Cost Optimization at Scale
•
4 likes•3,638 views
Intended for customers who have (or will have) thousands of instances on AWS, this session is about reducing the complexity of managing costs for these large fleets so they run efficiently. Attendees will learn about common roadblocks that prevent large customers from cost optimizing, tools they can use to efficiently remove those roadblocks, and techniques to monitor their rate of cost optimization. The session will include a case study that will talk in detail about the millions of dollars saved using these techniques. Customers will learn about a range of templates they can use to quickly implement these techniques, and also partners who can help them implement these templates. Presented by: Guy Kfir, Senior Account Manager, Amazon Web Services Customer Guest: David Costa, CTO, Fredhopper
Report
Share
More Related Content
What's hot
What's hot (20)
Similar to Cost Optimization at Scale
Similar to Cost Optimization at Scale (20)
More from Amazon Web Services
More from Amazon Web Services (20)
Recently uploaded
Recently uploaded (20)
Cost Optimization at Scale
- 1. © 2016, Amazon Web Services, Inc. or its Affiliates. All rights reserved. Guy Kfir Sr. Account Manager Amazon Web Services May 24, 2016 Cost Optimization at Scale David Costa CTO Fredhopper
- 2. What to expect…. We will introduce our approach for building the business case for moving to the cloud and share tips from some of our most innovative customers who are able to successfully architect for cost optimization in order to realize the economics of the AWS cloud.
- 3. In the beginning . . . …there was TCO
- 4. What is TCO? Definition: Comparative total cost of ownership analysis (acquisition and operating costs) for running an infrastructure environment end-to-end on-premises vs. on AWS. Used for: 1) Comparing the costs of running an entire infrastructure environment or specific workload on-premises or in a co-location facility vs. on AWS 2) Budgeting and building the business case for moving to AWS
- 5. So how do we do it? ≠
- 6. TCO = acquisition costs + operations costs Hardware—server, rack chassis PDUs, Tor switches (+maintenance) Software—OS, virtualization licenses (+maintenance) Facilities cost Hardware—storage disks, SAN/FC switches Storage admin costs Network hardware—LAN switches, load balancer bandwidth costs Network admin costs Server admin virtualization admin4 The diagram doesn’t include every cost item. For example, software costs can include database, management, and middle-tier software costs. Facilities cost can include costs associated with upgrades, maintenance, building security, taxes, and so on. IT labor costs can include security admin and application admin costs. Space Power Cooling Facilities cost Space Power Cooling Facilities cost Space Power Cooling Server costs Storage costs Network costs IT labor costs 1 2 3 illustrative
- 7. Resources to get you started AWS TCO Calculator https://awstcocalculator.com Case studies and research http://aws.amazon.com/economics/
- 8. So you’re feeling pretty good.
- 9. Cost optimization is… going from… to… pay for what you use pay for what you need
- 10. Where do you start?
- 11. The four pillars of cost optimization Right-sizing Reserved Instances Increase elasticity Measure, monitor, and improve
- 12. Right-sizing Right-sizing • Selecting the cheapest instance available while meeting performance requirements • Looking at CPU, RAM, storage, and network utilization to identify potential instances that can be downsized • Leveraging Amazon CloudWatch metrics and setting up custom RAM metrics Rule of thumb: Right size, then reserve. (But if you’re in a pinch, reserve first.)
- 13. Reserved Instances Commitment level 1 year 3 year AWS services offering RIs Amazon EC2 Amazon RDS Amazon DynamoDB Amazon Redshift Amazon ElastiCache * Dependent on specific AWS service, size/type, and region
- 14. Reserved Instances Step 1: RI Coverage • Cover always-on resources. Step 2: RI Utilization • Leverage RI flexibility to increase utilization. • Merge and split RIs as needed. Rule of thumb: Target 70–80% always-on coverage and 95% RI utilization rate.
- 15. Increase elasticity Turn off nonproduction instances • Look for dev/test, nonproduction instances that are running always-on and turn them off. Autoscale production • Use Auto Scaling to scale up and down based on demand and usage (for example, spikes). Rule of thumb: Shoot for 20–30% of Amazon EC2 instances running on demand to be able to handle elasticity needs.
- 16. Using right-sizing and elasticity to lower cost More smaller instances vs. fewer larger instances 29 m4.large @ $0.12 /hr $2,505.60 / mo* 59 t2.medium @ $0.052/hr $2,208.96 / mo* *Assumes Linux instances in US-East at 720 hours per month
- 17. Putting it all together: case study
- 18. Challenge: Minimizing unit costs under period of massive growth. A consistent measure of CPU processing power Elastic compute unit (ECU)
- 19. The growth challenge August 2014 August 2015 584 ECU 1,192 ECU 2x YoY Compute Growth 33% decrease in monthly EC2 costs!
- 20. Solving the growth challenge
- 21. Step 1: Right-size and update instances m1 on demand $0.07 per ECU c4 on demand $0.02 per ECU
- 22. The impact of right-sizing 70% reduction in unit cost
- 23. Step 2: Reserve
- 24. The impact of reservations 30% reduction In unit cost
- 25. Putting it together 85% reduction in unit cost!
- 26. Sounds pretty easy, right? Not really. In reality, it is very complex.
- 27. David Costa CTO dcosta@fredhopper.com @davidcosta Netherlands May 24, 2016 NBC Congrescentrum
- 29. The solution of choice for eCommerce Search, Navigation and Merchandizing in the Enterprise
- 30. Fredhopper Cloud Services Product Fredhopper Cloud Services Query API Data API Business Configuration API PIM Analyti cs CRM REST endpoints query.published.live1.fas.eu1.fredhopperservices.com/fredhopper/query/ bm.prepublished.live1. … .com/fredhopper/config/campaigns/fashion/en_US/list?label_id=<ID>
- 31. Search Box
- 34. Facet Navigation
- 35. Banner Campaigns
- 36. Promotions
- 37. Search Box Instant Search Suggestions Facet Navigation Flyout Navigation Banner Campaigns Promotions … much more
- 38. Our vision is to be the “most flexible and agile” SaaS- solution for managing digital merchandising strategies & brand experiences on a global scale.
- 39. How are customers using Fredhopper?
- 41. Adoption and scalability drivers 42 Create, manage and deliver engaging cross-channel product experiences on a global scale. Select, rank and serve the most relevant content to the right person at the right time Provide online the level of personalized, dynamic, rich experience of traditional high-end stores from International to Global from Static Shelf to Digital Isle from Transaction to Experience
- 42. Large fleet of computing infrastructure deployed globaly in production
- 44. Including RI amortization and depending on the RIs coverage $ 200-250k monthly Ready to be deployed in 6 of the AWS regions around the globe 4/6 Regions production Computing resources steadily growing over the past 7 years % Constant growth
- 45. We use RIs to obtain a significant Cost reduction 4
- 46. It provides a framework to build a cost structure 4
- 47. EC2 Instances deployed in production across 4 AWS regions EC2 Instances deployed in Development 212 5 195 0 175
- 48. 2 080,10% Reserved Instances 8 5 19,90% On Demand Instances 1 0 >90% Instances Utilization 2015 RIs coverage scheme for production
- 49. Summing it up 50 Around 2000 instances in production, (>5000 vCPUs) On demand monthly around $ 300k With 80% RIs we have a $ 150k EC2 cost per month Savings of 150k per month and around $1.7M per year on EC2 costs
- 51. Realtime dynamic dashboard of OD vs RI utilization Rules RI coverage according to scheme Usage above RI coverage RI coverage above usage Usage outside RI coverage
- 52. Conclusion
- 53. Takeaways from Fredhopper Cost structures beside cost savings provides a culture of utilization and cost awareness It does work! We saved $1.7M last year in EC2 costs The team is rewarded by those cost reductions: part of the savings funds innovation and green fields.
- 54. Thank you!
- 55. Sounds pretty easy, right? Not really. In reality, it is very complex. • Scale • Behavioral change • Visibility • Ownership
- 56. Cost optimization governance (Remember the fourth pillar?)
- 57. Uncovering the cost optimization opportunities 1. Auto-tag resources. 2. Identify always-on nonprod. 3. Identify instances to down-size. 4. Recommend RIs to purchase. 5. Dashboard our status. 6. Report on savings.
- 58. AWS options
- 59. Reserved Instances and right-sizing options
- 60. • Vast selection of software solutions optimized for AWS • Flexible Pricing: Hourly, Monthly and Annually • No cost trials • 1-Click deployment • Easy provisioning • One invoice that includes AWS usage and AWS Marketplace Software AWS Marketplace
- 61. Putting it all together
- 62. Where to start Set up a Cloud Competency Center Bring in the right tools Use metrics to reinforce behavior Use partners to accelerate!
- 63. Cycle of cost optimization ✔ ✔ ✔ ✔✘ ✘ ✘ ✘ $ $ $ $ $
- 64. Thank You!