SATURN 2020

The Open Source Foundation for Application Security (OWASP) reports misconfiguration as one of the top 10 most critical web security risks. While various fault tolerance and recovery mechanisms are useful in handling hardware and software failures, they are less effective in handling system administrators' human errors. For example, the March 13th, 2019 Facebook outage affecting millions of users was caused by a server configuration error. In addition to reliability, configuration errors also can lead to security issues. In 2017, a configuration error of Amazon S3 storage exposed the personal information of 200 million users. As cloud infrastructure becomes increasingly more complex with integrated AI and advanced data processing, human errors have become one of the major causes of failures in cloud and Internet systems. In this talk, Yuanyuan Zhou will focus on a few current challenges of the human dimension of cloud computing and management. Due to legacy solutions and other reasons, most of today's data center system management requirements (in particular system configuration) do not follow the primary design principles of human-computer interaction (HCI) – namely simplicity, feedback, and consistency – making cloud management error prone for system admins.

Coming soon!

To paraphrase an old saying: “It takes a village to get insights from data.” Today data analysts, data scientists, and data engineers are hard at work delivering insights. These groups have a lot of data, a lot of tools, and a lot of talented people. So why are the majority of projects failing? They are failing by functioning in ways similar to how software teams operated ten years ago. Data and analytic team leaders need to innovate, yet struggle to keep up with customer requests and not let embarrassing data errors slip into production. The answer is to take ideas from Agile, DevOps, and Deming and apply them to the complex world of data analytics.

As Gene Kim wrote: “So much of the genesis of the DevOps movement was to address the huge problems of getting code into production. These days, there is an orthogonal problem of getting data from where it resides (often trapped in data warehouses) to where it can be used by teams to help their organizations win in the marketplace. I’m hoping that this book [The Phoenix Project] conclusively demonstrates to all that data is ultimately a software effort and can be massively helped by DevOps principles and patterns.”

In this presentation you’ll learn why the booming world of AI, data analytics, science, and insights is failing and how principles from Agile, DevOps, and Lean Manufacturing are the way forward.

Play Kahoot after a discussion of the day's events!

Join Paulo's kahoot and answer questions about software architecture on your device. Participants will need to open a browser, go to kahoot.it, enter the PIN number that will be shown on the screen, and enter their name/alias. Prizes (and bragging rights!) will be rewarded for 1st, 2nd, and 3rd place. Game on!