Home MENU

Schedule

16 V - TUESDAY
17 V - WEDNESDAY
Track 1
Track 2
09:00 - 09:40
REGISTRATION STARTS
09:40 - 10:00
OPENING CEREMONY
10:00 - 10:40
Building Better FLOSS Community Relationships @ FB
Davide Cavalca, Marcin Sawicki Building Better FLOSS Community Relationships @ FB

This talk will cover the work Facebook has done to become more involved with various upstream open source communities. We will start with why we believe it's important for companies to build strong relationships with the communities around the  software they use - particularly in infrastructure. Next we will look at the steps we took to become better community citizens and finally we will discuss some case studies. Specifics covered will include various projects we've contributed to, technical work such as back-porting various OS components from Rawhide to CentOS 7, benefits we've received and lessons learned.

11:00 - 11:40
Superkomputer CIŚ - przykłady wykorzystania i administracji
Michal Wojcik Superkomputer CIŚ - przykłady wykorzystania i administracji

Narodowe Centrum Badań Jądrowych to jeden z największych instytutów naukowych w Polsce, dysponujący m.in. jedynym w Polsce jądrowym reaktorem badawczym Maria. Ponad 1000 fizyków, inżynierów i pracowników pomocniczych stale zajmuje się badaniami podstawowymi i stosowanymi z dziedziny fizyki subatomowej (fizyka cząstek elementarnych i jądrowa, fizyka plazmy gorącej itp.) oraz fizyką biomedyczną i rozwijaniem technologii jądrowych.
W roku 2013 na terenie NCBJ w Centrum Informatycznym Świerk rozpoczęła się budowa superkomputera o mocy 1PTFLOPS.
Główne zadania CIŚ, to:
    * wsparcie obliczeniowe Polskiej energetyki jądrowej i konwencjonalnej,
    * rozwój i udostępnianie zaawansowanych usług przetwarzania danych,
    * symulacje i monitoring zagrożeń radiologicznych,
    * udostępnianie zasobów i usług naukowcom,
    * prowadzenie badań naukowych i rozwojowych

W swoim wystąpieniu przybliżę historię budowy naszego Klastra, do czego można go wykorzystać oraz dlaczego uważamy, iż jest to wyjątkowa instalacja w skali Polski a może i Europy.
Opowiem jak żyć i pracować z prawie 1500 serwerami za ścianą i dlaczego chłodzimy je gorącą wodą.

Kubernetes 101 - Deploying Your First Cloud Native Application
Catalin Jora Kubernetes 101 - Deploying Your First Cloud Native Application

In this hands-on presentation you will learn what Kubernetes is and how you can use it to deploy highly available applications. The most important moments in the lifecycle of an application would be covered in a demo:

- creation of the environment where the application will run
- deployment of an containerized app
- application debugging
- exposing the application to users
- scaling up
- zero downtime updates

Take-aways:
After this presentation you'll be familiar with the main Kubernetes concepts and you'll have the knowledge to deploy your first application on top of Kubernetes.
The demo will be made available as an open-source repository.

12:00 - 12:40
Dlaczego docker@localhost to nie DevOps?
Wojciech Gawronski Dlaczego docker@localhost to nie DevOps?

W tak dynamicznych czasach jak dzisiejsze - gdy biblioteka staje niemodna po roku, a każdy projekt który trwa dłużej niż 3 miesiąc to "legacy" - wydajesię jakby "ludzie kodu" zapomnieli o tym, że systemy informatyczne żyją. W konsekwencji w erze wszechobecnego ataku buzzwordów, także tych związanych z DevOps, programiści dalej w imię tego podejścia oddają aplikację do zupełnie innego zespołu, a ten zmaga się z utrzymaniem jej przy działaniu na produkcji. Chciałbym wam opowiedzieć historię nawróconego programisty, którego seria pewnych doświadczeń czegoś nauczyła. Będzie to opowieść jak zobaczyć, że zabawa z dockerem na localhost nie ma nic wspólnego z DevOps. Podzielę się własnymi doświadczeniami oraz tym jak nawrócić innych - jak kultura pracyDevOps zmienia pogląd (także mój) na kwestię odpowiedzialności za żyjący system.

Why Did Devs Teamlead Ask Me to Implement OpenShift?
Piotr Baranowski Why Did Devs Teamlead Ask Me to Implement OpenShift?

Small software house implements PaaS. The story of why and how did we implement OpenShift to develop and deliver our own software. From manually managed VMs to automatic PaaS implemented with OpenShift. From puppet, vagrant, virsh, qemu to buildconfig/deploymentconfig/dockerfiles. 

13:00 - 13:40
Improving Security by adding InSpec to Your Workflow
Mandi Walls Improving Security by adding InSpec to Your Workflow

InSpec is an open-source testing framework for infrastructure with a human-readable language for specifying compliance, security and other policy requirements. You can use InSpec to easily integrate automated tests that check for adherence to policy into any stage of your deployment pipeline. InSpec can help you manage your compliance needs on multiple platforms and for multiple applications. This talk will examine the key aspects of InSpec for creating comprehensive checks for your systems, as well as the InSpec tools that allow you to incorporate security and compliance checks into your pipeline tools.

DevOps-Ready Database Automation
Dawid Szymaniuk-Paryz DevOps-Ready Database Automation

While introducing the DevOps into organization we faces many challenges. The one non-trivial challenge is an approach which enables database patching. In the ideal world there should be several tools already implemented in an organization which support version control and management of database schemata. But, the most often, the reality is opposite and the situation is that the next version of a database schema it just a single archive package containing sql statements whithout the logic and described only in few words in quasi-documentation. In such situation, system administrator, very gently, has to deploy such packages on each environment combining the process with the other changes, which in case of failures are delegated to a developer. The question is; whether big companies should resign implementing end-to-end DevOps solutions because of burdensome process of database patching? Or, is there a possibility of painless transition into position where database packages are under version control and automatically deployed into every environment, and, an administrator's interventions is required only in case of deployment failure?

In our presentation we would like to show our observations in this area and our approach to practical realization of the deployment of database packeges.

13:40 - 15:00
LUNCH
15:00 - 15:40
LIGHTNING TALKS
16:00 - 16:40
Platform Architecture & App Development w/systemd
Maciej Lasyk Platform Architecture & App Development w/systemd

So systemd is here in major Linux distributions. Even Facebook uses it in scale of it’s servers farms. Do you know what systemd brings to the table? How to benefit from it as architect, developer and sysadmin? Do you know, that it is not only an init system, but a core part of Linux created by developers for developers having in mind syssadmins sanity? During this talk You’ll see how to stop crafting imperative bash init scripts and enter declarative world of unit files. How to analyze boot time of your VM operating system (or container) and check performance with cgroups. You’ll learn how to confine Linux processes for stable VM performance using systemd-run. We’ll go through dark corridors (that are very useful) of systemd-bin revealing some of its treasures. I’ll show you how to start using systemd-journal for aggregating and analyzing logs. We’ll even see HTTP logserver! After this talk you’ll be able to start your application with systemd-based health checks and self - healing protocols. And you will see how to make application (or Linux container) start when your client pings defined TCP port with predefined payload (this is called a “socket activation”). Last but not least - forget about Docker containers and start using systemd-nspawn. Why use anything else when you have containers already installed as core component? Learn how to benefit from it! My main goal for this presentation is showing you that systemd helps - but you have to know how to use it. It’s a huge beast, but when enthralled it may charm you! Be prepared for huge data upload in a very short time - stay caffeinated!

From Zero to Production in 1 Conference Talk Time
Milen Dyankov From Zero to Production in 1 Conference Talk Time

NoSQL, Serverless, DevOps, Social Auth, ... Developers these days want to forget about infrastructure and simply build great scalable apps quickly. This just got easier and I'll prove it to you by live coding an application that you can interact with during the talk (bring your laptop and/or smartphone). While having fun playing with what I code, you'll learn how to easily build any web application - from simple SPAs trough a full micro-service architecture to complex enterprise applications.

17:00 - 17:40
Machine Learning by Example
Michal Matloka Machine Learning by Example

Are the machines learning on their own? Wait, is Skynet already here? During this session we will tackle an easy Machine Learning problem, show how can it be processed on Spark including data cleaning, normalization and a learning process. Live coding session, but only if the machines don’t rise against us.

Web Caching at Allegro
Jan Taczanowski, Szymon Jasinski Web Caching at Allegro

With over 12 million of active users and up to 20Gbps traffic to serve, ensuring the stability and speed of the Allegro platform requires a reliable caching layer. Having the right tool do the right job is crucial. For that reason, we use both Varnish and nginx in our reverse caching proxy clusters depending on what we want to achieve. In our presentation, we will talk about how we route traffic and how we perform operations on HTTP headers and why we sometimes use SQL in VCL. We will talk about some of the tools we develop internally and about our open source Varnish cluster management tool called VaaS. We will also spend some time describing how we monitor our caching infrastructure. On the nginx side of things, we will demonstrate how things like trim, page cache, sendfile affect caching and why they matter. We will tell you how we reduced our image caching farm by half to serve twice as much traffic by choosing the right solution. We will also demonstrate how to feed empty cache quickly without impacting backend performance.

19:30 - 23:59
AFTERPARTY