Video: YouTube

Summary: Data scientists in both industry and academia have been using GPUs for AI and machine learning to make groundbreaking improvements across a variety of applications including image classification, video analytics, speech recognition and natural language processing. In particular, Deep Learning – the use of sophisticated, multi-level "deep" neural networks to create systems that can perform feature detection from massive amounts of unlabeled training data – is an area that has been seeing significant investment and research. Although AI has been around for decades, two relatively recent trends have sparked widespread use of Deep Learning within AI: the availability of massive amounts of training data, and powerful and efficient parallel computing provided by GPU computing. Early adopters of GPU accelerators for machine learning include many of the largest web and social media companies, along with top tier research institutions in data science and machine learning. With thousands of computational cores and 10-100x application throughput compared to CPUs alone, GPUs have become the processor of choice for processing big data for data scientists.

Bio: David Williams is a Solutions Architect for NVIDIA, working with Enterprise and Startup companies over the Southeastern section of the United States. Born and raised in Houston, Texas, David left the South for the freezing shores of Lake Michigan to attend Northwestern University for his Bachelor's and Master's degrees in Computer Engineering. Making a much needed return to warmer climates, David moved to Chapel Hill after graduation and joined NVIDIA. Solution Architects serve as customer engineering resources, investigating questions and evaluating proof of concepts for companies interested in NVIDIA technology. As NVIDIA has become the leading artificial intelligence company, the technical challenges faced in this new market cover topics of GPU hardware, system software, data engineering, datacenter architecture, deep learning frameworks, and neural network data science. David is excited to discuss the key drivers and introductory concepts of the world of artificial intelligence and deep learning.

Sponsor:

Summary: Alert the office, school, city, or your family to the tornado warning for their specific area. Automatically initiate a pre-recorded announcement or activate an alarm to encourage everyone to take shelter – but only if the alert affects the area. This information is available via the internet, but the internet might fail in the event of a serious storm. Redundancy via the weather band provides additional certainty that the message is received correctly. This open-source project takes us through python coding, GitHub, continuous integration, integration of messages from different sources with different specificity, and testing of a device that should carry out its primary function perhaps annually. The finished product will be suitable for offices, schools, and outdoor alerting.

Bio: From the country’s hot spot for tornadoes per-capita, Huntsville, Alabama, Jim brings together an understanding of the severity of tornadoes and concern for fellow Red Hatters’ safety as a Safety Warden into a nighttime coding gig to create the ultimate tornado warning device.

Summary: This talk with showcase how to use Docker to orchestrate multiple services like node.js and Golang and I will be demonstrating how to interrogate Docker containers in order to debug what your Docker Containers are doing. We will also utilize TDD with both Golang and Vue.js and demonstrate via live demonstration good software practices.

Bio: Jean-Marcel Belmont is a Software Engineer at FoodLogiq. He is an organizer of 2 different meetups. He is also a former 82nd Airborne Paratrooper and combat veteran. He enjoys teaching others about Software Craftsmanship and spending time with his beautiful family. He loves nature and spending time learning new things in technology.

Ring in the end of the year with TriLUG! Our annual Holiday Social is a great time of delicious food, cool demos, and socializing with other FOSS enthusiasts. This year, we're back at the Red Hat Annex in downtown Raleigh.

Demos of all kinds are welcome. It can be as simple as a cool piece of software you found on the Net, or as complicated as a 3-D printed electronic Halloween costume. Just let us know in the signup form (below).

This year's event will be a potluck, so bring something tasty along to share with your fellow TriLUG members!

Please fill out the quick sign-up form at the link below so we know if we need to fill in any culinary gaps (TriLUG will provide one or two dishes as well), and so we can make sure we have everything we need for the demos people are bringing.

RSVP here!

We have a lot of really talented members, who do a lot of really interesting things. Come on down and share some of those things with your fellow members! Let us see your hobby hacks, your nifty shell incantations, or the latest tricks you've managed to make your Raspberry Pi or Arduino do. This doesn't have to be long - a few minutes is more than sufficient.

Note that this could also be a nice way to introduce whatever demo you plan to bring to the December Holiday Social - whet the appetite for more, as it were.

The Sign-Up Form is up and ready for entries!

Talks:

• Presenter: Brian Gerard Topic: Creating a Self-Installing Package Summary:I'll show a method for creating a self-installing package out of a git repo using gmake and a little scripting glue. While it's more a neat trick than a best practice, it has come in handy at work as a stopgap while we get more battle-hardened solutions in place.
• Presenter: Michael Hrivnak Topic: The Lounge IRC client Summary:The Lounge is a web-based IRC client, and I love using it.
• Presenter: Bill Farrow Topic: Building Debian packages for ARM with Docker and QEMU Summary: Building Debian packages is fun, but building them on a slow embedded system is not fun. Cross compiling is messy and complicated. Using the superpowers of Docker and QEMU you can build those packages on your screaming 16 core x86_64 desktop in no time at all.
• Presenter: Barry Peddycord III Topic: Mastodon: The OSS Alternative to Twitter Summary: Learn about Mastodon: A modern, open source implementation of the GNU Social protocol for microblogging.
• Presenter: Dwain Sims Topic: Rapidly Changing Storage Formats Summary: I will take a really quick walk through the rapidly changing world of storage form factors, and make some predictions!

Video: YouTube

Summary: If you pay attention to tech news at all, you're probably hearing a lot about Linux containers. In this talk, we'll take a look at what's behind the hype and dive into what containers are really good for, what they are, and the best ways to start getting your feet wet where it makes sense to use containers. This will include a look at Atomic Host, an opinionated take on Fedora, CentOS, or Red Hat Enterprise Linux designed strictly to run Linux containers and help organizations focus on applications inside the container.

Bio: Joe Brockmeier is a long-time participant in open source projects and former technology journalist. Brockmeier has worked as the openSUSE Community Manager, is an Apache Software Foundation (ASF) member, and participates heavily in the Fedora Cloud Working Group. Brockmeier works for Red Hat as a senior evangelist for Linux & containers.

Sponsor:

October 12 - Meeting Announcements

Slides: Here

Note the venue - we are in our new location!

Summary For those of us still configuring physical servers and storage, using RAID to ensure the availability of storage is part of our everyday jobs. But things are changing rapidly in the types of storage devices being used, as well as their characteristics. Standard RAID techniques may carry consequences that solution architects may not realize are present.

This talk will examine the beginnings of RAID, how we got to where we are at now, and some of the issues that currently exist with RAID. Alternate techniques (including open source solutions) can provide more resiliency with fewer gotchas. There will also be a quick tutorial on "fio", the flexible I/O tester.

Bio Dwain Sims has been working in the UNIX/Linux world for the last 30 years. He started out using BSD UNIX on a PDP-11 while in Graduate School (West Virginia University), and he has been hooked ever since.

Dwain has worked as a system admin and data center architect, product manager, trainer, and as a Systems Engineer for various companies (including Sun Microsystems and Fusion-io). He currently works for Western Digital / SanDisk / HGST. (yes, he is confused at times!)

He has been working in the storage (mainly Flash) world for the last 5 years, helping bring the storage revolution to the masses. He lives in North Raleigh with his wife Cleo (of 28 years), 4 children (1 married and living in Charlotte - a Nurse Practitioner, 2 at NC State, and one still living at home), and 5 hens.

September 14 - Meeting Announcements

NOTE the venue change! We're at the Red Hat Annex, where we have held our last couple of Holiday Parties, in downtown Raleigh.

Summary Conflation between open source, open data and open government has been problematic in fitting open data related open source projects to meaningful outcomes. Jason Hare will discuss data management and the link between open data and performance management and measurement. To do this Jason will run through a survey of some of the open source technologies used to manage data and create visualizations. Jason will also discuss issues around open data and its adoption by public sector agencies in the Triangle and how civic tech groups are adding value to open data programs through open source efforts.

Bio Named one of the top 100 global influencers in Government Technology by Onalytica in 2017, Jason Hare is the Open Data Evangelist for OpenDataSoft. He has worked in the public sector and in data dissemination for most of his 25 year career. He serves in several organizations on the topic of data governance and open data: In the previous Presidential Administration, he was a frequent participant at the White House OSTP on the subjects of data quality and open data; he was a member of the White House OSTP Open Data Advisory Group; he is listed as an Open Data Expert on the World Bank's Open Data Tool Kit; he is an advisor on data governance to the Chemical Genome Project.

In 2013 he became a charter founder of the Open Data Institute Node for North Carolina. In the same year he became an Open Government Partnership (OGP) Delegate. He guest-blogs on the Open Government Partnership (OGP) website and is an Open Data Moderator for OpenSourceWay, the Red Hat Open Source Blog. He previously served as an advisor for the Open Data Working Group on behalf of the US OGP Civil Society Delegation. In the past he has also consulted with the Open Knowledge Foundation to help craft Ireland’s OGP submission and spent several years working on the ground with Irish Civil Society.

His Open Data projects include: The North Carolina Department of Commerce; Town of Chapel Hill; Durham City and County; Durham Public Schools; City of Raleigh; The City of Newark; The City of Cambridge; The City of Chelsea; Ireland’s Ministry of Reform; Colombia’s Ministry of Information and Communications; The Commonwealth of Puerto Rico; The Colorado Secretary of State’s Office.

You can find him on twitter @opendatasoft and @jasonmhare.