
Our last blog introduced the Cloud Optimization Strategy (COS) concept and its 3 optimization levers – Commercial, Technical, and Provider. All 3 levers must work interdependently to maximize COS efficacy and efficiency. But what does this synergy look like within your cloud Software Development Lifecycle (SDLC)?
This blog will examine each lever’s optimization practices, improvements, opportunities, and how they work at each of the 3 stages of your cloud software development lifecycle.

Stage 1: Solution (Target) Design
In the Solutions Design phase, Solution Technical Assessments (STAs) assess application compatibility with both cloud environments and business objectives.
Commercial Lever
Optimize resource configurations to limit under-utilization and standardize best practices across solutions for accurate data capture and evaluation:
•
Allocate resources based on real-time utilization reports, not anticipated peak projections
•
Consider “burstable” VMs (e.g. Azure B-series or AWS EC2 instances) when continuous full performance is not needed
•
Replace premium cloud resources (servers, storage, or databases) with standard-level contracts when their performance is acceptable
•
Provision resources in less expensive regions where possible (e.g. if Azure West-US-2 costs more than East-US-2, move!)
Technical Lever
Prioritize Platform-as-a-Service (PaaS) cloud server infrastructure and database configurations. Focusing on services over infrastructure affords a wider range of service options and superior optimization by aligning solutions to needs as closely as possible. Design for, or transition to:
•
PaaS VMs when running common, current operating systems (e.g. Windows or Linux). Avoid attempts to “recreate the wheel” to suit less precise Infrastructure-as-a-Service (IaaS) configurations.
•
Database-as-a-service (DBaaS) rather than more expensive IaaS VMs when running common databases (e.g. Azure SQL DB, Amazon RDS).
•
Database types and configurations as close to solution needs as possible (e.g. key-value store, relational, columnar, or DBaaS).
•
Auto-scaling and load balancing capabilities using PaaS and Solution-as-a-Service (SaaS) services.
Provider Lever
Designate a Primary Cloud Provider (PCP) to host the majority of your workloads, outsourcing those with specific requirements to specialized secondary providers.
•
Ensure your PCP is aligned with and supports your cloud journey. Base selection on:
○
Current and planned workloads, solutions, and capacities
○
Organization-wide technology standards and preferences
○
Technology synergies with provider agreements and discounts
•
Remember to consider provider-specific commercial and technical differentiators when optimizing costs and performance
•
Engage in periodic remediation to ensure CSPs are aligned with your cloud goals and agreements remain competitive

Stage 2: Solution Development And Testing
Application synergies, cloud platforms, and Cloud Service Providers (CSPs) are adjusted to suit evolving needs. Set expected performance efficiencies as targets before benchmarking them against business objectives to measure performance.
Commercial Lever
Service subscriptions require continuous attention throughout the SDLC as cloud objectives evolve. A dedicated cloud vendor management team should lead optimization efforts:
•
Automate Start/Stop VMs based on timed needs and patterns across subscriptions (e.g., development environments, testing and staging, data pipeline usage)
•
Secure 1- and 3-year term VM reserved instances for steady workloads, improving cost efficiencies
•
Acquire Developer/Test subscriptions for heavily discounted access to development software and DB servers
Technical Lever
Focus on re-factoring application designs for PaaS cloud environments. Leveraging automated PaaS services can offload application management to specialist providers, reducing technical complexity and team workloads. Re-factoring examples include:
•
Replace web applications on VMs with Web App Service (Azure) or Elastic Beanstalk (AWS) services
•
Offload manual VM API management to specialized cloud platforms, e.g. Azure’s API Management or AWS’ API Gateway
•
Replace VM-based applications with more efficient, server-less computing configurations like Azure’s Logic Apps or AWS’ Step Functions
•
Eliminate expensive, proprietary database services wherever possible
Provider Lever
Provider optimization can support technical processes by driving the development of cloud-native solutions tailored to each CSP:
•
Direct developers to both PCP- and CSP-specific tools and technologies, with multi-cloud capabilities and cross-provider optimization in mind
•
Ensure migrated workloads with specialized needs are allocated to CSPs with the required core competencies
•
Prioritize PCP cloud-native projects wherever possible, as well as secondary and tertiary providers when the business case supports it

Stage 3: Solution Deployment and Monitoring
Deployed solutions require continuous monitoring and adjustment to stay sustainable and effective. Designing and deploying fixes to resolve emergent inefficiencies quickly is critical to long-term cloud performance.
Commercial Lever
Monitor your solutions’ IT components and change processes with an up-to-date Configuration Management Database (CMDB). Effective CMDBs provide central oversight over deployed solutions, enabling cloud teams to:
•
Audit environments to remove unattached or unused resources
•
Remove orphaned storage after VMs are deleted
•
Remove abandoned subscriptions and associated infrastructure
Technical Lever
Execute continuous reassessment of solution efficiencies against performance targets, implementing operational improvements where possible:
•
Leverage VM Scale sets to automate VM management based on demand thresholds and adjusting them as operating conditions evolve
•
Consider cost-efficient alternatives to database infrastructure once operational cadence has stabilized, such as:
○
Cosmos DB
○
Postgres
○
Mongo DB
○
Key-Value Stores
○
Azure or AWS DBaaS
○
Open-source databases
•
Outsource as-a-service solutions to take advantage of consumption elasticity
Provider Lever
The cloud vendor management team must maintain an optimal mix of CSPs for continuous cost and performance optimization. Consider re-positioning solutions to resolve inefficiencies and sustain performance. Remember to:
•
Fit and mix providers as the technical situation demands. Avoid inefficiencies for the sake of a short provider portfolio.
•
Take the opportunity to consider re-positioning solutions based on resultant provider efficiencies when evaluating solution design.
Coordinating all 3 levers throughout your cloud SDLC is a complex process, even when you know what to do. Misalignments cause inefficiencies that are hard to spot and harder to fix, potentially dragging your cloud journey to a grinding halt. Expert advisory is recommended to optimize your SDLC for sustained performance.
Consult a Wavestone expert for guidance on leveraging your Cloud Optimization Strategy to maximize SDLC performance.
CONTACT USLaying the Foundations: The 4 Core Competencies of Effective Cloud Solutions Security
Sep 21, 2023
Implementing effective cloud solutions security is complex. Solution security architectures must be initially done with minimal mandated standards, then built progressively, with additional layers of security built on foundational capabilities. In this first of a 3-part series on safeguarding cloud solutions, we examine the 4 foundational capabilities of effective solutions security and best practices to implement them.
Wavestone Named to Forbes World’s Best Management Consulting Firms 2023 List
Sep 19, 2023
Forbes has once again recognized Wavestone in its annual “World’s Best Management Consulting Firms 2023” list. Read our blog for more information about the ranking and Wavestone’s award-winning business and technology consulting services.
Have a Question? Just Ask
Whether you're looking for practical advice or just plain curious, our experienced principals are here to help. Check back weekly as we publish the most interesting questions and answers right here.