agones-dev/agones

Features

• Game Server Orchestrators - Acts as a specialized Kubernetes controller for hosting, scaling, and managing dedicated multiplayer game servers.
• Game Server Managers - Provides a system for maintaining pools of warm server instances with automated scaling and buffer management.
• Game Server Hosting - Provides specialized hosting for dedicated multiplayer game servers across various engines with health monitoring.
• Automatic Compute Scaling - Automatically adjusts the number of active servers based on real-time load and scaling policies.
• Server Lifecycle Managers - Manages the operational state of servers, marking them as ready, allocated, reserved, or shutting down.
• Capacity-Based Server Allocation - Matches players to a server based on the specific amount of remaining available slots.
• Capacity Scaling - Adjusts the number of running server instances to maintain a minimum threshold of aggregate free capacity.
• API-Driven Fleet Scaling - Adjusts active server replicas through API calls to match demand.
• Cluster Scaling Orchestrators - Manages the lifecycle and scaling of server processes using standard Kubernetes orchestration tooling.
• Multi-Cluster Coordination - Distributes server allocation requests across multiple clusters to balance load and optimize placement.
• Cluster State Reconciliation - Continuously aligns the actual state of game servers with the desired specification through a reconciliation loop.
• Container Lifecycle Signaling - Uses SDKs to communicate readiness and shutdown signals between server binaries and the controller.
• Containerized Server Deployments - Creates individual game server instances with specific container configurations, port mappings, and health checks.
• Declarative Game Server Deployments - Provides declarative management of individual game server instances and fleets via Kubernetes custom resources.
• Custom Resource Definitions - Uses Kubernetes Custom Resource Definitions to represent and manage game servers and fleets.
• Demand-Based Fleet Scaling - Automatically adjusts the number of active game servers in a pool based on current player demand.
• Dedicated Hosting - Provides infrastructure for deploying stable, headless dedicated game server instances.
• Scaling Policies - Implements automated rules for adjusting game server replica counts based on various scaling policies including buffers and schedules.
• Cluster Resource Management - Provides a programmatic API for managing the lifecycle and configuration of cluster resources.
• Kubernetes Cluster Provisioning - Provides automated deployment and lifecycle management of Kubernetes clusters configured to host dedicated game servers.
• Orchestration Client Libraries - Provides dedicated client libraries in multiple languages to connect server processes to the controller.
• Slot-Aware Allocation - Finds servers with sufficient available slots or assigns a fresh server if none exist.
• Ready Server Buffers - Maintains a specified number of servers in a ready state for instant player allocation.
• Remote Server Fleet Management - Maintains pools of server replicas using templates and scheduling strategies to ensure optimal infrastructure usage.
• Fleet Health Maintenance - Ensures a minimum number of servers remain ready by automatically replacing failed instances.
• Session-Aware Fleet Management - Manages the deployment and updating of server groups while preventing the shutdown of active player sessions.
• SDK-Based Lifecycle Control - Provides an integrated SDK for tracking health, managing state, and handling configuration for server processes.
• Server Allocation APIs - Provides a dedicated API for requesting secure game server allocations to ensure player access to the cluster.
• Server Configuration Retrieval - Allows servers to fetch their current status, public IP, and port or watch for real-time configuration updates.
• Server Lifecycle APIs - Agones API for automating server lifecycles through language-specific clients or direct programmatic calls.
• Load-Based Fleet Scaling - Automatically adjusts running server instances based on real-time load and specific buffer requirements.
• Server Metadata APIs - Provides APIs to write and read labels and annotations on server records for observability and discovery.
• Server Provisioning - Automates the provisioning of dedicated game server instances by deploying pods and monitoring sidecars.
• Warm Buffer Management - Maintains a specific number of idle servers to enable instant player connections without startup latency.
• Heartbeat Monitors - Uses periodic heartbeats from servers to the controller to prevent automatic restarts of healthy processes.
• State Reconciliation - Automatically synchronizes the actual state of game servers with the desired specification.
• Game Engine Integrations - Ships toolkits that connect various game engine server logics to a central hosting controller.
• Game Server SDKs - Provides a dedicated SDK and client libraries for integrating game server processes with the hosting controller.
• Server Logic Integrations - Agones language-specific toolkits that link a server process to a hosting controller to manage the server lifecycle.
• Container Port Mapping - Assigns and exposes UDP port ranges on host nodes to route external player traffic to containers.
• IP and Port Allocations - Assigns network ports using static, dynamic, or passthrough policies to enable external client connections.
• Latency-Based Ranking - Determines the optimal cluster for a client by ranking endpoints based on measured network response times.
• Network Port Configuration - Manages the opening of specific UDP port ranges on worker nodes to allow external game traffic to reach servers.
• Readiness Notifiers - Provides mechanisms for servers to signal their readiness to the controller after starting.
• Dedicated Server Provisioning - Automates the creation of dedicated server instances with configurable network ports and container templates.
• System Resource Tracking - Maintains user-defined counters and lists on server instances to track resource usage and player availability.
• Automated Unhealthy Instance Replacement - Automatically replaces servers failing health checks to ensure the target number of healthy replicas is maintained.
• Comprehensive Readiness Checks - Maintains a pool of warm servers and verifies their readiness for immediate player allocation.
• Server Shutdown Signaling - Signals the controller that a server is exiting so underlying Kubernetes resources can be reclaimed.
• Server Health Monitoring - Tracks the operational state of servers using periodic signals to detect and replace failed instances.
• Host Status Tracking - Monitors server lifecycle states to determine readiness and retrieve network identity information.
• Liveness Signaling - Implements periodic pings to a coordinator to prove server liveness and prevent termination.
• Game Server Lifecycle Controls - Implements integrated signaling for server readiness, health pings, and shutdown triggers.
• External Webhook Scaling - Delegates scaling logic to external HTTP endpoints to compute target replicas based on custom business requirements.
• Allocation State Atomicity - Performs atomic updates of server labels and state during allocation to control player availability.
• Fault Domain Isolation - Deploys servers across diverse geographic regions and availability zones to isolate faults and minimize player latency.
• Automatic Restart Policies - Automatically restarts game server processes if health monitoring detects they have entered an unhealthy state.
• Capacity Counters - Tracks and modifies integer counters with defined capacities to monitor server load and available slots.
• Local Development Servers - Enables connection of local server processes to the management environment for rapid development and iteration.
• Custom Orchestration Controllers - Allows the creation of custom controllers to monitor game server events and automate metadata modifications.
• Placement Optimizations - Provides strategies to pack game servers onto minimum nodes for cost efficiency or distribute them for load balancing.
• Multi-Cloud Deployments - Supports provisioning and distributing managed clusters across multiple cloud providers using infrastructure-as-code.
• Cloud Infrastructure Cost Optimization - Schedules server placements across environments to reduce overall infrastructure costs.
• Cloud Network Security Groups - Opens specific UDP port ranges in cloud security groups to permit external traffic to reach servers.
• Shutdown Prioritization - Controls the sequence of server deletion during scale-down to ensure the least impactful instances are removed first.
• Scheduled Capacity Events - Allows pre-scheduling of server capacity increases to prepare for known upcoming traffic spikes.
• Cluster Upgrades - Manages rolling updates of controllers and server fleets within a cluster to apply new versions without downtime.
• Wasm-Based Scaling - Executes WebAssembly modules to calculate required server replica counts using custom external business logic.
• Image Pre-pulling - Monitors fleet requirements to pull container images onto nodes before servers start, minimizing startup latency.
• Control Plane Resource Caching - Implements an in-memory cache of cluster objects with event watching to reduce load on the Kubernetes control plane.
• Cross-Cluster Workload Migration - Enables the transition of server allocations between clusters to support version upgrades with minimal player interruption.
• Custom Resource Controllers - Supports the creation of custom controllers to monitor server state changes and programmatically update metadata.
• Canary Deployment Controllers - Implements the deployment of a small subset of new server versions to validate stability before a full rollout.
• High Availability Control Planes - Ensures continuous availability of the orchestration plane by distributing controller responsibilities across multiple replicas using leader election.
• Game Workload Primitives - Provides foundational primitives for building complex infrastructure and proxy setups for game workloads.
• Regional Allocation Routing - Redirects allocation requests across multiple clusters to ensure regional instance assignment.
• SDK-Driven Shutdowns - Allows the orchestrator to trigger a running server instance termination via a shutdown command to the SDK.
• Leader Election - Implements leader election to distribute controller responsibilities and ensure high availability of the management plane.
• Policy Chaining - Supports a prioritized list of scaling rules in sequence to provide reliable fallback mechanisms for server capacity.
• Infrastructure as Code Tools - Supports provisioning managed clusters and dedicated node pools using declarative infrastructure-as-code tools like Terraform.
• Production Cluster Deployers - Automates the provisioning and lifecycle management of production-grade clusters and their controllers using infrastructure-as-code.
• Cluster Autoscalers - Integrates with cloud provider autoscalers to dynamically adjust the number of worker nodes based on resource demand.
• Version Upgrades - Manages the upgrade process for the controller and underlying cluster to ensure service continuity during version changes.
• Azure Kubernetes Service Deployments - Supports provisioning and managing dedicated game server workloads on Azure Kubernetes Service (AKS).
• Label-Based Selection - Selects available server instances by filtering custom metadata and state labels through the API.
• Server Metadata Synchronization - Updates labels and annotations on running servers to track operational state and configuration.
• Pod Disruption Control - Manages the eviction of server pods during node replacement to prevent unplanned player disconnects.
• Lifecycle Time-to-Live Policies - Provides automated resource reclamation by deleting game server fleets after a specified time-to-live period.
• Fleet Configuration Updates - Deploys new server configurations across a fleet using various update patterns to modernize server instances.
• Resilient Infrastructure - Configures orchestration services to maintain continuity and resilience during infrastructure disruptions or component failures.
• Server Distribution Control - Controls whether servers are packed for cost efficiency or spread across the cluster for higher availability.
• Topology-Aware Distribution - Uses topology constraints to spread server replicas across nodes and zones to prevent single points of failure.
• Session-Aware Node Eviction - Controls whether a server can be terminated by the cluster autoscaler based on active session lengths.
• Sidecar Containers - Deploys auxiliary containers like log shippers or monitoring agents alongside the primary game server.
• Health Tracking Sidecars - Uses sidecar processes to perform health checks and maintain operational counters for server instances.
• Wasm-Based Scaling Logic - Executes custom WebAssembly modules to compute the optimal number of server replicas based on real-time fleet state.
• Lobby & Matchmaking Services - Allows servers to announce their availability to external matchmaking services for player routing.
• Matchmaker-Integrated Orchestration - Orchestrates the starting, allocation, and shutdown of dedicated servers to integrate seamlessly with matchmakers.
• Server Resource Lists - Maintains string sets via appending and removing to track server attributes within capacity-limited lists.
• Ingress Controllers - Automatically configures network ingress rules to manage external traffic flow to game servers.
• Network Address Resolution - Identifies the best reachable network address for a server using priority DNS or IP options.
• Allocation Tie-breaking - Selects the most appropriate server by filtering for capacity and applying priority tie-breakers.
• UDP Port Range Configuration - Opens specific port ranges in network security groups to permit client UDP traffic to worker nodes.
• Weighted Cluster Allocation - Redirects allocation requests across multiple clusters based on priority levels and weights.
• Cross-Cluster Security - Configures mutual TLS and trusted CAs for secure communication between allocation services across different clusters.
• RBAC Manifest Generation - Automatically creates the required service accounts and roles to enforce RBAC for server management.
• Pod-Level RBAC Configuration - Defines RBAC roles and service accounts to restrict the actions a game server process can perform.
• Rolling Updates - Implements rolling update strategies to replace game server instances with new versions while protecting active player sessions.
• Port Range Definitions - Defines minimum, maximum, and named port ranges used for dynamic allocation to game server processes.
• Request Batching - Queues multiple allocation requests against a local cache to increase throughput and reduce control plane latency.
• Capacity Monitoring - Exports aggregate gauge metrics to monitor total server capacity and real-time usage.
• Cluster Latency Probing - Provides echo endpoints for timing roundtrip requests to determine the most responsive cluster.
• External Event Scaling Triggers - Enables the use of external HTTP webhooks as triggers for automated server scaling based on external business logic.
• Game Session Statistics Tracking - Maintains real-time counters and lists to monitor player counts and server capacity.
• Health Checks - Provides synchronous signaling to the controller to verify that the game server process is responsive.
• Microservices Health Monitoring - Deploys sidecar containers to track heartbeats and report the operational health of server processes.
• Server Metrics - Allows updating predefined counters and lists on server records to track operational data.
• Server Status Monitors - Monitors real-time server instances for configuration or status updates with triggered callbacks.
• Service Account Management - Specifies custom service accounts in server configurations to manage RBAC and security identities.
• Network Identity Services - Creates internal DNS records using headless services and custom subdomains for consistent server communication.
• Network Latency Testing - Provides a ping service for clients to measure round-trip time to servers for connectivity testing.
• Performance Metrics Exporters - Collects and exposes operational data on server counts and allocation latencies via monitoring tools.
• Real-Time Server Activity Monitors - Enables retrieval of server details and real-time subscriptions to configuration and status updates.
• Kernel Resource Limiting - Enforces CPU and memory requests and limits for server processes to ensure predictable resource allocation.
• Unified State and Metrics Tracking - Implements combined monitoring of server health, player capacity, and custom integer counters for comprehensive observability.

Star history