakeholders in the discussion and prepare time for multiple interactions with them. For the areas that are not as obvious to quantify values, make sure your organizations can select the sets of KPIs you value and track them diligently. And again, in order for us to make the right decision, we need to present the overall value, not just the TCO.
- Your team can get started learning with our free digital training courses. We offer over 550 courses, including the 6 hour AWS Cloud Practitioner Essentials course We offer a variety of learning paths by solution and by role to help advance skills or prepare your teams for AWS Certification. Plus, we also offer enterprise-specific resources, such as case studies, e-book, and whitepapers, to help you accelerate and scale your modernization.