Components and Services within the Kubernetes Cluster

First Time User? Free Credits on us!

Get up to $200 worth of Free credits when you sign up for the first time. Use this link to create a new account on Condense and claim your free credits!

https://www.zeliot.in/try-now

The heart of Condense lies within the Kubernetes cluster, where various services collaborate to process data:

Nginx

This versatile web server acts as the network traffic controller for Condense.

• Manages API access for user interaction with Condense.

• Handles Communication Protocols like TCP/IP and MQTT(s).

• Configures ingress resources and port mappings for API access and device data ingestion.

• Facilitates TLS termination for secure communication.

Redis

Condense leverages a Redis cluster for data caching and performance optimization.

• Stores metering data for various dimensions used in consumption-based billing.

• Manages authentication tokens used by Condense services for API interactions.

• Acts as a checkpointing mechanism for user-configured pipelines, storing sequence information, metadata, and real-time data like throughput and status.

• Stores asset ping information displayed on the Condense dashboard.

Also read our article on Redis - Harnessing the Power of In-Memory Data Storage here: https://www.zeliot.in/blog/redis-harnessing-the-power-of-in-memory-data-storage

Kafka

Condense employs Kafka as a message queue, serving as the backbone for data communication between services after transformations.

• Kafka brokers and Zookeeper run within the cluster, ensuring fault tolerance and data integrity.

• Operates as a one-way queue, facilitating data progression through various transformation stages.

• Event-driven auto scalers automatically adjust consumer group offsets to maintain optimal performance and prevent lag.

• Kafka version upgrades are managed by Condense with user consent for seamless adoption of new features and security patches.

Condense inherently deploys with a Fully - Managed Kafka

Condense provides a fully managed Kafka service within its Fully Managed BYOC model, ensuring seamless data streaming, high availability, and zero infrastructure complexity. By embedding Kafka into its industry-specific verticalized streaming ecosystem, Condense enables real-time event processing, data communication between services, and scalable pipeline orchestration—all within the customer's cloud environment.

With Condense fully managed Kafka, enterprises get enterprise-grade performance, dynamic scaling, and automated operational management, eliminating the need for manual intervention in cluster provisioning, maintenance, scaling, and security updates.

Kafka as the Messaging Backbone

Event-Driven Data Streaming

Condense uses Kafka as a high-throughput event streaming system, ensuring fault-tolerant, distributed message queuing between microservices and streaming applications.

One-Way Queue Model

Kafka operates as a one-way queue, moving event data seamlessly through various prebuilt transformations, custom transformations, and real-time processing stages before reaching downstream applications.

Decoupled Service Communication

Kafka ensures service independence, meaning applications can produce and consume data asynchronously without blocking.

High-Performance Streaming

Optimized for millions of events per second, ensuring low-latency streaming and high-throughput data processing.

Cluster Management: Kafka Brokers and Zookeeper

Fully Managed Kafka Cluster

Kafka brokers and Zookeeper run within customer-controlled cloud environments, managed by Condense, ensuring high availability and resilience.

Fault-Tolerant Architecture

Multi-cluster redundancy and automatic replication prevent data loss and ensure continuous streaming even in the event of node failures.

Data Integrity & Consistency

Kafka maintains strong replication and partitioning strategies, ensuring data consistency across distributed environments.

Seamless Kafka Version Upgrades

Condense handles Kafka version upgrades with user consent, allowing smooth adoption of new features and security patches without disrupting operations.

Automated Scaling with Event-Driven Auto-Scalers

Consumer Group Auto-Balancing

Kafka dynamically scales consumer groups based on event throughput, preventing backlogs and maintaining near-real-time processing.

Adaptive Resource Scaling

Event-driven auto scalers optimize cluster performance by dynamically adjusting consumer group offsets, ensuring Kafka maintains low-latency event processing at scale.

Intelligent Partition Rebalancing

Kafka automatically redistributes partitions among consumers during scaling, eliminating bottlenecks.

Reliability and High Availability (99.95% Uptime SLA)

Multi-Zone Deployment

Kafka clusters are deployed across multiple availability zones within customer cloud environments, preventing downtime due to infrastructure failures.

Automatic Failover & Leader Election

If a Kafka broker fails, Zookeeper automatically elects a new leader, ensuring data availability and seamless failover.

Log Retention & Durability

Kafka persists event logs based on configurable retention policies, preventing data loss and ensuring replayability.

Security, Compliance, and Observability

Data Encryption

Condense ensures end-to-end encryption for Kafka data in transit (TLS) and at rest (AES-256).

Role-Based Access Control (RBAC)

Fine-grained permission management ensures that only authorized applications and users can interact with Kafka topics.

Audit Logging & Monitoring

Built-in observability tools provide real-time monitoring of Kafka performance, including throughput, latency, and partition distribution.

Enterprise Compliance

Kafka operates within customer-controlled BYOC environments, ensuring GDPR, HIPAA, and SOC 2 compliance.