I am a big fan of programatic interfaces and have written extensively about it while working for Nutanix, EMC or VMware. Programmatic interfaces allow administrators to create custom automation workflows using scripting languages or workflow engines like vCenter Orchestrator, vCAC, Puppet, Chef, BMC and others. This is the ultimate goal for a true Software-Defined Datacenter, enabling a powerful application ecosystem that make automated decisions about the use of infrastructure resources in a simple, efficient and easy manner.
When I provided LiquidwareLabs with a space for a sponsored article I could not be happier to learn that they are now fully embracing programatic interfaces within their platform.
Byline: Kevin Cooke, Product Director, Stratusphere Solutions for Liquidware Labs
Multi-platform and heterogeneous virtual environments can be tricky to manage. Responsibility is often divided, and the silos are very much interdependent. For example, different IT groups may own the delivered application or service, the virtualization layer, and the supporting hardware; and lastly, the operations group, who are accountable to meeting SLAs and keeping things running smoothly. Knowing where constraints may be present or staying ahead of the curve can be a difficult process.
Consider desktop virtualization for a moment. You have the hosts and storage infrastructure, which may be managed by your server and storage team. There’s the hypervisor and allocation of virtual resources, possibly managed by a dedicated virtualization team. Let’s not forget the network team, who provide the pipes, supporting directory services and remote connectivity. Out on the fringe are the folks that manage data center operations, and of course, there are the magicians who provide the desktop VMs, application delivery and image management. And we certainly cannot forget the business units and end users, who expect nothing less than user experience nirvana.
Complicating matters somewhat, each of the above groups likely has adopted their own tools and workflow to best support each area of purview. And while the overall approach itself is not problematic, it does open cracks between the groups that can pose problems when users complain about poor experience.
“Is it the image?” “The hardware?” “The network … who knows?”
“Storage is showing some performance challenges, but is the issue really due to the data store or is it a lack of IOPs?”
Pressure is mounting. Management is requiring a status report, the storage team is under fire and no one has thought to look at the rogue application processes wreaking havoc on vCPU queueing and overflowing vRAM paging.
Of course, having everyone use the same monitoring solution can help to avoid some of the challenges in this scenario, but the reality is most of the discrete IT teams in the thick of this mess already have an existing tool and workflow to support their troubleshooting approach. So how do you preserve these existing methodologies and approaches, while providing information to the most relevant IT teams in a way that supports their approach and choice of tool?
Enter the Stratusphere API and Shared Visibility
Stratusphere UX from Liquidware Labs, is a solution that supports the functions of monitoring, performance validation and diagnosing the complexities highlighted in the very common real-world example outlined above. The solution—which includes the Stratusphere Hub, Database, Network Station and Connector ID Key—gathers extremely detailed metrics and information about all aspects of the virtual desktop user experience.
In a 1,000 or so desktop environment, Stratusphere UX will gather a few million data points per hour. Details about the machines, users, applications, network (pipes and services), as well as the contributing infrastructure will be captured and stored in the Stratusphere Database. This information can be correlated to constraining events and tied to a composite metric that quantifies the user experience. Better still, this information is exposed and easily accessible through the Stratusphere application programming interface (API).
In the above VDI downtime event—and with the visibility provided via the Stratusphere API—relevant metrics and information about the performance, user experience and state of other areas of the architecture can be flowed from Stratusphere UX to the native tools each IT group has previously chosen to gain visibility of their little corner of the overall architecture. For example, the JSON-outputted metrics and information can be sent to HP Operations Manager to assist the server, storage and networking groups, while BMC Remedy receives a helpdesk ticket to alert operations of the groups, users and applications affected.
The power in this approach is preserving workflow. IT groups are able to work within their existing monitoring tools and can leverage proven methods to troubleshoot and determine how their component of the architecture may be contributing to the poor performance. Further, the use of the Stratusphere API greatly facilitates inter-group trouble ticketing as a defined composite metric like the VDI User Experience can be leveraged to baseline and correlate activities across functional IT groups.
Stratusphere UX was designed to provide visibility into complex multi-platform and heterogeneous virtual environments. Its user-centric approach to monitoring, performance validation and diagnostic capabilities take much of the complexity out of next-generation desktop workspaces. And with the Stratusphere API, IT groups are able to support these virtual environments in a way that meets business goals, minimizes risk and supports both the organizational and IT changes ahead.
This article was first published by Andre Leibovici (@andreleibovici) at myvirtualcloud.net.