← PRODUCTS / CLOUD OPERATIONS
P-02○ Beta

DBOps

ETL & data pipeline console

A web-based ETL console that manages the entire lifecycle of your data pipeline on one screen — from source connection to mart loading, quality, anomaly detection, and cost. An information-dense data operations console in the style of Notion / Retool.

Web
DBOps — Console
subscriptions.sqlusers.sql+▶ RUN
1SELECT id, email, plan, status, mrr
2FROM public.subscriptions
3WHERE status IN ('active', 'trial')
4ORDER BY id DESC
5LIMIT 5;
Result5 rows · 12 ms
idemailplanstatusmrr
8421yuki@studio.jpProactive¥12,800
8420tanaka@acme.coTeamactive¥48,000
8419lee@uni-sys.devProtrial
8418ops@northtokyo.ioTeampaused¥0
8417park@seoulpay.coProactive¥12,800
/02PROBLEM

The problems DBOps
solves

This product was built to solve the recurring operational and development challenges found in real environments like the ones below.

CONTEXT

A web-based ETL console that manages the entire lifecycle of your data pipeline on one screen — from source connection to mart loading, quality, anomaly detection, and cost. An information-dense data operations console in the style of Notion / Retool. Existing solutions were either not operator-friendly or built in a way that piled up maintenance costs. UNISYS redesigned this into a product that's owned all the way through operations.

/03FEATURES

Key Features

The core features DBOps provides. Every feature is delivered in an operable form.

F-01

Operations dashboard · Real-time monitoring

Monitor pipeline health, throughput, and system status on one screen. Provides a Dashboard (6 KPIs · sparklines · ingestion line chart · day×hour heatmap), a live running-job stream (rows/sec · progress bar), CDC monitoring (lag · throughput, 1.5s refresh), and real-time log streaming of job run history.

F-02

Visual ETL builder

Connect the full source → staging → mart path node-by-node, no code. An 8-step pipeline Wizard, node-based Table Mapping / Column Mapping (ReactFlow), 22 transform functions + 8 table operations (30+ transforms), and per-step YAML/SQL preview with Dry Run.

F-03

Data governance

Manage the meaning, owner, sensitivity, and contracts of data assets and block changes upfront. Includes a data catalog (PII · sensitivity labels), Schema Diff (auto migration SQL), Schema Contract (automatic breaking-change detection), an immutable audit log (before/after diff · 24-month retention), and an RBAC permission matrix.

F-04

Data quality · Anomaly detection

Catch issues statistically before data breaks. 9 quality rules (not_null · unique · min/max · accepted_values · row_count · freshness · referential · custom_sql), a rule builder (auto SQL generation), and anomaly detection with 3 statistical methods (z-score · EWMA · PSI) plus incident tracking.

F-05

Operations · Cost optimization

Go beyond observation to control and cost. Cost tracking (rows · bytes · CPU · storage) with savings recommendations, safe historical reprocessing (Backfill · impact estimation · Dry Run), freshness SLA & violation heatmap, Incidents (MTTA/MTTR · postmortems), and GitOps/PR-based change tracking.

F-06

Multi-environment · Localization · Command palette

A console with built-in safeguards for global teams. An environment switcher (prod / stg / dev, color guards to prevent destructive actions), a command palette (⌘K) for unified search/navigation, localization (Korean · 日本語 · English), and full token-based light/dark themes.

/04ARCHITECTURE

Technical Architecture

An overview of the technologies DBOps runs on.

C-01
Frontend
React 19 · Vite 6
C-02
Visualization
@xyflow/react 12 (ReactFlow) — 5 custom node types
C-03
Sources / Targets
MySQL · MariaDB · PostgreSQL → S3 / R2 (Raw) · PostgreSQL (Staging) · ClickHouse (Mart)
C-04
Cloud / Deploy
Cloudflare Workers · Pages (wrangler)
/05FAQ

Frequently Asked Questions

The questions we hear most often when evaluating adoption.

A basic setup typically takes 2–3 business weeks; 6–8 weeks when integrating many sources and custom transforms. The 8-step Wizard and automatic schema inference (PK · incremental candidates · PII) shorten initial pipeline setup.
Yes. We provide a separate license to run it on your own infrastructure (read-only accounts · read replicas). Source passwords are protected with Vault, and sensitive columns with Mask / Tokenize / KMS encryption.
It supports direct connections to MySQL · MariaDB · PostgreSQL and CDC (WAL / binlog) streaming, with alerts delivered to Slack · Email · Webhook · Teams channels. Pipeline YAML · dbt models are managed via a GitOps · PR flow.
Schema Contract's automatic breaking-change detection blocks failures upstream, and PII detection · sensitivity labels · KMS encryption · RBAC · MFA · 24-month immutable audit log meet security and compliance needs.
Depending on the package, you can choose a business-day response SLA or a 24/7 operations SLA. The in-product SLA & freshness monitor tracks per-table freshness thresholds (5m–1d) and violations in real time.
START A PROJECT

Ready to start your project?

If you're planning a software product, SaaS, or cloud infrastructure build,
let's discuss your project with UNISYS.

Start a Project View Products