ServePath Blog

This week we rolled out a new graphic on the ServePath homepage that when I thought about it, really shows a change in mindset. ServePath has been doing managed, dedicated hosting for over 9 years and will continue to do so in the years to come. However, what has changed is the mainstream adoption of Cloud Computing and Cloud Hosting. Our Cloud division, GoGrid, has been providing Cloud infrastructure for well over a year now and has seen tremendous adoption by the likes of Web 2.0-ers, SMBs, Government Agencies and the Enterprise.

One feature that we were first to market with, is Cloud Connect. Other terms that we have been using is “Hybrid Hosting” or “Hybrid Networks”. Essentially, it enables you to combine the dynamic scalability of a Cloud-based front-end (on GoGrid) with a traditional managed backend of dedicated high-performance servers (with ServePath). The two hosting environments are combined via secure, gigabit private networks, thus enabling best-of-breed solutions to meet any infrastructure requirements.

As always, a picture is worth a thousand words…

If you have any questions about these new hosting strategies and infrastructure possibilities, I encourage you to contact a ServePath Sales Representative. There are many scenarios and opportunities available with this incredible flexibility of hosting, but we want you to surprise us with your ideas and designs. You shouldn’t wait to start planning out your Hybrid hosting solutions, we can get you there now!

Perhaps that subject was not strong enough. The Financial Sector is currently weathering a hurricane, recently suffering the largest drop since 9/11. Merrill Lynch fell into the hands of Bank of America. Lehman Brothers is in bankruptcy and looking for a buyer with Barclays buying some of their assets. The Airline industry is failing. AIG and other financial companies are looking for some sort of an economic bailout. HP is eliminating 24,600 jobs. And this was all over just a few days. If one extends the look a bit further, the perspective is just a grim: gas prices going up, the dollar losing value and housing going down. One simply cannot be surprised by any of this.

The Tech Sector is getting hammered as well, but this time, it isn’t “our fault.” The Dot Com bust managed to drag down the other sectors last time, but we learned our lesson. Long gone are unproven businesses and their associated models. Venture Capitalists and Angel Investors are taking long looks at business, not just getting in the car for a drive but doing a full check under the hood, looking at the road both ahead and behind and fully vetting the drivers and passengers. To get money as a start-up is truly an accomplishment nowadays. You have to have a proven business model, installed user base, and a clear direction of where your company and your industry will go.

I recently attended TechCrunch 50 which showcases 50 startups and allows them to present their business or service to a panel of experts. I saw about 1/2 of the companies’ presentations and I noticed that the companies where they couldn’t articulate or prove their monetization strategy, these companies got an earful of criticism from the experts. Similarly, at a meetup in San Francisco, the question asked every presenter is “How are you making or going to make money?” It’s a very simple question, but one that must be answered or the company loses credibility.

Perhaps we should apply these same simple questions to the Financial, Housing and Airline Industries? I guess the markets are already doing that.

It will take a long time before all of these markets start to recover, and corporations and businesses are currently challenged to prevent the hemorrhage of money and capital expenditures within their IT infrastructure. I recently read an article in the Wall Street Journal called “Cutting Tech’s Energy Bill” by William M. Bulkeley that discusses how large companies are looking at ways to cut electricity usage within the Enterprise. With energy costs directly and indirectly rising, it’s critical for the embattled IT manager or director to make fiscally sound and environmentally responsible decisions to keep their business moving forward will simultaneously ensuring that their technology progresses.

Bulkeley gives several examples:

• IBM has launched a “Green Data-Center Services” business line to help customer redesign their datacenters
• HP purchased a company that specializes in designing datacenters with a focus on energy-efficiency
• EMC has worked to eliminate unneeded equipment and use their cooling infrastructure more efficiently
• Hartford Financial Services Group has shut down 6 of 7 datacenters and host within a “green” IBM facility
• IBM and HP has introduced water-cooled servers and others are hard-selling blade computers that use significantly less power than traditional servers

Outsourcing one’s infrastructure is a core way to tighten your belt of your IT Budget. If you can allow for a managed hosting provider to run your IT infrastructure, you save on capital expenditures as well as human capital running it. Colocation and dedicated hosting were all the rage a few years ago and, while it may be losing its sex appeal nowadays, there are still plenty of IT traditionalists who demand it.

Towards the end of the WSJ article, Bulkeley starts to discuss virtualization as a Green technology that can help cut costs. This is where I pick up the thread and run with it. Virtualization is a key component to helping Corporations reduce their IT expenditures significantly. The ability to consolidate multiple low-end servers onto one or handful of higher-end servers is an obvious and logical cost and energy-savings option. The heavier adoption of virtualization technologies such as Xen or VMware or even Hyper-V is giving corporations or even those self-same outsourced data center and dedicated hosting providers a way to stretch their money and efficiencies even further. To over simplify, reduce the number of servers through virtualization and your datacenter space demands go down, your dependency on IT staff to manage those locations reduces and your CapEx shrinks, giving you efficiencies immediately.

This is where the Technology Sector is starting to get “Cloudy.” I’ve used this metric before to illustrate my point, simply look at the Google (Insights for Search) chart comparing “cloud computing” against “dedicated hosting” keyword searches:

Dedicated hosting will not go away. It’s a viable outsourced technology option that companies depend on. It makes fare more sense (cents?) than doing it yourself within your corporation. There are only a niche of companies that can afford to make the technology and capital investment to support large-scale IT infrastructures, and even those (as exemplified in the WSJ article) are looking to re-architect their infrastructure.

Could “Cloud Computing” be the silver bullet to help corporations survive the financial hurricane? I think so. But there are challenges ahead for both the providers of the Clouds (and even traditional dedicated hosting providers) as well as corporations.

• For Cloud Providers, education of the “Cloud” concept and overcoming the “this technology is only for early adopters” status will be critical. However this can be achieved through collaboration with other Cloud Providers and Enablers as well as standardization of protocols and APIs, for starters.
• For the Enterprise to view the Cloud as a viable alternative to traditional or even self-hosted infrastructures and datacenters, the challenge is larger. While Cloud Computing may be obvious to many  in terms of “green-ness” and cost savings through zero CapEx, IT managers of the Enterprise tend to not quickly jump on board with the latest technology. Some might say they are a bit gun-shy and would rather someone else test the waters and learn from their mistakes. This wait-and-see attitude will be the end of many. Given the current financial weather and outlook, the Enterprise should be looking at the non-traditional and emerging technologies just as hard as within traditional practices.

While I may be sticking my head in the sand by saying that this financial storm will pass soon, I also  have my head in the “clouds” by stating that dedicated/outsourced hosting and Cloud Computing are viable alternatives to “doing it yourself” that all businesses should seriously consider and get on their short term strategic plans. To jump back into the car metaphor, it’s time to dump the old 1970’s Plymouth and get the 2008 Hybrid!

ServePath to host Philippine Airlines web site . Dedicated server provider and colocation hosting specialist, ServePath, has been selected for its cost-effectiveness and scalability, by Philippine Airlines (PAL), to host its corporate web site.

John Keagy, President of ServePath added, ”Philippine Airlines came to us to improve the performance of its online services, which are accessed by thousands of customers per day. Our dedicated hosting and managed services allow PAL to deliver the best possible customer experience. ServePath’s hosting solutions are ideal for organizations like PAL that need to focus on core business issues–not IT issues–and lack the in-house expertise to manage their hosting themselves.”

For more than six decades, PAL has been serving passengers through a route network that spans 32 foreign cities and 20 domestic points, the airline continues to grow and expand its list of destinations.

Switch & Data, a colocation and hosting company, settles law suit for 3 million. The original claim was for more, when the complaint

“sought $29.7 million in damages for the alleged breach of a lease at a colocation facility in West Palm Beach.”

Ouch. I guess they worked it out.

According to this article on Colocation and Disaster Recovery Facilities:

“There are many reasons you could be faced with a disastrous situation. Especially if you are in the IT sector and/or if most of your data resides in electronic format. If you depend on computers for storing, accessing and using your data, you are potentially at risk.”

This is why Servepath offers Data Backup & Storage Area Network services to our customers.

Here are a couple vid clips of some data center workers moving the new Liebert CRAC Units into the data center. These units are so big they had to take the back of the elevator off to fit the thing inside!

Every morning I would walk through our data center to get to my desk. I always liked walking through there, just to see what was going on. Sometimes a colo customer would be in there setting up his server, or the data center technicians would be at work on our Cat6 cabling upgrade. On a warm day it was nice to get cooled down even if the noise was a bit overwhelming.

But since we recently moved to a new office to make room for our data center expansion, I no longer get my morning walk-through. Instead, I have a sweet window cube with lots of sunlight and a view of the street, so that certainly makes up for it. And, of course, I can always pop over to the data center whenever I want to see what’s going on.

By moving our sales, marketing and accounting department we freed up 3,000 square feet of data center space. This will allow our existing customers to expand as they grow their business, plus it will enable us to meet increasing demand for San Francisco data center space. The expanded space features Cat6 cabling designed to support applications such as IPTV, and it allows up to 10 gigabits per second hand offs to our cabinets or cages.

This is what my former office looks like now:

If you are looking to place your servers in a colocation facility or data center, you might have come across the term U. U is the standard unit used to measure the height of colocation racks, cabinets and the hardware which is stored there. 1U equals 1.75 inches, so a rack described as 20U has 20 spaces for hardware equipment and 35 inches of vertical usable space. 1U is the smallest unit and 7U the largest.

The width and height of a rack mount might be standard, but the depth sometimes changes. When you decide to store your server in a data center, be sure to check the exact specifications of your rack or cabinet. As I mentioned in an earlier post on Colocation Installation Tips, it is also good to let the data center technician know about the kind of hardware you’re bringing in, so they can determine where it is going to be stored. It’s no fun for anyone if you show up with gear that doesn’t fit into your assigned space.

When we last left our hero (me) he was telling us how to avoid cardiovascular illness. Really. Heart disease is the number one effect of data loss in IT. Or it feels like it, anyway. Probably the number one effect of data loss is simply failure. If you’re a small business, and you lose all your data, there’s a 70% chance that, this time next year, you won’t be a small business anymore.

Of the four points we’ve listed below, we’ve covered the what and the where; we still need to examine when and how to back up.

How to back up?

As I said below, ideally everything that needs to be backed up is in one spot. Ideally this spot is the backup media, which is in a government-controlled facility at the bottom of a mine under a mountain with four or five OC-768s. The real world is never ideal, unfortunately, so some of this just isn’t possible. However, you can get closer or further from ideal depending on how you set things up.

Try to keep whatever needs to be backed up on a separate partition; on a separate drive if you can manage it. This allows you to use tools such as dump (nww), which is pretty much the best tool for the job (on FreeBSD anyway). You can even keep configuration files in this space. The fewer places you have to hunt for files, the better.

If you use backups as a kind of undelete function (say, if you are backing up personal files and sometimes need to restore a file you’ve deleted), then you should keep incremental backups. Incremental backups store everything that has changed since the last backup. These allow you to make frequent, short, small backups. If your work changes often (for example if you are backing up an office fileserver) you can run such backups several times during the business day. Then, when a file is needed, you can grab it from an older incremental backup.

Alternatively you can make full backups each time. This minimizes restore time, since with only one backup file you can restore all your data.

More complicated structures, such as the Tower of Hanoi differential algorithm that is suggested in dump’s manpage, are nifty but not strictly necessary. They achieve an optimum between space consumed, time to back up, and time to restore.

A further consideration needs to be given to databases. You can’t simply dump the actual database files; when you restore, you’ll find the database is corrupted. Instead, regularly dump the database (a real database can dump without locking whole tables) and then back up the dumpfiles.

When to back up?

At what time to back up isn’t really hard to determine: whenever the load is lightest.

You might want to put some thought, however, into deciding how often to back up. As I said below, not everything need be backed up at once, but only as often as it changes significantly.

If you are making incremental archives or backups, you can run the backup three or four times daily, and then push those archives offsite at night.

Further reading

(all links open in a new window–sorry)

• Backup strategies
• Backup basics

Most of this has been directed at Unix-like operating systems. Some of this applies to Windows as well, though in Windows you can’t simply back up your configuration scripts and then restore them to have your applications work just as you remembered them. If you enjoy paying for software, there are many third party backup utilities availables, such as Veritas Backup Exec. But most of these guidelines should apply.

This week I thought I’d share some installation tips with you. For those of you who are first time colocators or perhaps just starting to think about colocation it is nice to have an idea of what to expect when bringing your gear to a colocation facility, like our San Francisco data center.

1. Please make sure you install your servers facing the right direction
I know, this sounds like a joke, right, but it’s actually not Our rows are designed so the backs of the servers face each other, creating a current of heat that rises every other row. Because servers draw hot air in the front and push it out the back, hot air would come out of one side and be pulled into the servers in the next row if each row had servers installed the same direction.

2. If you are mounting a server on sliding rails, attach the rails first loosely.
It will be easier for you to mount the server if the rails have some give and can move to accommodate the server. After the server is in place, you can finish tightening the screws that hold the rails in the rack.

Also, we provide screws to standard equipment, but if you have special stuff, you need to bring your own

3. If you are using a hub or switch that has an AC adapter type power cord (the block kind), tape the plug to the hub or switch with electrical or duct tape.
Now that we have been doing this for some years, we have found that these cords can easily become disconnected from your hub or switch, especially in shared environments. Taping the cord to the hub will ensure that it does not become disconnected. Another good alternative is purchasing a larger 16 port rackmount switch, which normally use a standard AC power cord that is more difficult to knock loose.

4. Fill out all the forms and paperwork beforehand
If you submit your completed paperwork before showing up at the data center, your installation will go a lot smoother. When you are prepared, it makes it easier for everyone, especially for the data center employees who are planning and scheduling their time.

Also, it is good to let the data center employees know about ALL the equipment you’re bringing in – the size of your gear determines where it will physically be located in the data center and it takes time and money to move everything around.