By: Megan Oster
While mainframe users will continue to rely on existing architecture and networks as the presence of cloud computing grows, infrastructure as a service (IaaS), platform as a service (PaaS) and software as a service (SaaS) will become a major piece of the puzzle – connecting mainframe applications to new cloud services as well as existing infrastructures. Speakers Chris Leonard and Mayur Raja of IBM UK Laboratories Limited addressed the impact of this convergence on MQ networks during their session on the topic at SHARE in Orlando 2015.
First and foremost, there is the issue of whether or not a traditional MQ network provides value in the cloud era. Part of MQ’s advantage in this matter is its continued reign as the market leader in enterprise messaging. This strength is due in large part to its design, which allows it to provide flexible, scalable solutions. MQ offers a diverse platform, API coverage and integration with other enterprise products such as application servers and databases. Taking all of this information into account, it is safe to say that MQ networks are a decent match for cloud and most issues should be minor.
One of the main challenges cloud presents is the effective management of interconnected devices. Several years ago, the industry recognized that users would need a common protocol to handle each new iteration of embedded devices that would inevitably come into existence. MQ Telemetry Transport (MQTT) was born in response to this need. MQTT has proven a viable solution due to a number of critical capabilities built into its design:
- Expects and caters for frequent network disruption – it is built for low bandwidth, high latency, unreliable, high cost networks.
- Expects that client applications may have very limited resources available.
- Designed with publish/subscribe messaging paradigm as required by the majority of SCADA and sensor applications.
- Provides traditional messaging qualities of service where the environment allows.
- Offers OASIS standard for ease of adoption by device vendors and third-party client software.
Provisioning messaging systems is another important consideration for MQ networks and cloud computing. One approach that has demonstrated success involves using MQ Hypervisor Editions in tandem with IBM Workload Deployer/PureApp. Benefits include:
- Standardization of software images reduces risk and uncertainty.
- Automated provisioning reduces errors and speeds time to value.
- Repeatable configuration across sets of machines is quicker and less prone to errors.
- Applying software maintenance is simpler and faster when using IWD/IPAS GUI or CLI.
- A comprehensive history/audit is maintained.
- License tracking is integrated.
In conclusion, far from limiting their reach, cloud could actually broaden the capabilities and impact of MQ networks. Ensuring the latter happens involves careful selection and application of solutions.