RelyNet will be hosting the Sacramento Ruby October Meetup at our offices on November 19th. What’s on the agenda? Ruby Quiz! We’ll be holding our irregular “Ruby Quiz Competition”. Come test your Ruby chops (or learn Ruby!) by solving a fun problem in friendly competition with other members. We team up Ruby novices with seasoned developers so everyone can participate and learn from each other. After the competition, we walk through the teams’ efforts to learn what tricks of the trade were employed.

As always, a number of folks will be around the hour before the formal meeting (6:30pm) to conduct the “Ruby Clinic”. Bring your questions, problems, or other issues in and we’ll try to help you out. Whether you are a new learner of Ruby, someone trying to set up a new machine, or an experienced developer with a tough problem at work, the Ruby Clinic is the place to turn to.

Hope to see everyone at the meeting! More Info Available at the Sacramento Ruby Meetup Page.

This is just a little write-up that we ran across derived from a talk at the 2009 Startup School. “What Startups are Really Like”. In the past, we’ve been involved in our share of startups and know that many of you have done the same. This is a little lengthy, but a pretty accurate breakdown that really resonated with us here. Hopefully, you enjoy it and take something away from it too.

Building for Evolving User Needs:

We try to keep an eye on the trends, identify what our users are responding to, and look for new opportunities to create in these areas. Recently the focus had shifted to virtualization technologies and the cloud. With zero overhead and the flexibility of a virtualized environment, its widespread intial appeal is no mystery. However, after some careful examination of the popular cloud offerings available, we discovered a few things that could be improved with their models.

While the appeal of a virtualized environment was obvious, people seemed to be glossing over the potential hidden costs, possible scaling issues, and the new hurdle of conforming their environment to a cloud provider’s API. With that, we set out to create an environment that would offer the best that virtualization could promise mixed in with some of the straightforward appeal that traditional colocation had to offer.

All The Benefits of the Cloud. None of the Haziness:

Well after a substantial amount of planning and development, we are now pleased to announce that we now offer our customers a zero overhead, affordable solution, with very high degree of flexibility and control. The solution is an entire compute environment: rapidly scalable, instantly deploy-able, platform agnostic, and completely accessed and controlled from anywhere.

What is EcoSystem?

Built from the ground up on enterprise-proven standards, EcoSystem encompasses premium hardware resources and expert configured layers for redundancy, system management, and connection management. Simply put, EcoSystem gives you all of the benefits of the cloud. None of the haziness.

Always worked with Traditional Colo?

Unlike traditional colocation, there are no upfront equipment costs. Clients can entirely avoid hardware maintenance. This means no more midnight runs to the colo to replace a drive! Ultimately, watching equipment depreciate and fall off the map is now a thing of the past.

Been Taking a Second Look at the Cloud?

Customers considering the cloud now have a serious contender. This compute environment offers them a totally hands-off option with complete control at a distance. An upfront pricing structure means no outrageous or hidden costs when it is time to scale. Having built this solution on top of a proven virtualization platform eliminates the need to conform their environment to an API or any constraints the cloud had. They now have their very own resource cluster completely at their disposal.

A Compute Environment to Encompass the Best of Both Worlds.

We’ve chosen the name EcoSystem because it encompasses all the elements that your company needs in a compute environment. Each EcoSystem is built on top of premium physical resources, backed up by at least 2TB of high performance storage & RAID 10 redundancy, robust management tools through XenServer’s cloud-proven virtualization platform, and an expert configured connection management layer. Our EcoSystem offering is built to give you the best of what both traditional colo and the cloud have to offer. Check it out and see for yourself. All of the benefits of the could. None of the Haziness.

RelyNet will be hosting the Sacramento Ruby October Meetup at our offices on October 15th. This month’s meet up will focus on a DataMapper Overview. As always, a number of folks will be around the hour before the formal meeting (6:30pm) to conduct the “Ruby Clinic”. Bring your questions, problems, or other issues in and we’ll try to help you out. Whether you are a new learner of Ruby, someone trying to set up a new machine, or an experienced developer with a tough problem at work, the Ruby Clinic is the place to turn to.

Hope to see everyone at the meeting! More Info Available at the Sacramento Ruby Meetup Page.

Enjoy cooking up some Ruby code and looking for a challenge? Well, come join the The Sacramento Ruby Meetup group as they present their “Summer Hack-a-thon” event here at RelyNet.

Participants will split into a couple of teams and tackle building a Ruby app from ground up in six hours or less. The activities will include brainstorming the project, some lighting-agile planning, and accomplish ing a handful of iterations with the goal of deploying our handiwork before the evening is done.

If you’ve wanted to work on a fun project, check out the latest Ruby, Rails, other Ruby-oriented tools, and experience a little Extremely Extreme Programming and Design, now is your chance.

The group is trying to keep this to a reasonable size so please RSVP if you are coming. More information about the event can be found here at the official Sacramento Ruby Meetup page.

If you’ve been inside the data center lately, you’ve probably noticed our most recent project developing in the southeast corner of the data center floor.

Well, we are pleased to announce that the project is now complete and that we have two new caged spaces available to current and potential clients. As you might expect, we are very excited about this new addition to our colo space. We take a lot of pride in the growth and development of this place and it is certainly fun to watch each new piece of the puzzle fall into place. More importantly however, it is something that we’ve received a lot of interest about from potential clients.

Both facilities are generous in size and competitively priced. If you know of a potential client that would be interested in on of the new floor spaces, please let us know. As always, we appreciate your referrals and are happy to reward you for your loyalty.

I decided I couldn’t wait any longer to upgrade my MBP with a SSD drive. After a few months of reading reviews I concluded that an Intel MLC SSD was the best choice at the moment. Not wanting to spend $800 I opted to get the 80GB Intel X25-M which was onsale at newegg.com for $363. The only problem is that my new drive would only be 74GB formatted and my used space was 140GB. The following are steps I took to reduce my primary HD footprint.

1. Get an external disk. I had a 320GB Western Digital Passport I was already using for Time Machine backups. I affixed this drive to my MBP with some classy velcro.

2. Move iTunes Data. I freed up several gigs of music, movies and TV shows by telling iTunes to store my iTunes Library on my external drive.

3. Move iPhoto Data. I moved my 24GB of iPhoto data the external drive. I couldn’t find a way to make iPhoto use a nonstandard location in its preferences, but a symlink worked fine.

mike$ cd ~/Pictures
mike$ mv iPhoto\ Library /Volumes/Storage/
mike$ ln -s /Volumes/Storage/iPhoto\ Library

4. Remove foreign language files. I was blown away at how much HD space is dedicated to foreign language support. I downloaded and ran a useful tool called Monolingual. It took about 2 hours to run but in the end it removed 4.7GB of foreign language files. The fine print said it could possible cause some problems with newer Adobe products which try to be self healing and rely on check sums. My CS3 installation seems unaffected. I made sure to keep English and English (United States) in my Monolingual config.

5. Remove old iPhone backups. If you have an iPhone and update your software regularly, odds are you have about 2GB of iphone OS backups. These can most likely be removed completely. I copied them to my external drive since there is plenty of room for them there.

6. Move Downloads. I moved my default Downloads directory to my external disk.

7. Clear caches and adjust settings. There were a few large caches that I felt weren’t necessary. For instance, Google Earth was using about 1.3GB of its 2GB cache allotment, which I lowered.

All this got my HD usage down to about 50GB without removing any applications leaving plenty of room on my new SSD. Hopefully someone else doing some HD spring cleaning will find this useful. I’d like to hear any other ideas!

UPDATE: Monolingual did actually cause a problem with Adobe Acrobat. You can exclude certain directories in your Monolingual configuration before it runs. I would suggest leaving out your Adobe application directories.

Mike

RelyNet will be hosting the Sacramento Ruby April Meetup at our offices on April 16th at 7:30pm. The “Ruby Clinic” will start 1 hour (6:30 - 7:30) before the Meetup and will go up to the starting time of the meeting. Bring questions about Ruby, Rails, or web development in general and group organizers and participants will do their best to help you out with your application feature, hard to squash bug, or just deepen your understanding of a particular subject.

More info on the Sacramento Ruby Meetup Page.

An earthquake with a preliminary magnitude of 4.3 and epicenter near Morgan Hill struck at 10:40 a.m. today, according to the United States Geological Survey.

Fortunately no damage has been reported at this point, just a very rattling moment for the residents and businesses of the area.

According to the San Jose Mercury News, “The southern segment of the Calaveras Fault, between Paicines and San Felipe Lake along Highway 152, is one of the fastest creeping fault segments in the San Francisco Bay region. Historic surface measurements show that the fault is creeping in the range of 11 to 19 mm per year. Geophysical investigations show that as much as 174 kilometers of offset has occurred along the Calaveras Fault in the past 12 million years. This translates to roughly 13.7 mm of offset per year.”

For the full story as it unfolds you can visit the San Jose Mercury News directly here.

RelyNet Status: Our facility was in no way affected by this incident as we are located safely away from this fault region.

Here is a quick list of tips and reason for keeping a clean network cabinet. Most are common sense but if you are mindful of them it will make your data center life much easier.

Avoid The Rat’s Nest
That giant tangled mess of wire not only looks bad but it blocks airflow exhausting from your equipment. All those wires at as insulation, traping heat and forcing your equipment’s fans to work harder and fail sooner. Tracking down cables becomes difficult and its easy to accidently unplug other equipment when you are tugging on a cable connected to a tangled mess.

Don’t Exhaust Into Your Intake
Be mindful of how your equipment exhausts hot air. Data centers that utilize hot isle/cold isle configurations like ours work best when all equipment is racked to intake air from the cold isle and exhaust into the hot isle. Although this might seem like common sense, it is not uncommon to rack smaller equipment such as switches, routers and firewalls in the reverse. This heat exhausts to the front of the cabinet and ends up being sucked into servers.

Use Wire Management
You don’t need anything fancy. A simple 1U or 2U wire manager can do wonders. The hinged wire management devices work well but can also restrict exhaust air exiting the cabinet. Try using velcro ties instead of plastic zip ties to bundle cables. Plastic zip ties cinched down too tight can damage cables and they can’t be undone and repositioned.

Mind Your Cable Length
You don’t have to make custom network cables to length in order to have a clean orderly cabinet. Pre-made network cables are you best bet for reliability and can be ordered in any length. With your network switch mounted at the top of your cabinet use 3′ to 5′ cables for the top third 7′ to 10′ cables for the middle third and 10′ to 14′ cables for the bottom third. This should be enough length to create a nice bundle down the side of your cabinet. Your simple 2U wire manager we talked about above will absorb any slack.

Label Everything
Not only is it helpful to have all of your server’s labeled correctly, its also a good idea to label your network cables with the switch port number at server end of the cable. This helps avoiding doing the “search for the blinking switch port light routine” while unplugging and plugging back in the network cable. If you don’t have a label maker, ask your Data Center. We keep one on hand to lend customers.

If you are mindful of these tips you should have no problem building and maintaining a clean data center cabinet.