Redux Blog All Posts in General News

DrupalCon Returns to Portland, May 20-24

(May 13, 2013, Posted in General News by Andrea Hughes)

DrupalCon is coming to Portland this month, and something tells me that things will be a little different than at the last conference held here in town. Portland hosted the first North American Drupal conference in 2005, attracting somewhere in the neighborhood of 100 attendees to the two-day event.

For a hint of what’s in store for us this year, consider last year’s conference in Denver. More than 3,100 Web developers, themers, site architects, project managers, content authors, designers, front and back end developers, and other users and potential users of the Drupal open source content management platform gathered to exchange ideas, experience and insights.

Portland remains a very active open source community, and we’re primed to host the largest event yet put on by the Drupal Association. Over the course of five days, the DrupalCon experience will offer hands-on training, educational sessions, inspirational keynotes and contribution sprints, engaging Drupal users of every level every step of the way.

Although DrupalCon has grown in size and scope since it made its North American debut eight years ago, the collaborative spirit of the event remains unchanged. And what else hasn’t changed? Portland’s love affair with microbrews.

This year, we’re partnering with OpenSourcery to give conference-goers a taste of Portland with an added attraction: pinball. After DrupalCon on Thursday, May 23, we’ll gather at OpenSourcery to embark on a favorite local pastime: a Pinball Pub Crawl.

If you want to join us, but are a pinball novice, don’t sweat it. After all, what’s a conference without a tutorial or two? Greg Dunlap (heyrocker) will lead a brief “how to not suck at pinball” course, and you’ll get plenty of practice on one of four pinball machines at OpenSourcery before heading over to the mecca of pinball âۥ Ground Kontrol âۥ for three hours of free play.

Shuttles will be running between the Oregon Convention Center, OpenSourcery and Ground Kontrol from 4:30 p.m. to 10:30 p.m. To really drink in the local sites and, yes, I meant that literally and figuratively âۥ OpenSourcery and Ground Kontrol are an easy walk apart with plenty of local watering holes with pinball attractions in between.

RSVP or give us a jingle at 503-274-9905, ext. 506, to reserve one of the limited spaces available. Of course, you must be of that glorious age that allows for the enjoyment of local wine and spirits âۥ at least 21 âۥ to join.

We hope to see you there, and we’re looking forward to a fantastic week at DrupalCon!

Drupalcon PDX Pinball Pub Crawl

(May 10, 2013, Posted in General News by Andrea Hughes)

Cap off a full day at DrupalCon Portland with a unique Portland experience: the Pinball Pub Crawl.

We’re celebrating the largest gathering ever of Drupal developers at this year’s biannual conference, where attendees will be addressing open standards and exploring ways to build the Drupal community. With a theme like “Building Bridges, Connecting Communities,” there’s no better way to enjoy Bridgetown than with a little appreciation for some of this town’s favorite community gathering spots.

Who: You, of course! (So long as you’re 21 or older, that is.) Along with Network Redux and our partner, OpenSourcery.

When: Thursday, May 23 – 4:30 p.m.

Where: Start the evening at OpenSourcery for local beer, spirits, food, pinball and table games. If you’re a pinball novice, we’ve got you covered with a quick tutorial on “how to not suck at pinball.” Continue on to Ground Kontrol, pinball and video game mecca, for an evening of free play from 7:00 p.m. to 10 p.m.

How: A free shuttle will run between the Oregon Convention Center, OpenSourcery and Ground Kontrol from 4:30 p.m. to 10:30 p.m.

Or, hoof it from OpenSourcery to Ground Kontrol, taking in some of Portland’s best local bars – and pinball – along the way.

RSVP: ASAP! Space is limited, RSVP to reserve your spot.  If you have any questions please contact Andrea Hughes at

503-274-9905, ext. 506, or e-mail sales@networkredux.com.

Information: Visit http://portland2013.drupal.org/event/pinball-pub-crawl to learn more.

Slow Loading Web Pages, an Invitation to Lose Business

(April 16, 2013, Posted in General News by Andrea Hughes)

Have you ever wondered how your online audience perceives your website’s performance?  Read further to hear about how our partner, New Relic, and their application monitoring software can help you evaluate your entire platform.  Also listed are two popular performance enhancing tools that NR uses to fine tune websites and applications. 

New Relic is an application monitoring solution and a necessity for companies of all sizes. It delves into server resources such as CPU, memory, disk and the network helping track slowdowns and threadlocks.  Most importantly, New Relic displays measurable data, which provides a solid baseline revealing methods for improvement.  NR is a partner with New Relic and our customers have access to the Standard edition, valued at $24/month, for no additional cost.  As New Relic’s CEO Lew Cirne states, “slow web pages are an invitation to lose business”.  To signup visit http://bit.ly/NewRelicNRSignup. 

Opcode Caching is a way to accelerate your PHP based applications.  A server must interpret various PHP files and translate them into something your machine can understand.  By preserving previously generated code in cache, it will limit the amount of information called per request.  Enabling can help shave time off page load by up to 90%.  APC (Alternative PHP Cache) and eAccelerator are open source opcode caches that NR often uses.  The installation of one of these will raise memory usage slightly and can cause more harm than good if not configured properly or lacking memory to allocate.  Make sure that you have enough resources prior to installation.  If not, it may be time to upgrade to one of our Enterprise Virtual Server platforms.

Gzip is the most popular and effective file compression method currently available.  It lowers page response time by reducing the size of the HTTP response.  Each file is replaced by one with the extension .gz, all while retaining the original properties.  Gzip gives a lot of flexibility and agility as you can always restore compressed files to their original format and new files will automatically create based on the original if the compressed file is not a proper fit for its file system.  Approximately 90% of today's Internet traffic travels through browsers that claim to support gzip. 

If you would like further information about any of these applications, please contact us at 1-800-756-6518 or sales@networkredux.com. 

Lead Datacenter Engineer, Cameron Smith

(April 01, 2013, Posted in General News by Thomas Brenneke)

Behind the curtains at Network Redux, our infrastructure team ensure that our physical components are online for our clients 24/7/365.  This requires an array of talents, and we’re fortunate to have a phenomenal team lead, Cameron Smith.  As our Lead Datacenter Engineer Cameron’s team have eyes and ears into every physical component under the hood.

With that said, we thought you may like to get to know Cameron a bit better, both on a personal and professional level.

Cat or Dog?

I love all Animals. I have a 13 year old cat named Sashimi that looks like a panther and will probably get a dog when I have some room.

Portland or Seattle?

Portland (but we have some awesome toys in Seattle). I love the people here it is just the right size city in that it is large enough to have culture but small enough to still be friendly.

110 or 208?

208 and 8-10kW per rack actual draw.

Chassis or Stacked?

Chassis as a cluster node because we love High Availability.

Beer or Wine?

Very dark beer. Actually I love a wide variety of beer and living in Portland we have it good as there are many world class brewers here.

Dell or HP?

Dell kicks HP in the pants! I am hands on with both day in and day out and Dell chassis are rock solid.

Can you tell us about your role as the Lead Datacenter Engineer at Network Redux?

I am in charge of the installation and management of all physical infrastructure at the hardware level for Network Redux. I architect our space usage, all physical deployment, racking, cabling and power efficiency. I love to make the most out of what we have and do it not just to industry standards but to our standards.

You led an internal project to implement Dell Open Manage Essentials across our network, can you tell us a bit about this?

For years we had a reactive approach to hardware health. We did daily walk throughs of our multiple data centers and looked for amber lights. As our infrastructure grew and we moved away from traditional hosting into enterprise managed hosting I needed better visibility into our hardware health and a better way to manage it’s maintenance.

We needed a proactive approach and I chose Dell Open Manage Essentials. We are close to being 100% Dell for our servers and there were many perks to choosing this app over other off the shelf products or building our own IPMI 2.0 management tool.

All of our servers have Dell’s out of band on board card called DRAC and OME integrates directly with the DRAC cards allowing us to have a central point to have an in depth view into all the components in all of our servers. Not only do we have the ability to immediately access a complete inventory of all the parts in all our servers down to the part number and even MAC addresses of ports we now immediately know if a part is in warning or critical status. I have also connected these alert notifications that we receive in OME to our central NOC so our team can address issues 24/7/365.

The icing on the cake for me is how OME handles firmware updates on the remote servers.

We are able to see what servers have new firmware available as OME connects to Dell’s site for the latest releases. I am able to select what updates I want to apply and when and OME handles downloading the firmware files and running the remote updates all automatically. This helps dramatically if there is a critical update or if I want to bring a server up to the latest after it has been in a production run.

You’ve been known for stunning cable management and incredible visual diagrams, where do these skills originate?

I am an artist at heart and was a graphic designer in a past life so the making the things I work with look nice comes naturally for me but there is more to it than that. Regarding cable management we have to keep it organized due to the shear volume of cabling. Some of our private cloud chassis have 12 cat cables each with 3 or 4 nodes. That’s 36 to 48 cables for just one small build running to the network chassis. As our cabling bundles grew and our level of client changed I needed to keep in mind that improperly cabled runs could lead to data degradation due to near-end crosstalk. We use cat 6 exclusively but our bundles can grow quite large as we are now filling 50U cabinets top to bottom with servers so I made sure my design was allowing for those thresholds and I also changed our power routes to reduce contamination.

Your work includes a great deal of forecasting for growth while maintaining incredibly efficient usage rates.  In this effort what trends are you seeing in the datacenter industry which could be considered game changers?

I am noticing changes that are affecting many parts of the industry. Today we are able to do so much more with so much less hardware that it is changing the way data centers are being used. Hosted private clouds are now becoming standard for business growth. Larger public clouds and big data deployments are springing up everywhere so even though we need less hardware to to the same work there is much more work to do. There is just so much explosive growth that data centers are thriving but I see many newcomers to the data center market as well as old school points of view that are getting left behind in this flood as the needs of the clients change very quickly and the agile are able to change with them.

Other changes I am seeing are modifications in how data centers are offering services to large colo clients. They are offering different billing models for power and space to make the deal that much sweeter allowing clients to reduce their overhead and more quickly move up in power and space usage as they don’t have to bite off more than they can chew. If a company can spend those extra dollars these days on marketing or an extra body rather than wasted space or power drops everybody goes home happy.

Meet Alex From PlayHaven

(February 20, 2013, Posted in General News by Cameron Smith)

You may have heard of PlayHaven. 

Their innovative applications are influencing the gaming industry. Alex is one of the team members that is making this impact. We find his insight intriguing and informative.

---Tell us a bit about Playhaven, and your role there. 

PlayHaven is building a Lifetime Value Maximization platform. We currently provide an iOS and an Android open source SDK for game developers to integrate into their applications. This enables the platform to offer a number of player acquisition, engagement, retention and monetization features to game developers.  Most recently we deployed a new segmentation feature that allows developers to send targeted communications, like reward offers, virtual good promotions, opt-in data collection, and ads, to their users from PlayHaven. These features are all configured via a web dashboard alongside performance graphs and live previews, and configuration changes show up on devices in about 5 minutes. I came on board at the beginning of the current arc of PlayHaven, and I currently run the server engineering team, provide database administration, and work on overall software architecture. We are rapidly growing both in adoption by publishers and advertisers and in building features for the platform, so capacity and performance are always a concern. 

---Describe some of the unique attributes in your software development practices.  Any secret awesome-sauce to share? 

Our backend Python APIs are all first-class citizens. Each API has its own area of functional or logical expertise - one maintains user accounts, one maintains advertiser information, and so on - and when one API needs information in another API's domain, it creates a proper HTTP request like any other consumer. This allows us to scale our APIs independently and place them behind load balancers for redundancy and availability. We also try to use hosted solutions for tools that aren't core to our business in a big way. Our code lives in Github, our servers are monitored with Scout, our stories are in PivotalTracker, and we use Google Apps for our corporate email and calendar. The cost of these tools is far less than the development time and opportunities it would take to bring these things in house and host them ourselves - we weigh that cost against the impact it might have on our business if we rely on a third party. NetworkRedux fits well into this approach and allows us to focus on building software rather than hiring the systems and network administration staff it would take to manage the hardware ourselves. 

---What bothers you most about the "cloud" as it is defined today? 

The "cloud" is fiction. It has fundamentally changed the way we consider hosted computing, encouraging optimization for commodity hardware and the automation of deployment in a way that only the most advanced engineering teams might have tackled before. Unfortunately, the "cloud" didn't change the way we build server hardware yet.  This creates many places where that fiction breaks down - for example, Amazon's attached storage service is fundamentally tied to a physical datacenter, and in order to attach a drive to another server's in a different availability zone, you have to manually snapshot it and create a new volume in the second availability zone. This makes sense because the two servers are on separate physical networks, and Amazon hasn't hidden this physical separation from its users. 

---Your passion for gaming, where does this originate? 

Board games and miniatures were a large part of my childhood. My dad and his friends humored me and let me push small units of troops in their military games whether they were simulations of real and fictional battles in the American Civil War and Napoleonic Wars, or the imaginary armies of fantasy or science fiction. I had hand-me-down computers from a young age and was engrossed by games like Elite, Privateer, and Master of Orion. I also harbor a deep love for the LucasArts era of adventure games such as Grim Fandango, Day of the Tentacle, and Sam and Max Hit the Road. It's great for me to watch a resurgence in adventure games from companies like TellTale Games and following Tim Schafer's DoubleFine saga as they build their Kickstarter backed adventure game. 

---Tell us about one of the most complex tasks or adventures you've completed. 

Building the new architecture and codebase for the PlayHaven platform from the ground up was a pretty intimidating undertaking. I got to apply my experiences from previous endeavors where I lived through growing pains of several businesses to work on building a solid foundation. We launched our platform with a giant of mobile game publishing, so there was a lot riding on that initial launch. The launch was a success, but the challenge is an ongoing one as we grow. 

---Tell us something about yourself people would be surprised to know. 

I think like a lot of Portland engineers, I'm a bit of a grab bag of interests and skills. Music has always been a big part of my life, and I was a flutist until college, when I did a lot of choral performance. After a brief affair with penny whistles, I got the digital drum set that I had been wanting to get for ages and have been dabbling in percussion. I'm a bit of a beer snob and a foodie, enjoy cramming as many outdoor activities as possible into the summer months, and skiing Timberline in the winter.