2015 · 26 citations · 16 references
ComputingComputer ScienceComputer SystemsComputer NetworksDistributed SystemsFault-tolerant SystemsDistributed ComputingDistributed Systems ArchitectureDistributed Systems PerformanceFault-tolerant CommunicationReliable Distributed SystemsReliable Message Delivery
Nowadays, with the rapid development of distributed computing, message distribution has been more and more important to distributed systems. Kafka was developed to collect and distribute massive messages for distributed systems. It is a high-throughput distributed messaging system. Kafka can always keep stable performance even if it processes millions of messages per second. Messages are persistent in Kafka, and Kafka is able to scale out easily. Besides, it is suitable for both online and offline processing simultaneously. Existing messaging systems all have some disadvantages. In comparison with these messaging systems, Kafka has superior performance. Therefore, Kafka has already been successfully used in production by some companies for a certain time. Though Kafka is still able to be improved in some aspects, it's expected to have broad development prospects.
16
Streaming Big Data Processing in Datacenter Clouds
Rajiv Ranjan · IEEE Cloud Computing · 2014 · 233 citations
Stream Bench: Towards Benchmarking Modern Distributed Stream Computing Frameworks
Data Infrastructure at LinkedIn
Aditya Auradkar, Chavdar Botev, Shirshanka Das et al. · 2012 · 76 citations