#VFD6 Preview: Cirba
Virtualizaion Field Day 6 Preview: Cirba
I’m thrilled to be invited to be a delegate at Virtualization Field Day 6. I’m looking forward to the opportunity to learn about some companies that I’ve never heard of, and learn more about some that I’ve seen before. To wet everyone’s whistle, I thought I that I’d throw together a few preview posts with what I’ve gathered about these companies so far, and then follow up after the event with what I’ve learned as a result of the deep dive. Today we’re going to be looking at the “Control Analytics for your Software-Defined Infrastructure” company Cirba.
Quick Company Background:
By utilizing a feature rich analytics engine coupled with a supply and demand based algorithm, Cirba is able to provide virtual machine workload placement recommendations, future capacity planning, and workload optimization for your virtual environments. I’m looking forward to Cirba’s upcoming presentation at Virtual Field Day 6. Here’s what I’ve gathered so far:
- The workload optimization bit appears to function as a rightsizing engine for your environment that can work in either a recommendation only or in an automated fashion.
- The placement recommendation bit takes into account newly placed workloads and existing workloads and optimizes the number and types of VMs on each hypervisor host in your environment so that CPU/Memory/Disk/Network are all as balanced as possible (hopefully allowing for greater consolidation)
- Workload Optimization/Placment functions taking into account your VM placement rules (think of these like VMware affinity/anti-affinity rules, these VMs can’t exist on the same host, these VMs must exist on the same host, these VMs have to run on this type of storage, etc…)
- These capabilities appear to be functional across multiple platforms: VMware, Hyper-V, KVM
My thoughts around usecases:
I’m seeing Cirba as a potential platform to help drive the addition of more economical hypervisors into the enterprise. Perhaps production virtualization workloads run on a trusted platform with years of organizational run-rate (VMware), while development environments could be moved to a more economical tier of Hyper-V because the organization is already paying for expensive Microsoft Datacenter licensing? By having a tool like Cirba, could I unify my capacity/performance management needs under one pane of glass? Could I leverage Cirba to give Hyper-V some additional smarts to free up budget for my production environment to improve other areas (automation, service catalog, new services)?
The capacity management niche in the cloud management space is a muddy area. If you run a VMware platform (vRA/vRO), you have the option to use the vRealize Operations suite for capacity trending and performance monitoring, but from a provisioning perspective in vRealize Automation the most leveraged way is by placing a reservation out there in your environment and reserving a dedicated amount of capacity. I could see an extension or a replacement of this reservation based policy as an vRealize Advanced Services Designer workflow call to Cirba’s API that could be leveraged to determine if the workload truly can fit based on a smarter algorithm than a reservation. I’d like to flesh out the API integrations that Cirba offers to see if this could work.
Dashboards are cool. Everyone likes feature-rich dashboard eye-candy. Does Cirba allow for me to leverage the dashboards that it creates for my environment outside of the normal control console so I could integrate this into a larger environment health dashboard?
What I’m hoping to learn:
- How are Cirba’s customers using to platform to provide placement decisions in automated environments?
- How is the “codification” of existing policies in today’s IT environments are being translated in Cirba environments today?
- I’m considering the possibility of leveraging Cirba’s API to provide placement recommendations for automated environment provisioning. I’d like to dive into the API capabilities of the platform.
- How are Cirba’s capabilities limited based on platform? If I’m using HyperV/KVM (more recently supported platforms), what features of the platform are stunted? How do these different platforms impact the placement algorithms underlying the placement decisions/requirements?
- How does Cirba differentiate itself from others in this crowded market? Where is the company winning?
- Are we really seeing value in the capacity planning space for the software? General observation – I’m finding most application deployments follow the deploy and then measure process. General recommendations from software manufacturers or application teams are good but oftentimes miss the end performance requirements of a deployment.
- How are customers leveraging the automated re-sizing/re-balancing options available in the platform?
What are you hoping to learn?
As a delegate for Virtualization Field Day, I’m honored to represent the tech community onsite with great companies like Cirba. If there are questions that you have for Cirba, or use-cases that you’d like considered, let me know by pinging me on Twitter or commenting below. I’m more than happy to work to get them answered/addressed during our session.