IBM Watson IoT Platform: Message Gateway and MessageSight are nearing end-of-support - What next?
IBM Watson IoT Platform – Message Gateway and MessageSight are going end-of-support in a little over 4 months. Are you all set with an alternative?
Recently, a large electronics company with 600,000 monitored devices deployed in the field reached out to us. They were worried about what their world will look like after September 30th, 2022 when support for IBM IoT MessageSight and the IBM Watson Message Gateway will be discontinued.
Since many businesses are working their way out of this situation, this blog post explains the perfect alternative: HiveMQ. We surmise your priority to make sure HiveMQ (at a minimum) has feature parity with what your current (and soon to be EOS solution) can do.
We reviewed publicly available documentation and created a summary list of key features that an enterprise is likely to use.
HiveMQ MQTT Platform Vs. IBM Watson IoT Platform
IoT Message Gateway/MessageSight 5.0 features both publish and subscribe (topic-based) and point-to-point (queue-based) messaging domains. Whereas, HiveMQ is a PubSub-architecture MQTT broker that can extend to work with message queues and ESBs. You can use our pre-built extensions or build custom extensions with our enterprise SDK. HiveMQ professional services team can review and approve your custom extension to assure performance.
IoT Message Gateway/MessageSight 5.0 features MQTT versions 3.0, 3.1, and 5.0. HiveMQ too, including 3.1.1.
IoT Message Gateway/MessageSight 5.0 features both persistent and nonpersistent qualities of service. HiveMQ too.
IoT Message Gateway/MessageSight 5.0 features MQTT over HTML5 WebSockets. HIveMQ too.
IoT Message Gateway/MessageSight 5.0 features Java Messaging Service (JMS) 1.1. HiveMQ is an MQTT broker and an MQTT platform.
IoT Message Gateway/MessageSight 5.0 features Message expiry. HIveMQ too.
IoT Message Gateway/MessageSight 5.0 features Message discard. HiveMQ too.
IoT Message Gateway/MessageSight 5.0 features MQTT shared subscriptions. HiveMQ too.
IoT Message Gateway/MessageSight 5.0 features clusters for scalability. Whereas, HiveMQ uses a masterless cluster architecture that provides scalability and high availability. Adding new nodes increases the system throughput to multiple hundreds of thousands of messages per second. The HA approach with internal data replication enables you to perform fail-over in case of hardware or software failures without message loss. Furthermore, this also enables you to perform rolling updates of a cluster with no downtime and message loss.
IoT Message Gateway/MessageSight 5.0 features Restful administration API. Whereas, HiveMQ gives the possibility to manage key aspects of the broker (e.g. client connections) both via our portal named Control Center and via REST APIs for a programmatic approach.
IoT Message Gateway/MessageSight 5.0 features developer-friendly APIs and libraries, such as
MQTT clients and libraries for a variety of platforms, which include C and Java-based APIs, Libraries for Google Android and Apple iOS, JMS client libraries, JavaScript API for HTML5-based applications. Whereas, HiveMQ supports MQTT and MQTT is standardized on the protocol level and therefore MQTT standard-compliant brokers like HiveMQ work flawlessly with all standard-compliant MQTT libraries provided in many programming languages including Java, C, C++, C# and many more. HiveMQ provides an extensive overview of all available libraries.
IoT Message Gateway/MessageSight 5.0 features assured delivery of messages, has support for MQTT qualities of service (QoS) 0, 1, and 2 (best effort, at least once, and exactly once), has local transaction support, supports for high-availability configurations, has SNMP monitoring and supports for log file rotation, purging, and offloading – all for reliability. Whereas HiveMQ ensures the delivery of messages and also supports all QoS levels defined by the MQTT specification (0,1 and 2). It’s crucial to note that many ESBs and hyperscalers do not comply with the QoS values defined by the specification and can not support features like “exactly once” delivery. The risk in using such a product is that some of your applications will have to be re-engineered to parse for duplicates. Also, HiveMQ provides out-of-the-box integrations into Prometheus, InfluxDB, and Splunk, and JMX-based integrations into other Observability tools such as Datadog. The HiveMQ logging system provides support for log rotation, purging and Syslog integration capabilities for offloading.
Security and Integration Feature Comparison of HiveMQ MQTT Platform and IBM Watson IoT Platform
Security
IoT Message Gateway/MessageSight 5.0 features:
Authentication through local user store or external Lightweight Directory Access Protocol (LDAP) servers.
Secure Sockets Layer (SSL) and Transport Level Security (TLS) support (SSL v3 and TLS 1.0, 1.1, and 1.2).
Fine-grained messaging authorization policies, which restrict access that is based on combinations:
User of group
Client identifier
Protocol
Network interface
Listening address or port
Client IP address or range
Destination (topic and queue) name
Black list with client revocation lists
OAuth 2.0 support designed to securely delegate authorization to a remote server. OAuth authorization servers that can be used with MessageSight include the DataPower appliances that contain an OAuth authorization server feature.
Single sign-on support by using LTPA tokens within cookies in the HTTP header when connecting with WebSockets.
HiveMQ delivers:
Authentication via Username and password with support for Lightweight Directory Access Protocol (LDAP) servers and databases as user data sources. Using the Enterprise Security Extension, you can create custom authentication logic that lets you chain multiple modes of authentication.
In addition, HiveMQ offers overload protection and throttling mechanisms to provide robustness in case of large (re)connect scenarios.Secure Sockets Layer (SSL) and Transport Level Security (TLS) support (SSL v3 and TLS 1.0, 1.1, 1.2 and 1.3).
Fine-grained messaging authorization policies on all MQTT interactions, which restrict access that is based on combinations:
User of group
Client identifier
Connect information
Network interface
Listening address or port
Topic filter including wildcards and dynamic variables
Whitelist approach
HiveMQ also supports OCSP Stapling for certificate lifecycle management to avoid overwhelming your OCSP responder in case of frequent or mass re-connections.
OAuth 2.0 support to delegate authentication to a remote server, securely.
Web Portal Single sign-on support via OAuth 2.0.
Integration
IoT Message Gateway/MessageSight 5.0:
Extends and connects to IBM MQ infrastructures.
Supports mapping MessageSight destinations to and from IBM MQ destinations.
Supports connectivity to WebSphere Message Broker via JMS or MQTT nodes or both.
Integrates with Java environments
Integrates with rich HTML5-based web applications.
Allows development of interactive mobile messaging applications.
Is extendable with a protocol
HiveMQ:
Extends and connects Kafka Protocol based infrastructures such as Apache Kafka, Confluent Platform or Cloud, and Azure Event Hub.
Supports message bridging (bidirectional message forwarding) to other MQTT 5 compliant brokers
Integrates with most programming languages via MQTT open protocol standard and MQTT open libraries.
Integrates with rich HTML5-based web applications.
Allows development of interactive mobile messaging applications.
Is extendable with an Extension SDK.plug-in feature.
IoT Message Gateway/MessageSight 5.0:
Allows complete configuration and operation within 30 minutes.
Delivers easy-to-use, web-based command-line interfaces.
Requires minimal configuration options
Is available to support nonproduction, idle standby, and production environments.
Supports Linux RPMs and can be installed on both Red Hat and CentOS Linux
Can be deployed as a Docker image, which improves portability and enables automation across your software delivery pipeline
Can be deployed within VMware ESXi on either a Red Hat or a CentOS base image
Can be deployed on public cloud infrastructure in IBM Cloud, Amazon, and Azure cloud environments
HiveMQ features Deployment ease:
We don't offer RPMs for HiveMQ however HiveMQ can both be installed on Red Hat and CentOS Linux.
HiveMQ offers an easy-to-use k8s Operator including helm deployment chart to simplify the operations of HiveMQ in k8s environments such as AWS EKS or Azure AKS.
HiveMQ software can be deployed on AWS, Azure and Google Cloud Platform cloud environments.
In Conclusion
For enterprises looking to migrate away from the IBM IoT MessageSight / IBM Watson (Message Gateway) solution, HiveMQ is a straightforward choice. While your priority at the moment is to find a system that offers parity and ease of migration, this is also an excellent opportunity to evaluate products which can adapt to your future needs. In addition, a cloud-native solution that doesn’t lock you in gives you the ability to be agile. HiveMQ offers an easy immediate solution and will be a long-term partner for your IoT data movement requirements.
We are ready and eager to discuss how you can evaluate and implement HiveMQ as a replacement for IBM IoT MessageSight. Reach out to discuss how we can help.
Gaurav Suman
Gaurav Suman, Director of Product Marketing at HiveMQ, has over a decade of experience in roles like Solutions Architect and Business Development Manager. His journey includes launching market-first products and achieving a 2X revenue increase in the past year. Eager to connect with industry peers, Gaurav pushes the boundaries of what Product Marketing can achieve for businesses.