If it’s true that a picture is worth a thousand words, then nascent software company Uila has written a novel with their breakout software product for monitoring, and performing root cause analysis on, virtualized and physical network infrastructure, bridging the gap between both. Rarely does a breakthrough product come out of a startup on first try—usually there is much refining—but Uila may have done just that. The visualizations are nothing short of stunning, and the backing data is attained fast, and is highly accurate. Add cloud management and you really do have the makings of a great product.
Before we get to the product, however, in this case it is worth talking for a moment about the founders of Uila. Chia-Chee Kuan, Dean Au, and Miles Wu have, collectively, some 76 years of experience between them in the industry they’re trying to change; experience that comes in the form of patents, in the founding of companies like AirMagnet and Cinco Networks, and in the many years of R&D at companies from Cisco to Fluke. Oh, and let’s not overlook their credentials in the form of computer science degrees and master’s degrees from some of the best engineering schools across the world. Background and experience do not equal success, but certainly the weight of experience here lends itself to a high level of credibility right out of the gate, and begs at least a second glance.
What’s the Problem?
As data centers and infrastructure in general have become increasingly complicated, the risk of one thing breaking and causing the whole ball of yarn to come undone has increased exponentially, while the tools to analyze such problems with an eye towards root cause analysis and resolution have not matured. Vendors are still selling, and users are still using, point products designed to troubleshoot one particular aspect of a failure. What we should be doing is not isolating failure domains from a trouble-ticket point of view, but rather starting with a larger domain and shrinking from there. It’s a different methodology, but in today’s data centers it can be challenging to go with the traditional bottom up techniques we’ve all learned to love.
Don’t take that to mean that we shouldn’t use good, solid methods for isolation of failures, just that we should be doing so on subsets of the larger whole, with that larger whole firmly visible and watched as we work. Too often we toss flaming bags of shit over the wall to the <insert team here> and hope that they figure out the problem (validating that the problem was theirs to fix) before tossing it back. In this way, each team in succession, goes through the insular troubleshooting steps relevant to only their domain, with no interest in, or view of, the bigger whole. This is inefficient and slow, and a terrible way to solve any problem.
How Does Uila’s Product Work?
Uila’s product aims to bridge those inherent divides in troubleshooting and application visibility spheres by utilizing virtual smart taps (vST) in concert with more traditional (SNMP, SMI, SSH) means. Virtual taps get their information straight from the distributed virtual switch (DVS) in a virtualized environment, and utilize agents to grab the physical device data. All of this gets rolled up to the Uila Management & Analytic System, which is a cloud-based service handling the intelligence and analysis behind and of the data–think: Meraki for application visibility.
Uila’s product can perform deep packet inspection (DPI), auto-discover over 4000 applications, track application transactions and dependencies, and track network and TCP performance all while remaining distributed and agent-less. For a lot of systems and network operators the latter point is a big one. Having to install agents on a multitude of devices, many of which can’t actually host agents, can become unwieldy. An agent-less product like Uila’s allows for a quicker rollout overall, making for a quicker time to value for the business. And because of the various inflection points for data into the application, full stack visibility is more than just lip-service.
What Else Can It Do?
One of the challenges of modern software-defined networking, systems, cloud, etc., is application discovery mapping. Assuming that one of the primary reasons for moving to a new modality in network fabrics and automation is agility, understanding what the hell your applications are doing is of paramount importance. Understanding which ports are used by which processes, which backend databases talk to which front-end or middle-ware software services, and how all of this can be orchestrated and automated is not as easy as it might appear at first glance. Finding this information can be frustrating and error-prone, and very often dependencies get missed, leading to downtime, roll-backs, or missed milestones.
Uila’s software, due to it’s full-stack visibility, is actually a very good tool for analyzing a software stack, for performing application dependency mapping at a fast clip and with high accuracy and confidence. Having used a variety of tools to perform this analysis in the recent past, I can say that this tool is one of the best on the market. The visualizations are a nice touch, as during these exercises I have found that many application teams are surprised to learn the complete scope of what their stack is doing, reinforcing the fact that a tool like this is key to grabbing all pertinent information possible.
Conclusions, More Information, and Next Steps
In watching the presentation that led to this article, researching the product online, using the product myself, and talking with the founders in person, one thing has become abundantly clear to me: this application has a lot more capabilities and features than I can properly capture in a single sitting. Storage analytics, virtual machine (VM) cross-talk, latency and jitter, and myriad more options for troubleshooting are all covered through this tool, and the uses are limited only by the time you have to delve in and push buttons. I haven’t even brought up the automated root cause analysis capabilities, which in and of themselves warrant at least an article, if not a whitepaper.
If you have any interest in the product, I would suggest you take a look at Uila’s website (http://www.uila.com) and poke around a bit. They have an impressive list of customers already, as well as some whitepapers and other information available. They have a generous 30-day free trial which is fully featured and includes support and training, which goes a long way to getting people in and using the product in enough time to actually use the demo—something some more established industry players might want to take note of. You can also see some videos of Uila presenting their solutions to an audience of industry folks of varying backgrounds, by going to the Tech Field Day site at: http://techfieldday.com/companies/uila/.
At the end of the day, Uila may have written a novel, but the market will determine if it’s worth reading.