Since I have been onboard with Tintri for a while now, and after spending time in the field with customers, I thought I would focus on one of the biggest advantages Tintri customers love. The founders of our company wanted to solve a specific pain point. Virtualization and traditional storage arrays do not work well together and VMware environments need visibility, simplicity and performance. General purpose storage is difficult to manage when it comes to VMware environments because, well, it’s not designed specifically for virtualization! Many customers that I worked with at VMware and now with Tintri, strive for a ~100% virtualized environments. The reason behind this is two fold. Virtualization brings agility and simplicity to the data center engineers and drives CAPEX/OPEX costs down for the C level folks. That’s hard to ignore when you can bring two camps together (aka a compelling event). Being purpose built for virtualization will help drive your organizations efforts towards a denser, more optimized, virtualized datacenter.
What is purpose built and why do I care? I chose the image above to help illustrate this blog post. Being purpose built has great advantages over something that is not purpose built. Consider if the woman above had a general purpose, metal clunky row boat rather than a sleek, fast, transparent kayak that was specifically designed for her journey. Yes the metal row boat can get her from point A to point B but it will be hard to navigate and she won’t be able to see what is coming. With her transparent Kayak she can now see the dangerous coral below her boat, navigate faster and easier, and also take in the sites of the ocean life below her. Tintri brings these same characteristics to your virtualized workloads by showing you the performance metrics across the VMware stack when dangerous conditions are occurring.
Top down vs Bottoms up
Most storage solutions take a top down approach towards management of the underlying infrastructure. They manage constructs such as LUNS or volumes. Some of you I have spoken with at local VMware user conferences have seen the Tintri “LUNS Suck” bumper sticker we have floating around. It’s true, they do suck, and who wants to spend cycles managing a spreadsheet of logical unit numbers?
Tintri’s management construct is the virtual machine. Makes sense when you are trying to virtualize your datacenter as aggressively as possible. We take a bottom up approach since we start at the virtual machine itself. In the image below you are looking at the Tintri user interface that shows a virtual machine not performing well. You can quickly see across all of the infrastructure that touches this VM. Notice the 4 metrics called out which are Host, Network, Disk and Storage. You can see in this example that this virtual machine is having CPU contention on the physical ESX host that he happens to be living on. This is powerful stuff! This specific metric has nothing to do with storage at all, but is certainly important when troubleshooting performance issues in your VMware environment. This granularity is something you can bring up in 3 mouse clicks rather than sifting through layers of irrelevant management constructs. By the way, this visibility is not an additional license, it is included with the Tintri storage array.
Tintri solves VMware storage challenges
Virtual sever workloads are not the same as traditional physical workloads. Performance needs to be addressed as you scale out and manage your VMware environment. Visibility across the entire virtualized stack needs to be present so you can virtualize with confidence and success. End users come to IT for services that they expect to be top notch so they can drive the organization forward and be successful. You need the ability to satisfy these requests with a storage solution that is purpose built for your environment. Tintri can help you be successful.