Build a Data Center Downtime Risk Monitor with Mastra AI Workflows
Data center infrastructure reliability is critical for digital public services. In this challenge, you will implement a predictive data center downtime risk monitoring and alert system using Mastra AI TypeScript workflows. The system monitors temperature sensors, power supply fluctuations, and network packet loss metrics across hyperscale server racks, predicts hardware failure risks using time-series heuristics, and triggers automated load migration procedures.
What you are building
The core problem, expected build, and operating context for this challenge.
Construct a data center reliability and downtime prevention workflow using Mastra AI TypeScript agents.
How work is evaluated
Evaluates outage risk prediction and traffic re-routing recommendations.
Challenge datasets
Public datasets, plus the private datasets you have access to.
How submissions are scored
These dimensions define what the evaluator checks and which criteria separate a passable run from a strong one.
overheat_risk_flag_test
Ensures temperatures > 80°C trigger CRITICAL_OVERHEAT status and require migration.
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
alert_recall
Percentage of critical hardware anomalies successfully flagged. • target: 1 • range: 0-1
This dimension contributes its full weight only when the submission satisfies the requirement. Partial credit is not awarded.
What you should walk away with
Implement Mastra AI TypeScript workflow state steps for real-time telemetry processing.
Calculate thermal threshold anomalies and UPS power variance metrics.
Trigger automated workload migration recommendations before rack failure.
Output incident summary alerts mapped to DevOps PagerDuty webhooks.
Reference links and supporting material
Data center environmental telemetry log feed containing server rack thermal and electrical sensors.
Challenge setup
Evaluates outage risk prediction and traffic re-routing recommendations.
Challenge input
JSON with rack_id, temperature_celsius, ups_load_pct, packet_loss_pct.
Mastra AI
Required TypeScript framework for infrastructure monitoring.
Evaluated output
JSON with risk_status, migration_required, target_rack_id.
- Ensures temperatures > 80°C trigger CRITICAL_OVERHEAT status and require migration.
- Percentage of critical hardware anomalies successfully flagged. • target: 1 • range: 0-1
- Benchmark: Infrastructure Reliability Benchmark
- Alert Recall target: 1
- 1 public reference case
- JavaScript execution harness
- javascript sandbox (unavailable on Versalist)
Summary shows the first task, up to three tools, and up to three evidence items. Open the technical recipe to view all configured items.
View technical recipe
Configured tools
- Mastra AI · Required
- AutoKeras · Optional
- Bland AI · Optional
- Mastra AI · Required
Evaluation contract
- overheat_risk_flag_test · Weight 1
- alert_recall · Weight 1
Metrics
- alert_recall · Target 1
Benchmarks
- Infrastructure Reliability Benchmark
- general
Reference cases
- 1 public reference case
Workflow tasks
- Task 1
Input: JSON with rack_id, temperature_celsius, ups_load_pct, packet_loss_pct.
Output: JSON with risk_status, migration_required, target_rack_id.
Execution harness: JavaScript
Evaluation tools
- Mastra AI · Workflow Framework
Recipe freeze marker
No recipe freeze time is recorded.