Stacktical, a scalability prediction platform

Hacker News

Share card

Stacktical, a scalability prediction platform

Hi guys, we're Wilhem from Paris and Jean-Daniel from Tokyo, software engineers with a passion for all things cloud (IaaS, PaaS, SaaS). We recently decided to tackle the problem of Capacity Planning with Stacktical, a Scalability Prediction service (https://stacktical.com). For a decade, we've been observing our clients and colleagues trying to nail down their strategy using repeated cycles of defining, collecting and interpreting load testing campaigns. It's funny how most people don't realize how demanding the work of infrastructure managers and their teams really is... While we've made great progress in the field of Continuous Delivery (we're huge Docker fans for instance), getting actionable scalability metrics remains a daunting process that is far from being streamlined. It has a ginormous operational cost that defies one of the main purposes of scalability: the ability for a company to fit the increasing and decreasing computing needs of apps so accurately that it never under or overspend its hosting budget. We think we can do a better job by leveraging predicting technologies to do most of the heavy lifting involved in answering the question "Will my application scale?". To do just that, Stacktical can generate a Scalability report with tons of useful insights within seconds, using just a dozen load testing metrics. For example, we can tell you how many max concurrent users your application infrastructure can handle before starting generating bad performance (or being down altogether!) due to the fundamental penalties any server is subject to (serialization, bad inter-processus communication, suboptimal caching etc). Our prototype is up and running at https://stacktical.com and we're looking for people willing to try out our solution and help us getting ready for primetime. Can you guys give us a hand on that regard? You can signup for our private beta at https://stacktical.com/signup Thanks a tons, Cheers, Wilhem and Jean-Daniel