Recognize when to use PubSub+, Apache Kafka, and the Confluent Platform

Confluent adds key features to Apache Kafka to address Kafka’s shortcomings in certain use cases. Keeping that in mind, here’s a high-level look at how each performs:

Legend: Great Good Other

Legend:
Great Good Other

KafkaConfluentSolace
IoT
  • Not suitable for large scale 
  • Needs to be fronted by other MQTT brokers
  • Good sensor data consumption
  • Great sensor data consumption and command and control
Multi-Cloud
  • Static routing between clouds
  • Poor reliability 
  • Complex remote client setup
  • Static routing between clouds
  • Complex remote client setup
  • Good IaaS services  integration
  • Poor throughput versus reliable performance
  • Great multi-cloud dynamic event routing
  • No remote client setup required
  • Great IaaS services integration
  • Good throughput versus reliable performance
Microservices
  • Limited exchange patterns 
  • Java and other community driven languages
  • Many uses not truly suited for Kafka batching
  • Limited exchange patterns
  • Good language support
  • Many uses not truly suited for Kafka batching
  • Great breadth of  languages, protocols, message exchange patterns
  • Messaging and streaming
  • Advanced event mesh
Digital 
Transformation
  • Good CDC tools
  • Less flexible multi-site hybrid cloud solution
  • Good CDC tools
  • Great when thinking of bi-model development in combination with hybrid cloud
Stream Processing
  • Core product use case
  • Great core product use case
  • KSQL defines well-used patterns
  • Messaging and streaming with good 3rd party stream processor APIs integrated
Low Latency
  • Batch architecture limits performance
  • Kafka batch architecture limits applicability
  • Great core product use case
Analytics
  • Great core product use case
  • Great core product use case
  • Nothing added
  • Good when thinking of bi-model development in combination with hybrid cloud
Transactional
Workflows
  • No individual message acknowledgement

  • Can use Kstreams and transactions which adds complexity but you're then locked-in to Confluent license agreement for use as a service (I.e., Amazon/Azure can't do it)
  • Individual message ack out of the box (i.e., Need to act upon a ticket purchase)

See How PubSub+ and Kafka Differ

Wondering how Solace compares to the capabilities Confluent adds to Kafka?

Here’s our take:

ConfluentSolace
Schema Registry
  • Supports compact wire formats
  • No lifecycle management or understanding of app interactions/dependencies
  • PubSub+ Event Portal enables design, discovery and management of events across brokers and environments
  • Does not have a runtime schema registry
Connectors
  • Many connectors do not provide source code
  • Must be certified to be supported by Confluent
  • Broker integrated connector or source/sink connectors as source code
  • Native transparent protocol and transport translations
Clients
  • Supports other clients based on their C library for C++, Python and Go.
  • Other language or wire format is only supported via connectors
  • Supports many client languages and transparent interoperability with multiple open source terminated transports and wire formats
REST Proxy
  • Allows retrieval of broker metadata and limited consumer/producer proxying between REST and native Kafka producers and consumers
  • Supports REST as a first-class producer and consumer
  • Full request/reply MEP
MQTT Proxy
  • Terminates MQTT producers so persistent MQTT producer messages can be stored in the Kafka topic
  • Can’t send messages to MQTT client through this proxy

  • Natively supports full MQTT broker for producers and consumers
  • MQTT persistent messages are stored in the same Solace store
  • Any other Solace supported API or protocol can easily access MQTT messages
KSQL
  • Proprietary SQL interface engine to Kafka streaming so consumers can get messages from topic and process SQL statement to move message to another topic
  • Many of KSQL’s capabilities are natively available via topic hierarchies, topic-to-queue bridging and SQL selector statements
Replicator
  • Simple topic replicator software that allows primitive and limited DR and hybrid cloud capabilities for integration with Confluent Cloud and/or multiple broker clusters
  • Slightly more robust than Apache MirrorMaker
  • Native enterprise-grade DR
  • Specialized routing protocols intelligently move data in a hybrid cloud environment
Cluster Linking
  • More advanced topic replicator to facilitate multi-datacenter, multi-region, and hybrid cloud deployments
  • It provides data distribution and HA/DR capabilities between multiple clusters
  • Solace solves multi-cloud data distribution using sophisticated dynamic bridges between broker clusters
  • Native enterprise-grade DR
  • These capabilities are available as standard offering, without additional components
Available Clouds
  • Available in most regions of Amazon, Google, and Azure, with additional regions available upon request
  • Available in most regions of Amazon, Google, and Azure, Huawei and Alibaba, with additional regions available upon request

And managing and governing your events – that’s an important capability to dive into.

Confluent Stream Governance and PubSub+ Event Portal solve different problems. 
Do you have multiple types of brokers? Do you need a tool to design, expose Event API products and operate your EDA? Then Event Portal is for you! 

confluent stream

Operations

  • Confluent Kafka brokers only
  • Data/schema management and Catalog
  • Schema Validation – Confluent Cloud/Platform
  • Operating data flows – Confluent Cloud ONLY
  • Tracking lineage – Confluent Cloud ONLY

Design to Operations

  • Supports all Kafka Distributions (Confluent, Apache, MSK) and messaging systems(Solace) – Mix and Match!
  • Event discovery to understand your existing event-driven assets (Schemas, Events, Applications)
  • Event Dependency Visualization at Design and Runtime
  • Lifecycle Management with versioning and promotions
  • Exposure of Event API Products for sharing across LOB and Orgs
  • Coming Soon: Audit your Design vs what was built/deployed, be alerted to design deviations

Discover, Visualize and Catalog
Your Kafka Event Streams

You know and love Apache Kafka, but have you ever tried to visualize Kafka topology, or figure out who owns what event stream in a Kafka cluster? Your event-driven architecture has evolved, and your system has grown to the point where you’re feeling a bit… out of control.

That is why we have developed PubSub+ Event Portal. This event management toolset makes it easy for you to discover, visualize, catalog and share your Apache Kafka event streams, including those from Confluent and Amazon MSK.

About Kafka Event Streams

Still have questions about how Solace PubSub+ compares to the Confluent Platform?

Get in Touch