Skip to main content

Core gRPC Services

The ARES architecture is decoupled into a set of specialized microservices that interact via high-performance gRPC (HTTP/2) interfaces. These services manage automation workflows, connect to physical equipment, handle analytical evaluations, and orchestrate real-time event streaming.

1. Core Workflow Orchestration: AresAutomation​

The AresAutomation service is the main orchestrator for the entire autonomous loop. It controls the lifecycle of a CampaignTemplate from startup to closeout, manages multi-user project spaces, handles live runtime execution streams, and saves historical summaries.

  • Campaign Management: GetAllCampaigns, GetSingleCampaign, AddCampaign, UpdateCampaign, RemoveCampaign.
  • Execution Control: StartExecution, StopExecution, PauseExecution, ResumeExecution.
  • Live Streams: Opens persistent server-streaming connections to monitor execution parameters in real-time (GetExecutionStatusStream, GetCampaignExecutionStateStream).
  • Intervention: SubmitUserDecision lets users handle non-trivial execution exceptions (e.g., triggering replanning or pausing for manual refills via an ErrorHandling enum).
  • Auditing: GetCampaignSummary retrieves compiled execution data logs for post-experiment analytics.

2. Hardware Middleware: AresDevices​

The AresDevices service manages the physical or simulated instrument layer. It keeps track of driver instances, relays device telemetry, and fires raw commands.

  • Inventory Tracking: ListAresDevices returns a full inventory of available equipment along with their capability specifications (DeviceInfo).
  • Telemetry Streaming: GetDeviceStateStream provides an async server-to-client pipeline of high-speed telemetry channels.
  • Instruction Dispatch: ExecuteCommand sends an uncompiled CommandTemplate down to the assigned driver layer for immediate execution.
  • Logging Control: SetDeviceLoggerSettings configures the persistent database recording filters (INTERVAL or ON_CHANGE).

3. Optimization and Closed-Loop Systems​

AI & Optimization Drivers: AresPlannerManagementService​

Manages the configuration and routing rules of AI/ML planning algorithms.

  • Algorithmic Inventory: GetAllPlanners returns all active planning capabilities.
  • Direct Seeding: SeedManualPlanner injects explicit point arrays or file paths directly into the trial queue, bypassing active learning models when doing baseline calibration runs.

Performance Analytics: AresAnalysisService​

Triggers the physical data parsing routines after an experiment wraps up.

  • Evaluation: Analyze takes the target inputs/settings and returns an evaluated performance Analysis outcome metric.
  • Validation: ValidateInputs verifies that data shapes conform to expected analytical bounds before processing.

4. System Support & Utilities​

Safety Controls: AresSafetyService​

The safety-critical channel within ARES.

  • Emergency Halt: RequestEmergencyStop triggers a zero-latency, high-priority safety override across all active physical device drivers to shift equipment into designated hardware safe states.

Language Server Engine: AresScriptingService​

Powers the autocomplete and developer IDE backend features.

  • LSP Methods: GetCompletions, ValidateScript, GetScriptSummary, GetSignatureHelp, and GetSymbolMetadata.

Global Event Mesh: AresNotificationRpc​

An asynchronous alerting engine that broadcasts system-wide notifications using a unified severity spectrum (INFO, WARNING, ERROR, DANGER, SUCCESS).

  • Subscription: Subscribe opens a long-lived push channel to relay warnings or milestones directly to standard out logs or front-end dashboard panels.

System Diagnostics: AresServerInfo​

Tracks node health and resource states.

  • Heartbeat: GetServerStatusStream broadcasts server states using a simple state enumeration:
    • IDLE: Open and awaiting a campaign assignment.
    • BUSY: Actively executing trial sequences or parsing data.
    • ERROR: A system fault occurred requiring user intervention.
    • STOPPING / STOPPED: Orderly service teardown states.

Telemetry Archiving: DeviceStateExportService​

Handles bulk data transportation.

  • Exporting: GetStateExport gathers vast historical time-series sensor readings, compiles them based on a time-range filter, and serializes them into bulk formats (COMBINED binary sheets or ZIPPED collections) for deep statistical offline modeling.