Service and Support
Amazon MSK customer support is generally well-received, with varying experiences. Users find the assistance helpful, though its effectiveness often depends on the support plan. Some note declining knowledgeability over time, yet others praise the prompt service and technical knowledge. While some rate the support highly, others feel improvement is needed. Despite not universally engaged, when assistance is sought, it is mostly reliable and provides guidance on unresolved issues, influencing varied satisfaction levels.
Deployment
Amazon MSK's initial setup is perceived as straightforward by those familiar with Kafka, while newcomers find it complex due to limited documentation. Deployment usually requires collaboration with multiple teams, yet it can be completed in less than ten minutes for those experienced. Some setups took several weeks due to networking configurations. Familiarity with AWS eases the process. Users appreciate its ease of use and integration with AWS, though they recommend evaluating its necessity based on specific needs.
Scalability
Amazon MSK exhibits strong scalability, effectively handling increased usage with minimal concerns. While the system requires manual scaling, it remains efficient for many despite lacking autoscaling. Users report variable experiences, with some encountering minor operational challenges. The cluster-based architecture supports easy expansion by adding nodes, yet operational overhead may arise when managing the infrastructure. Additionally, serverless options offer ease, while classic versions necessitate manual adjustments for increased processes.
Stability
Users generally consider Amazon MSK stable, with no significant downtimes or crashes reported. Any issues experienced were linked to integration with other services, like Debezium, rather than Amazon MSK itself. Stability ratings hover around eight or nine out of ten. A few concerns were noted with older server versions, but those were resolved. Amazon MSK boasts 99.9% availability, demonstrating reliable performance during high-load testing scenarios.