Presentations
Scylla is a new NoSQL database, capable of 2 million requests per second per node with Apache Cassandra compatibility. Scylla enables faster cluster scaling and more overhead to handle complex queries.
Tinc (http://www.tinc-vpn.org/) provides a secure mesh vpn for any number of hosts. My friends and I used this to build a network linking our homes, laptops, and various hosted machines. We started doing some cool things with it such as UPnP and NFS, things that would be impossible to do securely over the public internet. We then added Consul (https://consul.io/) which provides a service directory and health checks. In this talk I will cover the configuration of our VPN and the neat things that we do with it.
The question isn't "Can a blind person use a Linux machine?", it should be "How does a blind person use a computer running Linux?". The purpose of this presentation is to show several different applications and subsystems that have been modified in such a way to facilitate the use of a system without looking at the screen and use of keyboard input rather than a mouse.
What every application developer needs to know about when databases don't quite work right. Covers details of how Serializable transactions work in Postgres, various approaches to consistency in distributed systems and details of various anomalies and how to avoid them.
We had many VMs in AWS - were ingesting millions of metrics per second - and were having pain around service discovery and quick configuration changes. This is the story of how we integrated Consul into our environment, what it helped us with, mistakes we made and some tips for successful implementation in your own environment. 10 months later, our growing cluster was using Consul to facilitate 60 second cluster-wide configuration changes and make service discovery simpler and more flexible.
Server security is more important now than ever, yet many of the hardening guides out there read like they were written a decade ago (probably because they were!). Modern server hardening can be an obscure and complicated subject but it doesn't have to be. There are a number of simple steps you can take to dramatically increase the security of your infrastructure. In this talk Kyle will start with an overview of security best practices and provide a series of current yet simple hardening examples.
Unikernels offer an exciting opportunity to rethink kernel design choices and experiment with new low-level features that may affect the performance and security of applications in the cloud and ultimately change the way they are used. This talk is about my experience building Solo5, an open-source kernel library that runs directly on virtual hardware, at the lowest layer of a unikernel. The goal of Solo5 is to better understand the effect of the lowest layer of a unikernel on its behavior, bring MirageOS to more hypervisors, & also provide a platform for further experiment.
We developed and propose an alternative to /proc interface, called task_diag, which is based on netlink sockets and is way faster than traditional /proc. With /proc, one needs to open, read, and close many separate files, and perform a few conversions from binary to string representation and and back. With task_diag, one creates a request and receives a response in a single system call. This can be used by tools such as top and ps with considerable (up to 4x) speedup, as well as other tools that needs information about the processes running, such as perf and CRIU.
Stateless architectures are all the rage right now, but is that a good thing? Stateful architectures in general are just starting to take off, and here at VictrorOps, we’ve been using Akka to build several stateful projects and we’ve learned a lot along the way. This presentations will introduce the concept of stateful architectures using Akka and how to start building application using this stateful paradigm.
Building stateless http servers has been the standard architecture for SaaS companies for the last ten years. Requirements for continuous uptime, scalability and low cost have pushed companies such as Facebook, Twitter, Typesafe and VictorOps back towards architectures where processing and data co-exist in the same memory space. In this talk we will explore what you need in order to build a stateful system and the benefits and tradeoffs of the transition.