# Streamkap - Complete Product Information

> Zero Ops Data Streaming Platform for Real-Time CDC and Stream Processing

## Company Overview

Streamkap is a fully managed real-time data streaming platform that enables organizations to sync data between systems with sub-second latency. The platform is built on Apache Kafka and Apache Flink, providing enterprise-grade Change Data Capture (CDC) and stream processing capabilities without operational complexity.

**Key differentiators:**
- Sub-second latency (not batch - true real-time streaming)
- Zero infrastructure management required
- 3x lower total cost of ownership compared to competitors like Fivetran
- SOC 2 Type II certified with enterprise security features

## Core Technology

### Change Data Capture (CDC)
Streamkap uses log-based CDC to capture database changes by reading transaction logs (WAL for PostgreSQL, binlog for MySQL, oplog for MongoDB, etc.). This approach:
- Has near-zero impact on source database performance
- Captures all changes including DELETEs
- Maintains exact ordering of events
- Provides millisecond-level latency

### Stream Processing
Built-in Apache Flink integration (called Streaming Agents) allows:
- Real-time data transformations
- Filtering and routing
- Aggregations and windowing
- Custom processing logic in SQL, Python, or TypeScript
- VS Code editor with GitHub sync and version control

### Delivery Guarantees
- Exactly-once semantics where supported
- At-least-once delivery with deduplication
- Automatic schema evolution handling
- Built-in retry and error handling

## Pricing Model

Streamkap uses simple, usage-based pricing based on data volume (the higher of ingress or egress, in GB/month). No per-connector fees. All source and destination connectors included in every plan.

**Tiers:**
- **Starter** ($600/mo): 10GB/month, all connectors, Streaming Agents (SQL, Python, TypeScript), MCP Server, CLI, Terraform, REST API, SOC 2/GDPR/CCPA, 99.9% SLA
- **Scale** ($1,800/mo): 150GB/month, dedicated Kafka and Streaming Agents, dev/prod separation, AWS PrivateLink, SSO, sub-accounts, 99.9% SLA
- **Enterprise** (custom): Unlimited capacity, BYOC (runs in your VPC), VPN, HIPAA/PCI DSS 4.0.1, Multi-AZ, GEO replication, 99.99% SLA, 24/7 support

30-day free trial on Starter with no credit card required. Also available on AWS Marketplace and Snowflake Marketplace (pay with Snowflake credits).

## Source Connectors (30+)

### Relational Databases
| Connector | Description |
|-----------|-------------|
| PostgreSQL | CDC via logical replication (pgoutput/wal2json) |
| MySQL | CDC via binlog with ROW format |
| SQL Server | CDC via SQL Server CDC feature |
| Oracle | CDC via LogMiner or XStream |
| MariaDB | CDC via binlog |
| IBM Db2 | CDC support |

### Cloud Databases
| Connector | Description |
|-----------|-------------|
| Amazon RDS PostgreSQL | Managed PostgreSQL CDC |
| Amazon RDS MySQL | Managed MySQL CDC |
| Amazon Aurora PostgreSQL | Aurora-optimized CDC |
| Amazon Aurora MySQL | Aurora-optimized CDC |
| Azure SQL Database | Azure-hosted SQL Server CDC |
| Google Cloud SQL | Cloud SQL PostgreSQL/MySQL CDC |
| AlloyDB | Google's PostgreSQL-compatible CDC |
| Supabase | Supabase PostgreSQL CDC |
| Neon | Serverless PostgreSQL CDC |
| PlanetScale | MySQL-compatible CDC |

### NoSQL Databases
| Connector | Description |
|-----------|-------------|
| MongoDB | CDC via change streams |
| MongoDB Atlas | Managed MongoDB CDC |
| Amazon DynamoDB | CDC via DynamoDB Streams |
| Amazon DocumentDB | MongoDB-compatible CDC |
| CockroachDB | Distributed SQL CDC |
| YugabyteDB | Distributed SQL CDC |
| Vitess | Sharded MySQL CDC |

### Streaming Sources
| Connector | Description |
|-----------|-------------|
| Apache Kafka | Consume from existing Kafka topics |
| Redis | Key change notifications |
| OpenSearch | Document change capture |

## Destination Connectors (33+)

### Data Warehouses
| Connector | Description |
|-----------|-------------|
| Snowflake | Via Snowpipe Streaming API (lowest latency) |
| Databricks | Delta Lake streaming |
| Google BigQuery | BigQuery Storage Write API |
| Amazon Redshift | Streaming ingestion |
| ClickHouse | Real-time OLAP streaming |
| SingleStore | Real-time analytics |
| Firebolt | Cloud data warehouse |

### Data Lakes & Lakehouses
| Connector | Description |
|-----------|-------------|
| Apache Iceberg | Open table format on S3/GCS/Azure |
| Delta Lake | Databricks open format |
| Amazon S3 | Parquet, JSON, or Avro files |
| Azure Data Lake Storage | ADLS Gen2 streaming |
| Parquet on S3 | Columnar file streaming |

### Databases (as destinations)
| Connector | Description |
|-----------|-------------|
| PostgreSQL | Sync to PostgreSQL |
| MySQL | Sync to MySQL |
| MongoDB | Sync to MongoDB |
| Oracle | Sync to Oracle |
| SQL Server | Sync to SQL Server |
| IBM Db2 | Sync to Db2 |
| Elasticsearch | Search index sync |
| Redis | Cache sync |
| Timescale | Time-series data |
| CrateDB | Distributed SQL |

### Streaming & Messaging
| Connector | Description |
|-----------|-------------|
| Apache Kafka | Produce to Kafka topics |
| Amazon Kinesis | Stream to Kinesis |
| Google Pub/Sub | Stream to Pub/Sub |
| Azure Event Hubs | Stream to Event Hubs |

### Analytics & Search
| Connector | Description |
|-----------|-------------|
| Elasticsearch | Real-time search indexing |
| OpenSearch | AWS OpenSearch streaming |
| Imply/Druid | Real-time OLAP |
| Materialize | Streaming SQL views |
| Rockset | Real-time analytics |

### Emerging
| Connector | Description |
|-----------|-------------|
| MotherDuck | Serverless DuckDB |
| DuckDB | Local analytics |
| Starburst | Trino-based analytics |

## Common Use Cases

### 1. Real-Time Analytics
Stream operational data from PostgreSQL/MySQL to Snowflake or Databricks for live dashboards. Replace nightly batch jobs with continuous streaming for up-to-the-second insights.

**Example pipeline:** PostgreSQL → Streamkap → Snowflake (via Snowpipe Streaming)
**Latency:** < 1 second end-to-end

### 2. Database Migration
Zero-downtime database migrations by running CDC replication in parallel during cutover. Validate data consistency before switching traffic.

### 3. Event-Driven Microservices
Capture database changes as events and publish to Kafka for consumption by downstream microservices. Implement the Outbox pattern without application changes.

### 4. Search Index Synchronization
Keep Elasticsearch/OpenSearch indices in sync with source databases in real-time for instant search updates.

### 5. Cache Invalidation
Stream database changes to Redis for intelligent cache invalidation instead of TTL-based expiration.

### 6. Data Lake Ingestion
Continuously stream CDC data to Apache Iceberg or Delta Lake for cost-effective historical analysis with time-travel capabilities.

### 7. AI/ML Feature Stores
Feed real-time data to feature stores for ML models that require fresh data for inference.

## Security & Compliance

### Certifications
- SOC 2 Type II certified
- GDPR compliant
- CCPA compliant
- HIPAA ready (with BAA)
- PCI DSS 4.0.1 (Enterprise)

### Security Features
- End-to-end encryption (TLS 1.2+)
- Encryption at rest (AES-256)
- VPC peering / AWS PrivateLink support
- IP allowlisting
- SSH tunnels
- SSO (SAML, OpenID, Okta) for Scale+
- SCIM provisioning (Enterprise)
- Role-based access control (RBAC)
- Audit logging
- MFA support
- Bring Your Own Key (Enterprise)

### Deployment Options
- **SaaS Hosted**: Fully managed, multi-tenant
- **Snowflake Native App**: Runs inside Snowflake via Marketplace
- **BYOC (Bring Your Own Cloud)**: Data plane runs in your VPC (AWS, GCP, Azure). Control plane managed by Streamkap. Raw data never leaves your account.

### Data Handling
- No data stored at rest (streaming only)
- Data never leaves customer's cloud region (configurable)
- PII masking and filtering capabilities

## Comparison with Alternatives

### Streamkap vs Fivetran
| Aspect | Streamkap | Fivetran |
|--------|-----------|----------|
| Latency | Sub-second | 5-minute to hourly batches |
| Pricing | Usage-based, 3-8x cheaper | Per-row pricing, expensive at scale |
| Architecture | True streaming (Kafka/Flink) | Batch ETL |
| Transformations | Real-time with Flink | dbt (batch) |

### Streamkap vs Debezium (self-managed)
| Aspect | Streamkap | Debezium |
|--------|-----------|----------|
| Operations | Fully managed | Self-managed Kafka + Connect |
| Setup time | Minutes | Days to weeks |
| Maintenance | Zero | Significant ongoing effort |
| Cost | Predictable SaaS pricing | Hidden infrastructure costs |

### Streamkap vs AWS DMS
| Aspect | Streamkap | AWS DMS |
|--------|-----------|----------|
| Latency | Sub-second | Minutes |
| Destinations | 33+ including Snowflake, Databricks | Limited (mostly AWS) |
| Transformations | Full Flink processing | Basic |

## Getting Started

1. **Sign up**: Create account at https://app.streamkap.com/account/sign-up
2. **Add source**: Configure CDC from your database (PostgreSQL, MySQL, etc.)
3. **Add destination**: Connect to Snowflake, Databricks, or other targets
4. **Start streaming**: Data flows in real-time automatically

**Time to first data:** Typically under 15 minutes

## Support & Resources

- **Documentation**: https://docs.streamkap.com
- **Blog**: https://streamkap.com/blog
- **Resources & Guides**: https://streamkap.com/resources-and-guides
- **Request Demo**: https://streamkap.com/demo-request
- **Contact**: https://streamkap.com/contact-us

## Technical Requirements

### Source Database Requirements
- PostgreSQL 10+ with logical replication enabled
- MySQL 5.7+ with binlog in ROW format
- MongoDB 4.0+ with replica set or sharded cluster
- Network connectivity (direct, VPC peering, or SSH tunnel)

### Developer Tools
- Terraform Provider
- REST API
- CLI
- MCP Server (AI agent integration)

### No Client-Side Installation
Streamkap is fully SaaS - no agents, connectors, or software to install in your environment. BYOC option available for Enterprise customers requiring data residency.

---

For more information, visit https://streamkap.com or contact sales@streamkap.com