DevConf.cz 2016 has ended
Can’t make it to #DevConfCZ for any reason? You can still attend virtually: youtube.com/RedHatCzech

Sign up or log in to bookmark your favorites and sync them to your phone or calendar.

Storage [clear filter]
Saturday, February 6

14:50 CET

Finding Storage Neverland
Are you a developer working on a code that interacts with storage? Do you know which tools, libraries and APIs are available to make you life easier? I'll give you an overview of what and for which use-cases is available in the current storage stack and what lies in the bright future ahead. Let's write less and share more code! Collaboration works, remember?

avatar for Vratislav Podzimek

Vratislav Podzimek

Software Engineer, Red Hat
Vratislav Podzimek is a software engineer at Red Hat recently mostly working on storage-related code used by the Anaconda installer and other tools -- Blivet, libblockdev and libbytesize. He's also an active contributor to the Anaconda installer itself and related projects.http://blog-vpodzime.rhcloud.com... Read More →

Saturday February 6, 2016 14:50 - 15:30 CET
e. E104 (72 places)

15:40 CET

Ceph Rados Gateway overview and roadmap
Ceph is a highly available distributed software defined storage, providing object, key/value and file-system interfaces. Ceph RGW (Rados Gateway) provides HTTP REST api that is S3 and swift compatible. We will provide an architecture overview of ceph and RGW and will talk of RGW road map.

avatar for Orit Wasserman

Orit Wasserman

Senior Principal Software Engineer, Red Hat
Orit is a senior principal software engineer at Red Hat, focusing on Container and multi cloud storage. She was a principal architect at Lightbits labs working on NVMe/TCP software-defined storage. At Red Hat, she worked on Ceph object storage (Ceph Rados Gateway), a highly available... Read More →

Saturday February 6, 2016 15:40 - 16:20 CET
e. E104 (72 places)

16:30 CET

Automated GlusterFS Volume Management with Heketi
Have you ever wanted a simple way to create volumes in GlusterFS? Did you ever wonder how many bricks to create and where to place them? Well, worry no more! Heketi is here! Heketi is a RESTful service which provides automated creation, deletion, and expansion of GlusterFS volumes across not just one, but any number of GlusterFS clusters. We will show the architecture and usage of Heketi and discuss how it could be integrated with cloud services like OpenStack Manila.

Project Page: https://github.com/heketi/heketi

avatar for Luis Pabon

Luis Pabon

Principal Software Engineer, CoreOS
Luis Pabón is a software engineer at CoreOS. Prior to joining CoreOS in November of 2016, he worked at Red Hat Storage, NetApp Advanced Technology Group, and at EMC on various storage products. He also previously presented at Vault in 2016 and 2015.

Saturday February 6, 2016 16:30 - 17:10 CET
e. E104 (72 places)

17:20 CET

NFS-Ganesha and Distributed Storage Systems
Many enterprises still heavily depend on NFS servers to access their data from different operating systems and applications. Stand-alone systems have always been a bottleneck. It is immensely challenging to deploy a Highly Available, Clustered NFS server; especially on distributed storage systems. In this session, we will discuss the possible solutions and some important aspects in using NFS-Ganesha to scale out your storage needs. NFS-Ganesha is an extensible user-space NFS server which supports NFSv3,v4,v4.1,v4.2 and pNFS. It also has an easily pluggable architecture called FSAL (File System Abstraction Layer) which enables seamless integration with many filesystem backends (GlusterFS, Ceph, GPFS, Lustre, XFS etc.). We'll take a detailed look at the Clustered HA implementation by taking a specific example of a distributed storage, GlusterFS.


Kaleb S. Keithley

Sr. Principal Software Engineer, Red Hat
Kaleb currently works on Red Hat Gluster Storage and Community GlusterFS. Prior to the Gluster acquisition Kaleb worked on HekaFS, a multi-tenant storage proof-of-concept based on GlusterFS. Prior to coming to Red Hat he worked for EMC's Centera and Atmos divisions on a portable... Read More →

Saturday February 6, 2016 17:20 - 18:00 CET
e. E104 (72 places)
Sunday, February 7

09:00 CET

Freak show (#2): CTDB -- Scaling The Aliens Back To Outer Space
Samba is the well established server software implementing SMB and related Windows-protocols on Linux since almost a quarter decade now. Despite the fact that it has to follow Windows as the reference implementation of all these protocols, Samba was able to substantially raise the bar: Long before Microsoft announced SMB file server cluster with SMB 3.0 in late 2012, Samba created an all-active scale-out file server cluster when running on top of a distributed / clustered file system with the introduction of the CTDB software.

While CTDB is at its core a special purpose clustered database and messaging transport for Samba, it is also a slim little resource manager for virtual IPs and a few other services in the cluster.

This presentation describes the architecture of Samba and the history of the creation of CTDB. It demonstrates how easy one can set up a scale out Samba cluster on top of a distributed scale-out file system with the example of GlusterFS, and illustrates a few of the pleasantries that the SMB protocol brings with it in the clustered environment, such as fail over without aborting I/O by virtue of durable/persistent file handles.

Finally the current project to move towards a unified HA infrastructure for storage components is introduced.

(The talk will be co-presented with Günther Deschner (FAS: gd).)

avatar for Michael Adam

Michael Adam

Principal Software Engineer, Red Hat
Michael Adam is an enthusiastic open source software developer, interested in all things about storage and containers. One of the main developers of Samba since more than a decade, Michael is an engineering manager at Red Hat, leading two worldwide teams: The Samba team for Gluster... Read More →

Sunday February 7, 2016 09:00 - 09:40 CET
e. E104 (72 places)

09:50 CET

Lies, mapped drives & device-mapper statistics
Linux has provided block device IO statistics for well over 15 years, with familiar tools such as iostat, sar, and PCP able to benefit from the same sets of counter data maintained by the kernel.

Device-mapper, the kernel framework that underpins applications like LVM2, multipath-tools, and dmraid, as well as modern storage virtualisation features, including thin provisioning, and caching has provided its own in-kernel statistics support since 2013 (v3.11).

In addition to the basic statistics already provided by the kernel this includes several new features such as the ability to track a user-defined latency histogram, independent statistics for arbitrary regions of devices, and high-resolution timekeeping.

avatar for Bryn M. Reeves

Bryn M. Reeves

Principal Software Engineer at Red Hat, Red Hat
Bryn has worked at Red Hat since 2004, in roles including kernel developer training, support and sustaining engineering, and development. His interests include software defined storage, LVM2 and device-mapper, and the boot process.

Sunday February 7, 2016 09:50 - 10:30 CET
e. E104 (72 places)

10:40 CET

Build your own Scale-Out Storage with Gluster
During this workshop we will give the attendees a short introduction on Gluster. After that, the participants can install a Gluster environment on virtual machines on their own laptop (or in their own clouds). Several Gluster developers will be available to assist users with configuring applications (QEMU, Samba, NFS-Ganesha, ...) to use Gluster storage. While we have some examples of applications, attendees can bring their own as well.

Participants are highly encouraged to prepare at least two virtual machines (Fedora or CentOS preferred) that can be used for this workshop. There is no need to install Gluster packages in advance.

Duration: 90 minutes

avatar for Niels de Vos

Niels de Vos

Senior Software Engineer, Red Hat
Niels is a core-developer and maintainer for Gluster. He is employed by Red Hat and works together with other teams who provide professional support for Red Hat Gluster Storage. The main areas where Niels is active include network protocols, low-level/Operating Systems improvements... Read More →

Sunday February 7, 2016 10:40 - 12:10 CET
workshops A113 (64 places)

11:30 CET

The future of disk encryption with LUKS2
For years Fedora provides disk encryption option
in basic installer configuration. The LUKS (Linux Unified Key Setup),
implemented through cryptsetup library, provides convenient way to
configure such a basic disk-encrypted systems.

In this presentation we will focus on new requirements
for deploying disk encrypted storage in modern systems.
We will present the new LUKS2 format definition that will allow
implementation of these requirements in future.

These requirements are both technical (for example integration
to an enterprise key management systems) but also based
on new advancements in cryptographic algorithms (for example
new key-derivation functions more resistant to massive parallel
systems used by attackers for password cracking).

Another current requirement is an ability to change encryption
parameters without need of complete disk re-formatting.
We will describe prototype of a re-encryption tool that allows
such a change on a fully running system without any downtime.

Last but not least we will mention some interesting answers
from users participating on a survey questionnaire focused
on usage of disk-encryption systems.

avatar for Milan Brož

Milan Brož

Milan Brož is a principal software engineer working for Red Hat and upstream cryptsetup/LUKS maintainer.

Ondrej Kozina

software engineer, Red Hat
I'm software engineer working for Red Hat in storage/LVM team and also RHEL cryptsetup maintainer.You can discuss cryptsetup, LUKS2 and reencryption with me.

Sunday February 7, 2016 11:30 - 12:10 CET
f. E105 (72 places)