Control Panel

The Control Panel is a Next.js 16 web application that serves as the unified management and analytics interface for the entire AetherLake platform. It provides a cohesive, enterprise-grade design system across all screens with real-time Kubernetes telemetry, distributed query interfaces, stream orchestration, and Iceberg metadata exploration.

---

🏛️ Enterprise Design System & Core Features

Every view in the Control Panel adheres to a standardized corporate format:

• Standardized Header: Unified icon badge, title, subtitle badge, and real-time refresh action.
• 4-Card KPI Metric Grid: High-level operational metrics (health, active nodes, storage, throughput) displayed prominently above each view.
• Unified Filtering & Search: Instant, debounced search bars and category filters across all explorer trees and data grids.
• Bilingual Support (i18n): Fully internationalized in English (en) and Turkish (tr) with runtime switching and zero layout shifts.
• Enterprise Security: SSO integration with Keycloak OpenID Connect (OIDC), JWT role forwarding, and credentials-based fallback.

---

📊 Platform Overview (/)

The Platform Overview acts as the central mission control dashboard:

• Cluster KPI Metrics: Live summary cards showing Total Active Services, Overall Cluster Health %, Cloud/K8s Environment, and Telemetry connection status.
• Quick Launchpad: One-click action cards to jump straight to the SQL IDE, Iceberg Explorer, Flink Studio, dbt Lineage, or Observability.
• Category Filter Pills: Filter platform components instantly across All, Lakehouse, Compute, Streaming, and Orchestration.
• Instant Search: Search services by name, description, or internal/external endpoints.
• Service Controls: Trigger service restarts and view live pod phase indicators.

---

❄️ Iceberg Tables Explorer (/tables)

The Iceberg Tables page is a rich metadata and data inspection workbench connecting directly to the Apache Polaris Iceberg REST catalog:

• Namespace & Table Search: Search bar in the catalog sidebar to instantly filter through namespaces and tables.
• One-Click Clipboard Actions: Copy Table Reference (iceberg.<ns>.<table>) and S3 Storage Location directly to clipboard.
• 4-Card Metric Grid: Real-time snapshot statistics displaying Total Rows, Total Data Files, Total Table Size (formatted in bytes/MB/GB), Format Version, and Schema Column count.
• 5-Tab Deep Inspector:
  1. Overview: Snapshot summary, file format, and table location.
  2. Schema: Full column-level schema definitions with Iceberg types and nullability constraints.
  3. Partitions: Partition specifications, transforms (e.g. identity, bucket, hour), and source columns.
  4. Snapshots: Complete table commit history, operation types (append, overwrite), timestamp, and added rows.
  5. Data Preview: Live data querying executing SELECT * FROM iceberg.<ns>.<table> LIMIT 20 via Trino. Renders an interactive, scrollable data table with column type indicators, row counters, and error fallbacks.

Demo dataset

A fresh install seeds a small demo dataset — iceberg.demo.events (partitioned by event_type) and iceberg.demo.users — via a post-install hook, so this page and the SQL IDE have data immediately out of the box.

---

⚡ SQL IDE (/query)

The SQL IDE provides a browser-based Monaco SQL editor integrated with Trino's distributed query engine:

• Role-Based Execution: For Keycloak logins, the panel forwards your access token and Trino executes queries under your username. Role-based access control (RBAC) is enforced server-side.
• Schema & Table Search: Real-time filter input in the data catalog explorer to instantly search through catalogs, schemas, and tables.
• Quick SQL Snippet Pills: One-click insert pills for common queries (SHOW CATALOGS, SHOW SCHEMAS, SHOW TABLES, LIMIT 50).
• Execution Performance Stopwatch: Measures exact query execution duration in milliseconds (e.g. 85 ms) alongside row and column metrics.
• Persistent Query History: Slide-over drawer persisting past executed queries in localStorage with execution status, query duration, timestamps, and one-click re-run capability.
• Data Export:
  • Export CSV: Download query result sets directly as .csv files.
  • Copy JSON: Copy formatted JSON records directly to clipboard for API testing.

---

📡 Apache Kafka Management (/kafka)

The Kafka page (/kafka) introspects Strimzi custom resources to manage event streams:

• Cluster KPI Grid: 4-card overview displaying Cluster Status (Ready), Total Topic count, Total Partitions (calculated across all topics), and Online Broker ratio.
• Listeners & Endpoints Banner: Displays internal plaintext (aetherlake-kafka-bootstrap:9092) and external endpoints with one-click copy buttons.
• Topic Search Filter: Instant topic search filter bar.
• Deep-Link Shortcuts per Topic:
  • Query in SQL IDE: One-click shortcut that opens the SQL IDE pre-populated with SELECT * FROM kafka."default"."<topic>" LIMIT 50;.
  • Stream in Flink: One-click shortcut that opens Flink Studio pre-populated with streaming table DDL and query for that topic.
• Topic Inspector: Detailed table partition counts, replica counts, Strimzi reconciliation conditions, and topic configuration overrides.

---

🌊 Apache Flink Streaming Studio (/flink)

The Flink page (/flink) provides a full stream processing workspace backed by the Flink Kubernetes Operator:

• Stream Studio KPI Grid: Total Jobs, Running Jobs (with live status dot), Failed/Suspended Jobs, and Operator Mode (Flink K8s Operator v1.10 · HA Native).
• Streaming SQL Template Selector: Pre-packaged enterprise streaming templates ready to run:
  1. Datagen → Kafka: Synthetic event stream generator for pipeline testing.
  2. Kafka → Iceberg Bridge: Real-time event ingestion streaming directly into Apache Iceberg lakehouse tables.
  3. Tumbling Window Aggregation: 1-minute event-time tumbling window aggregation computing counts and revenue metrics.
  4. Stream Filter & Routing: Low-latency threshold filtering routing alerts to secondary Kafka topics.
• Kafka Topic Explorer Sidebar: Searchable list of Kafka topics with partition counts. Clicking a topic appends a ready-to-run streaming table DDL into the editor.
• State Filter Tabs: Instant filtering of submitted jobs across All, RUNNING, SUSPENDED, and FAILED.
• Job Lifecycle Management: Track parallelism, start time, execution errors, view submitted SQL statements in a modal, or cancel running jobs.

---

🌐 Apache Polaris Management (/polaris)

The Polaris Management page provides governance over the Apache Polaris REST Catalog:

• Polaris KPI Grid: Configured Catalogs, Namespaces in Selected Catalog, Catalog REST URI, and Credential Vending status (OAuth2 · RBAC Vended).
• Catalog & Namespace Search: Instant search filtering catalogs and namespaces.
• Explore in Tables: Direct shortcut link on each namespace to explore its tables in /tables.
• Client Integration Snippets Tab: Copy-paste integration snippets with one-click copy buttons for:
  • Trino: etc/catalog/iceberg.properties REST catalog configuration.
  • Apache Spark: PySpark SparkSession builder configuration for Polaris REST Catalog and S3FileIO.
  • PyIceberg: Python client initialization using pyiceberg.catalog.load_catalog.
• Catalog & Namespace Creation: Create internal REST catalogs with custom S3 warehouse paths and namespaces without CLI tools.

---

⚡ Trino Analytics Management (/trino)

The Trino Management page monitors and configures the distributed SQL query engine:

• Engine KPI Grid: Configured Catalogs count, Active Worker Nodes, Coordinator Health Status (HEALTHY), and Architecture (Distributed · Fault-Tolerant Execution).
• Catalog Search Filter: Search through configured catalogs by name or connector type (iceberg, hive, postgresql, mysql, tpch).
• Query in SQL IDE: Every catalog row includes a direct shortcut button to open the SQL IDE configured to that catalog.
• Catalog Config Expander: View raw catalog properties with sensitive secrets automatically masked (••••••••).
• Add Catalog Wizard: Interactive modal supporting Iceberg REST, Hive Metastore, PostgreSQL, and MySQL connector templates.
• Trino Web Dashboard: Embedded native Trino UI for viewing live worker threads, stage execution trees, and distributed query plans.

---

🩺 Observability & Logs (/observability)

The Observability page provides comprehensive cluster diagnostics without leaving the Control Panel:

• Workload KPI Grid: Total Pods, Healthy Pods, Degraded/Pending Pods, and Total Pod Restarts across the namespace.
• Pod Search Filter: Instant search filter in the pod selection list to quickly isolate specific services or failed containers.
• Live Log Streamer:
  • Live log streaming (follow=true) using native browser ReadableStream chunk decoding.
  • Auto-scroll lock to stay at the tail of incoming log streams.
  • Client-side log text search highlighting.
  • Container switcher for multi-container pods.
  • Tail lines control (100, 500, 1000, 5000).
  • Download: Export complete log files (.log) with timestamps.
• Kubernetes Events: Real-time Kubernetes events sorted chronologically with Warning events highlighted.
• Pod Resource Metrics: Live CPU (millicores) and RAM (MiB/GiB) metrics from Kubernetes Metrics API.

---

🔄 dbt Lakehouse Workspace & Lineage (/dbt)

The dbt workspace (/dbt) provides visual management for Lakehouse data transformations:

• Interactive Lineage DAG: Scalable React Flow canvas spanning Bronze sources, Silver curated tables, and Gold analytics marts with directional Bezier curves, arrowheads, animated flow pulses, Dagre auto-layout, interactive radar minimap, and fullscreen mode.
• Model Inspector: Embedded Monaco editor inspecting raw Jinja SQL, compiled Trino SQL, partitioning specs, and model documentation.
• Dependencies Explorer: Clickable upstream parent and downstream child navigation.
• Data Quality Assertions: Review column-level schema definitions and dbt test validations (unique, not_null, accepted_values).
• Run & Test Triggers: Trigger dbt run and dbt test against the Trino cluster and monitor live execution duration and logs.

---

💻 Running Locally

To run the Control Panel locally in development mode outside of Kubernetes:

cd control-panel
npm install
npm run dev
# Starts on http://localhost:3000

To run linting and compile production builds:

cd control-panel
npm run lint
npm run build