A managed service is a provider-operated offering that administers the underlying software and infrastructure. For Kafka, it shifts provisioning, monitoring, upgrades, broker management, and hardware maintenance from the customer to a cloud provider. section 1.6.2 section 11.0
The architectural choice is between operational control and reduced administration. On-premises deployment preserves control and can suit organizations with strong systems-administration capability; a managed service reduces setup effort and provides provider-backed service-level agreements. section 1.6.2
A managed Kafka service typically changes the following responsibilities and constraints:
- Reduces responsibility for hardware maintenance, infrastructure setup, monitoring, upgrades, and broker management. section 1.6.2 section 11.0
- Limits control over Kafka versions, broker-level tuning, cluster-management tools, and sometimes ecosystem components such as Schema Registry. section 1.6.2
- Supports scalable streaming architectures that integrate with cloud data sources, storage, databases, messaging systems, and machine-learning services. section 8.4.3
- Requires evaluating functionality, financial considerations, interoperability, and whether a cloud, on-premises, or hybrid approach fits the enterprise. section 1.6.3
Customer 360 deployment choice
A Customer 360 project can begin with a small, locally installed three-broker cluster when replication and low-level control matter. Choosing a managed service would simplify initial setup and conceal internal complexity, but reduce control over those deployment decisions. section 2.3.3
asyncapi: 3.0.0
info:
title: Customer 360 Project
version: 1.0.0
description: >
This project aims to aggregate data from various sources, providing a
comprehensive view of customer information.
channels:
customerProfile:
description: Storing profile information of customers.
address: CUSTOMER_PROFILES
bindings:
kafka:
partitions: 9 #1
replicas: 3
topicConfiguration:
min.insync.replicas: 2 #2
transactions:
description: Information about customer transactions.
address: TRANSACTIONS
bindings:
kafka:
partitions: 9
replicas: 3
topicConfiguration:
min.insync.replicas: 2
The configuration listing shows the alternative self-managed design: nine partitions, three replicas, and two minimum in-sync replicas for Customer 360 topics. section 2.3.3
