-
Resiliency
- HA
- What is the solution's SLA?
- Are there any service dependencies that will impact this solution's availability?
- What HA features are available in the current applications? (ex. SQL HA)
- BC/DR
- What is the RTO/RPO? Is the solution business-critical?
- Which regions do you need to deploy to? Are they Paired Regions?
- Will the DR solution need to meet the same production requirements (ex. same/reduced traffic, lower performance)
- How do users access the service now, via IP, URL etc..
- Do you have an existing DR or Incident Recovery Plan?
- How do you perform DR testing?
- HA
-
Security
- What are your biggest threat vectors?
- Do you need Firewall? Egress/Ingress restrictions?
- Are you subject to any regulatory/organizational mandates or policies?
-
Identity
- Do you need to support identities from identity providers external to Azure Active Directory (eg., Google, Facebook, Active Directory Federation Services)?
- Do you need to share your apps and resources with external users? (B2B collaboration)
- Do you need to build user journeys with a white-label identity management solution for consumer- and customer-facing apps (B2C collaboration)
- Refer to the 'Compare External Identities solutions table for more decisions points (https://docs.microsoft.com/en-us/azure/active-directory/external-identities/compare-with-b2c)
- What are you currently using to authenticate your users?
- What are you currently using to authorize user/group/service access to your services?
- Do you need to support identities from identity providers external to Azure Active Directory (eg., Google, Facebook, Active Directory Federation Services)?
-
Data
- Do you have retention policy? How is data backed up?
- Are their policies impacting how the data is shared?
- How much data does the solution generate? How much data do you intake daily?
- Is your data segregated by customer? How do they access the data?
- What kind of data is being stored? What is it used for?
-
AI
- What level of AI is of interest? (Bots, Cognitive Service API, Data Scientist platforms, Machine Learning models)
- What skill sets do you have on staff? (Data scientists, Developers)
- What data sets do you have to leverage? (public/proprietary)
- What feature/functionality are you supporting with the AI?
-
Observability
- Questions: What existing solutions are you using? Provided internally, by another provider?
- Metrics
- What resolution/metrics do you need to understand the performance of your app/understand if you are meeting SLA?
- What is needed at the network level, platform level, application level, user experience level?
- Logs
- How are logs collected? aggregated?
- Alerts
- How do you know if the service experiences problems?
- Who will be receiving the alerts?
- Tenancy
- Key Concepts:
- Application based Multi-tenancy
- Infrastructure based Multi-tenancy
- Hybrid
- Key Concepts:
- Questions:
- How do you onboard a new tenant?
- Does this support a multi-tenancy control plane? Ingress?
- What is the tenancy model of the solution? Is there a shared front end but seperate back-end/database per customer?
- How is the tenancy implemented in the database? Seperate tables, schema, databases on the same server, independent servers?
- What is the tenancy model of supporting services like storage, CDN, VMs?
-
Scaling
- Key Concepts:
- RPS
- Traffic Patterns
- App vs Infra Scaling
- Questions:
- What's your traffic volume at normal load? Peak?
- What is the bottleneck that causes you to scale? (e.g. Memory, CPU, bandwidth)
- Are there any leading indicators to help you scale?
- Do you have scalability/performance requirements? Ex. Number of users, transactions per second, Number of SQL statements running simultaneously
- What is your Y/Y, Q/Q growth? Are there upcoming events that you anticipate growth/traffic spikes for?
- Is your application written with microservices or as a monolith? Can the components be deployed as seperate services?
- Are there closely coupled systems that must also scale or will be impacted when the solution scales up?
- Key Concepts:
-
Governance
- What is currently in place for governance? Who controls environment access?
- Who dictates the enforcement policy and how is it enacted?
- Who dictates the audit policy and how is it enacted?
-
DevOps
- What environments do you have? (e.g. Dev/Test/Prod)
- How are you creating new software artifacts?
- How do you deploy the new software artifacts across your environments?
-
Compliance
- Key Concepts
- SOC, HIPAA, ISO, PCI
- Questions:
- Are you subject to security audits?
- Do you have any compliance requirements?
- How do you currently track compliance? Who is responsible for system compliance?
- Key Concepts