Gorgeous Cloud supported a mobile game studio with AWS architecture assessment, environment design, migration planning, infrastructure automation, and operational readiness. The engagement delivered deployment templates, monitoring dashboards, recovery procedures, release runbooks, and workload-based cost reviews.
Customer Background
The studio was preparing its SLG/MMO mobile game infrastructure for international expansion. Persistent worlds, player progression, guild interactions, and scheduled events placed different demands on the platform, from long-running world processes to application services that needed to respond to changing traffic.
The infrastructure had to support ongoing player activity alongside new releases and regional growth. Decisions about server placement, capacity, data storage, and maintenance therefore needed to work together. The team engaged Gorgeous Cloud to assess its AWS architecture and strengthen the foundations for deployment and daily operations.
The Challenges
Regional Player Experience
Regional deployment required more than selecting server locations. The studio needed to consider player access paths, world placement, and regional operating requirements together, with clear boundaries between game environments.
Launch and Event Capacity
New releases, season updates, and guild events created different capacity demands. Planning needed to account for both the regular workload and the additional headroom required around scheduled activity.
Persistent Player Data
Characters, inventories, progression, and transaction records had to remain consistent as infrastructure changed. Data protection and recovery procedures needed to reflect the relationship between stored records and running game worlds.
Safe Live Operations
Infrastructure changes and software releases had to account for players already online. The team needed documented deployment and recovery procedures that addressed the dependencies between application services, world processes, and data stores.
Infrastructure Efficiency
Long-running game servers required a stable capacity baseline. The studio also needed to assess event headroom, instance sizing, and resource utilization together when reviewing infrastructure costs.
The Solution Architecture
The architecture separated asset delivery, HTTP-based application services, and persistent world servers into distinct operating layers. This gave each workload an appropriate deployment and capacity model.
Amazon S3 and Amazon CloudFront supported the distribution of game assets, updates, and static content. This separated downloadable resources from the services responsible for player requests and world execution.
Load-balanced containerized services supported HTTP application workloads, with AWS WAF protecting the HTTP interfaces. Stateless services could scale independently of the game worlds they supported.
Amazon EC2 hosted the long-running world and shard processes. Regional environments maintained explicit world ownership and player-data boundaries. Capacity changes for these persistent processes required controlled procedures, state protection, and planned recovery steps.
Amazon Aurora provided transactional storage for persistent player data. Amazon ElastiCache supported frequently accessed data and leaderboard workloads. Amazon CloudWatch brought together operational metrics, logs, and alarms, giving the team visibility across the infrastructure layers.
The design connected workload responsibilities with operating procedures: resource distribution, application scaling, and world-server changes each had distinct requirements. AWS's game database guidance provides additional context on matching data services to gaming access patterns.
How Gorgeous Cloud Helped
Gorgeous Cloud assessed the architecture, designed the AWS environment, and developed migration and operational readiness plans. The work also covered infrastructure automation, monitoring, recovery procedures, release preparation, and workload-based cost reviews.
The engagement connected architecture decisions with the documents and configurations needed to operate the environment. Its deliverables included:
- Deployment templates for consistent environment setup and infrastructure changes.
- Monitoring dashboards and alerting configurations for infrastructure health and workload behavior.
- Backup and recovery procedures for persistent data and supporting services.
- Release and operational runbooks for planned changes and incident response.
- Workload-based cost reviews covering instance sizing, capacity, and resource utilization.
Business Value
Repeatable Deployment
Deployment templates documented a consistent approach to environment setup and infrastructure changes. They gave the team a reusable foundation for preparing environments and managing subsequent updates.
Operational Visibility
Monitoring dashboards, logs, and alarms gave the operations team a clearer view of infrastructure health and workload behavior. They brought information from the different architecture layers into the operational workflow.
Recovery Readiness
Documented backup and recovery procedures established a defined basis for incident response and recovery exercises. The procedures connected data protection with the operating requirements of persistent game services.
Cost Governance
Workload-based reviews connected instance sizing, capacity planning, and resource utilization with infrastructure cost decisions. They provided a framework for reviewing the capacity needed for regular activity and scheduled events.
Discuss Your Game Infrastructure
Planning an international launch or strengthening an existing game platform? Talk to Gorgeous Cloud about AWS architecture, migration readiness, and ongoing operations.
Discuss Your Game InfrastructureIndustry Research & Benchmarks
The following AWS publications provide broader infrastructure context, separate from the customer engagement described above.
Up to 25% Higher Compute Performance per vCPU
AWS reports up to 25% higher performance per vCPU for C9g compared with C8g instances. Suitability and performance gains depend on workload characteristics and compatibility testing. AWS announcement, June 30, 2026.
Microsecond Cache Response Times
Amazon ElastiCache provides microsecond response times for supported caching workloads. This describes cache-layer performance rather than end-to-end player latency. Amazon ElastiCache features.