SCALE 10x - 2012 Southern California Linux Expo - Advanced

Replacing Your Proprietary Scale-out NAS with GlusterFS

JacobShucart — Fri, 18 Nov 2011 08:01:39 +0000

Audience:

Categories:

GlusterFS is an open source, software-only, highly available, scalable, distributed storage system for modern data centers. GlusterFS is one way for public, private and hybrid cloud environments to scale their data storage capacity to hundreds of petabytes, and across multiple geographic locations. Learn how Pandora users GlusterFS to scale out their file serving operation around the globe. Highly Available GlusterFS delivers enterprise class high availability via local and remote replication capabilities. GlusterFS is the only highly available storage solution for Amazon Web Services (AWS) Elastic Compute Cloud (EC2) and Elastic Block Storage (EBS) and GoGrid’s Cloud Hosting environment. GlusterFS enables business continuity in the public cloud and enhances both business continuity and disaster recovery capabilities in the private cloud. Superior Economics With GlusterFS, superior economics is the rule. In both public and private cloud environments you pay only for what you use. In the private cloud you are empowered to choose and deploy on any certified commodity hardware. Only pay for performance and capacity as and when it is needed. In this session, attendees will learn what is required in a typical GlusterFS rollout - what are typical use case, where does GlusterFS shine, and where it doesn't. Attendees should walk away with a good understanding of how to start their storage project.

Presentation Slides:

Replacing Your Proprietary Scale-out NAS with GlusterFS.pdf

Audio:

FUSE with In-Kernel Meta-Data Caching

gwr — Fri, 18 Nov 2011 02:10:12 +0000

Audience:

Advanced

Categories:

Kernel

FUSE (File-system in User Space) traditionally uses a very "thin" kernel module that communicates using a message passing interface to a user-level process that does all the interesting work. That approach forces a large number of very fine-grained requests for each high level file-system requests. An alternative to the traditional FUSE design is presented, where the kernel file-system module can do meta-data caching and some aggregation of requests, allowing the rendez-vous with the user-level helper process to be less frequent and more efficient. This allows the elimination of a significant portion of the work normally done by the user-level FUSE library. Also new in this approach is the use of illumos "door calls" (an efficient inter-process communication mechanism inherited from OpenSolaris) for communications between the in-kernel file-system module and the user-level FUSE service program. Experimental results from a prototype of this design are compared with the more traditional implementation, demonstrating some advantages of the new approach.

Presentation Slides:

fuse-scale-2011.odp

Developing for illumos

gdamore — Tue, 15 Nov 2011 18:08:10 +0000

Audience:

Advanced

Categories:

Kernel

We will present an overview of the tools and APIs that affect device driver developers, including a sample very simple device driver. We will cover the core DDI, resources to find more information, building the software, the tools to load and test drivers, and some high level debugging strategies.

Video: