Planning for the unknown and unlikely can be difficult but having enough funds to implement an IT disaster recovery plan can seem close to impossible. Often companies are faced with more pressing issues to their business than something that may or may not happen, or so they think, which is why we hear about these stories on the news quite frequently. But how unlikely is it that an IT disaster could happen to your company? With the combination of natural disasters, hardware failures and human error, the reality is that data loss is only a matter of when, not if.

According to a NFIB National Small Business Poll, man-made disasters affect 10% of all small businesses and a staggering 30% have been impacted by natural disasters. A simple power outage could put your whole business at risk. In fact, research by the University of Texas has shown that only 6 percent of companies suffering from a catastrophic data loss survive, 43 percent never reopen and 51 percent close within 2 years.

That is why we compiled the top 5 reasons why your company should rethink your IT disaster recovery and prevention plans.

5. Machines and hardware fail. While modern IT hardware is fairly resistant to failures, most devices fall far short of a perfect track record. No one is immune to hard disk or internet connection failures. While it can be costly for your company to eliminate any single point of failure in your IT infrastructure, it is the only way to insure that a hardware failure doesn’t interrupt your service or cause data loss. The less expensive and more sensible option would be to have your data backed up regularly. Ideally, and more cost-effective than building your own top-of-the-line data center, would be to outsource your IT infrastructure to a leading Infrastructure-as-a-Service managed data center operator. This eliminates any capital expenses while ensuring the strictest protection from service interruptions due to IT infrastructure failures.

4. Much like machines, humans are not perfect. They make mistakes. Have you ever accidently saved over a word document or had your computer crash before you could save an important file? It happens to the best of us. Even the most cautious can forget a step in an important process causing data loss or the wrong data to being entered. While very common, these mistakes can often be the hardest to prevent and correct. Creating a series of incremental online data backups lets you easily restore your files to an error-free state. Having redundant firewalls, anti-virus, and anti-spyware software can ensure that security breaches are protected against if one were to be accidently disenabled or a port accidently left open. Often times the most important way to prevent human error is in process improvements and quality insurance activities. Checking and double checking is often the best remedy, with online backups as a safety net.

3. Customers expect perfection. The internet and the web 2.0 revolution have not only forced companies to be more transparent and accountable but competition in all in industries has greatly increased. With increased competition come lower prices, better service, and less forgiving customers. Most customers demand perfection because they know that if your company can’t provide it, the next competitor in line is ready to give it their best shot to win the business. It’s imperative not to give your customers a reason to leave because of something preventable like an outage in service.

2. Customer retention is costly, but customer re-acquisition is devastatingly expensive. While on average it is much cheaper to retain a customer then to acquire a new customer, re-acquiring an old customer after an IT disaster can be next to impossible. It takes a lot to earn customers’ trust but after an IT disaster like loss of data or an extended outage in service, trust quickly evaporates. This is especially true when your customers are losing up to thousands of dollars per minute of downtime, which is not uncommon in some industries. Most IT disasters are deemed absolutely unacceptable to end users. It is far less expensive to prevent an IT disaster than try to re-acquire lost customers after one has occurred.

1. You’re only as strong as your weakest link. The age-old saying is as true today as it was when first uttered. You can have one of the most well-crafted ships ever made, but if your crew can’t see the iceberg just beyond your line of sight, then it doesn’t matter how well-built your ship is upon impact. No business is immune to IT disasters, but there are many things you can do to prevent them or quickly recover. There is no excuse not to have an IT disaster recovery plan in place – today. The ideal situation would be to have your production servers in a top tier data center with no single point of failure on the power and network connections and your disaster recovery backup servers at another data center at least 45 miles away in case of a severe natural disaster. This would ensure that you’re virtually protected from any downtime and you have put your business in a great position to survive the most brutal IT disasters.

Conclusion: Save Money, Save Time, Save Your Business. Develop a Solid IT Disaster Recovery Plan.

No business is invulnerable to IT disasters, but speedy recovery due to a well-crafted IT disaster recovery is expected by today’s ever-demanding customers. Too many businesses fail because they were ill prepared for an IT disaster, even when a simple solution like online backup could have easily saved them. If your business hasn’t thought about developing an IT disaster recovery plan yet, it should be on the top of your list of priorities. Your business and customers demand it.

With a typical dedicated server stack, which is either managed by you or outsourced, you select your server, storage and networking needs, purchase them, and then live with that configuration for 3 to 5 years. During the course of those years, you might add memory, which isn’t too difficult. You might need to upgrade disk drives, which proves a bit more difficult. Over time, you might need to upgrade CPU’s, which is very difficult and expensive because this essentially requires changing the entire server.

When speaking to finance professionals, the following analogy can be used to describe the benefit of  Private Cloud computing:

Imagine you were responsible for bus transportation in a metropolitan area. You are getting ready to order new busses that your community would need to live with for the next 7 years. Busses are available in 20, 30, or 50 seat configurations. You need 1,000-1,100 total seats to accommodate your population and the routes you’ve designed to service their needs. Each line is designed to maximize the use of a 20 or 50 seat bus. So, you guess and buy twenty, 20-seat buses, and twelve, 50 seat busses, and then hope it’s efficient for the next 7 years.

Interesting puzzle huh? Even if done really well, at the end of the day, there’s still tons of room for inefficiencies and wasted resources.

Instead of being forced to live with the same resource configuration for several years, imagine you could purchase a “cloud with 1,000 bus seats” and re-configure those seats at will into whatever size busses you want, at any point in time. Busses could be as small as a single seat or you could have one bus with 1,000 seats.  Got a convention? Make 4 x 200 seat busses between hotels and the conference center. Having this flexibility is really powerful and valuable. Why? Because with the increased flexibility you can have more seats being used at any one time, thus requiring significantly fewer seats while providing a better experience.

That could be called “cloud transportation”, and it would be every Transportation Officials dream. Now think of each bus as a server and you have “private cloud computing” – every IT-finance manager’s dream come true.

An Important New Requirement

We just saw the power of the flexibility provided by Private Cloud computing. Unfortunately, it introduces a level of complexity that wasn’t present before. There are a few unique abilities that must exist in order to really gain the benefit of a Private Cloud.

Continuing with the transportation analogy… you now have this new on-demand flexibility for your busses.  All you need now is a staff of people monitoring the traffic flow and re-designing the busses with more or less seats based on demand and utilization. They would work feverishly to reduce the wait and route times for anyone wanting a seat on any bus. The flexibility is powerful but to get the most out of it you need fairly sophisticated and complex (i.e. expensive) abilities to dynamically change your configurations and obtain the full benefit of the flexibility.

In the computing world, what you need to do is watch the server, storage and network resources. When one appears to be causing a bottleneck, provide it more resources. You also have to remember to reduce the number of resources allocated to a server that doesn’t need it. If you don’t, it sits idle and unavailable for another server that might demand it. This resource allocation process is powerful, but not inexpensive.

That’s where you need to pay attention when designing your Private Cloud. You need it to automatically change the cloud configuration in real-time so resources are where they need to be when they need to be there! In the blink of an eye and automatically, it needs to turn cloud servers off that aren’t being used and turn them back on when they are needed. This means that a small, extremely smart piece of software will have to constantly monitor your server, storage, memory and network resources and compare that to work-loads. It will then estimate and forecast which servers need more resources. After estimating, it then needs to automatically, in real-time, re-allocate resources so that you are always using your cloud computing resources in the most efficient manner. How cool is that? The dream continues…

Cost

The beautiful ending to this story is that the ultimate savings of idle capacity can be passed on to you. The cost for a well designed Private Cloud computing platform is less than a dedicated server on a per server basis! So, not only is it more flexible and can deliver a lower total cost of ownership, a managed Private Cloud can be outright cheaper. That’s the benefit of Private Cloud computing.

About Online Tech

Online Tech owns and manages SAS-70 secure and reliable multi-tenant data centers across the Midwest.  Their private cloud computing services combine the benefits of cloud computing – flexibility and cost-effectiveness – with the security, data integrity, and service level agreements (SLAs) of Online Tech’s SAS 70-certified, dedicated environment. Visit http://www.onlinetech.com for more information.

The article discusses the need for in-depth remote server monitoring tools, monitoring dashboards, and skilled engineers who are capable of managing thousands of servers with millions of data points. Without the proper tools in place for monitoring servers, you may be lulled into a false sense of security about the health of your servers and the applications your customers depend on. You need proactive monitoring tools in place in order to know the server is in jeopardy – before it’s too late.

Think about it – how do customers feel in 2010 when they can’t access data?

Don’t you hate when you need to access a website and you receive “Server Too Busy” or another server/application error message? It almost always seems to happen when you need access to a website and you have limited time to grab the information and get your job done.

Receiving an error message is so frustrating! It’s bad enough when you squeeze in time on the weekend to pay bills online and can’t load your bank’s website. Just think about what paying customers feel like when they can’t access critical data that is absolutely necessary to complete their job.

The cause of the error is usually one of the following:

1)  Too many users trying to access underpowered server hardware
2) Server application error
3) Client browser, software, or end-user hardware error
4) Insufficient Internet capacity or bandwidth at the hosting location
5) Server or network security breach

Making sure all of these issues are resolved can be very difficult. There’s tons of pressure on IT managers to keep costs low and provide excellent server uptime, especially in the post-2009 economy. A good IT architecture (think VMware, great hosting infrastructure, capable staff, solid applications) and reliable remote server monitoring is a fantastic place to start.

In 2010, keeping your servers running, regardless of where you host them (in-house, colocation, or in a managed dedicated server environment) requires the right tool set to manage the infrastructure. Do you grab some open source server and network monitoring code, learn how to use it, deploy it, configure the probes and monitoring thresholds, and hope it works? Or do you beg for capital to purchase enterprise monitoring tools, support, and the corresponding training sessions? Or do you outsource 24/7 monitoring and support service to an outside vendor.

This can be a hard question to answer. Each presents different resource demands for capital and operational expenses, and risk of downtime exposure.

Regardless of the hosting environment, remote server monitoring can go a long way in helping to limit those problems and help you avoid downtime. While, not the solution for all “Server Too Busy” problems, it helps to ensure that problems and potential problems are brought to awareness as quickly as possible – and hopefully before negatively impacting your online customers.

Managed colocation is similar to a fully managed server except that the hardware is owned exclusively by the client. With raw colocation, the server owner is responsible for monitoring and tracking, responding to and repairing problems with their server and taking preventative measures like backing up their data. With managed colocation and managed servers, many of those operations can be outsourced to a data center operator that can offer those services at a lower cost than a company could provide by themselves.

One advantage of managed colocation is that a client’s hosting needs and problems are being addressed by industry experts with extensive knowledge and experience. The client will not have to worry about hosting issues because they are paying for the managed services each month. With managed colocation the client will also be lowering their overall cost of IT management. Because the staff and space is being shared across many servers in the data center, the client will pay a lower cost than hiring data center staff internally.

Other than who owns the hardware, there are very few differences between managed colocation and a managed server from most hosting providers. However, one advantage of managed servers over managed colocation is faster remediation time on hardware failures. With a managed server, the hardware is owned by the hosting provider who most likely will have spare parts readily available to replace hardware quickly and the availability to substitute a similar server to prevent downtime during maintenance. For managed colocation, each client’s servers can differ dramatically and hardware failures often require time-consuming delays while coordinating the order, delivery, and replacement of custom server parts.

Managed servers possess similar advantages to managed colocation in that they can provide clients with expert management at lower costs than what it might cost to host internally. However, managed colocation is best for companies that already have the server hardware and expert IT staff that want to be in control of the management and just need the security and reliability of a data center provider and basic management services. The managed server option is better for companies that don’t have the staff or hardware and want more in-depth services than basic maintenance and problem solving. Both options allow a company to leverage the complete expertise of the data center staff and get absolute security, management, monitoring and reporting.

All things considered, with both managed colocation and managed servers companies should think of the services as a function of their business rather than just the outsourcing of IT services. Excellent data centers allow companies to pick and choose their desired services and level of control. In the spectrum from raw colocation to managed servers, every business can find an IT management solution that will work best for their scenario.

About Online Tech

Online Tech owns and manages SAS-70 secure and reliable multi-tenant data centers across the Midwest.  With a full range of raw and managed colocation, fully managed dedicated servers, and server monitoring and management services, Online Tech reduces IT data center costs, operational risks and downtime and insures their clients’ servers are always on, always online, and always safe. Visit http://www.onlinetech.com for more information.

By choosing the right colocation data center partner, a company can ensure the proper security and reliability for their IT infrastructure. However, finding the right location for data center hosting can present a challenge. Factors that come into play typically include: proximity to company headquarters, disaster risk and price. Michigan’s low risk for natural disasters and cool, mild climate make it an ideal hosting location for data centers.

Until recently, the advantages of Michigan data centers meant little to companies across the country because of the inconvenience of installing and managing their servers to a location far from their business. However, some Michigan colocation providers have removed these obstacles with a set of hassle-free colocation options for clients near and far.

Some Michigan colocation providers have made it easy to ship servers directly to them for installation and easy to manage those servers remotely without ever setting foot in the colocation facility.  Michigan colocation services like server “rack and stack” simplify clients’ data center migration and managed colocation makes it easy to outsource server management to a remote colocation provider.  Services like remote monitoring and web-based portals make it easy to manage the IT infrastructure over the Internet.

A great example of a Michigan colocation service is “Rack and Stack”, a server installation service designed to eliminate the difficulties of installing a company’s servers at a data center located far from their place of business. The Rack and Stack service includes:

• Rack design and layout
• Receiving and uncrating the servers
• Inventorying, labeling and recording all server serial numbers
• Professional assembly and rack mounting
• All power and network cables required
• Intra-cabinet cabling as specified
• Fully documented cable list
• Final documentation of layout and cable runs
• Rack layout digital pictures

Typically, every step listed above is carried out according to each company’s exact specifications. All documented colocation rack design layout information and digital pictures are provided to each company in order to simplify ongoing support and ensure the satisfaction of every customer.

With this new service, clients never have to set foot in the data centers and yet can enjoy complete visibility of their server installation. This makes it easy and even more appealing to use a Michigan colocation data center facility from anywhere in the country. The need to send IT technicians to Michigan for the installation is eliminated, saving time and money. In addition, oftentimes the resulting rack is much better laid out, wired, and documented than the client had before the move.

Moving to a top-tier Michigan colocation data center facility has never been easier. With server rack and stack, managed colocation, remote monitoring and web-based portals, companies can maintain focus on their core IT mission instead of being concerned about the unpredictability that typically comes with data center migration. Michigan colocation installation professionals are knowledgeable, helpful and experienced in making the move and server management a seamless, simple process and eliminating the normal concern of whether the job is being done right.

The top Michigan colocation data centers can deliver server colocation more cost effectively, more reliably, and with a Midwest work ethic that is rarely available at other data centers across the country. Now any company can take advantage of these benefits and leverage Michigan colocation no matter where it conducts its business.

About Online Tech

Online Tech owns and manages SAS-70 secure and reliable multi-tenant data centers across the Midwest. With a full range of colocation and managed dedicated server offerings, Online Tech reduces IT data center costs, operational risks and downtime and insures their clients’ servers are always on, always online, and always safe. Learn about its Michigan colocation service on their website.

Traditional software companies teach their sales employees “to avoid disclosing any more information than the minimum necessary to close the deal”. Because of the one-time payment structure of those deals, it creates a short-term focus on both the deal and the relationship with your potential client. However, with SaaS services one can “try before they buy” and, depending on the contract length, walk at any time if the service is not to their liking. This creates a long-term focus, making transparency and trust between the two parties more natural.

As Joel points out, Google Adwords is a great example:

“Ask yourself how much you spend right now on Google AdWords without ever having spoken to a sales rep. How does this compare to your own average selling price for online transactions?  Now, ask yourself why.  The answer is transparency, from company reputation to cost-per-click.”

While Joel focuses his discussion on the software industry, this is true for all products that can also be services including SaaS hosting. This is seen with big ticket items like cars and houses as well as everyday purchases like music and books. The larger the purchase and commitment, the more risk involved and the less trusting the two competing parties become. Full transparency becomes the only logical play when dealing with Product-as-a-Service businesses, just as it is the only logical play when dealing with SaaS hosting providers.

Anyone familiar with the prisoner’s dilemma (a game similar to a negotiation process) knows when the game is only played once (like purchasing a good such as a car) both parties have the incentive to “defect” or lie (it is the rational strategy). However, when the game is played multiple times for an amount of time unknown to both parties (like a subscription based service) the rational strategy is to “cooperate”. This repetition in “games” creates the incentive to take the leap of faith into a cooperative strategy. If one party defects, the other will defect for the remainder of the negotiations and vice versa. With just one defection, there is no trust, and everyone is worse off for the rest of the series of games.

The as-a-Service model is built on this “repetition of games” trust and cooperation; while you might gain in the short term if you hide the truth, in the long term you will have lost a valuable relationship. This dedication to transparency is one of the lesser talked about benefits of not only moving your business to an as-a-Service company but purchasing from these types of SaaS hosting companies.

Many SaaS hosting companies see and live this phenomenon every day. When companies understand the value of transparency, they will have a conversation with all of their prospects to try to understand their needs. If the prospects’ needs don’t match what the SaaS hosting company offers, the company will let them know. There is no point for an as-a-Service company to sell something to a client that they don’t need. That is not a win-win deal. The client will soon figure out that they don’t need your service and move elsewhere. It creates unproductive work for both parties. This is something that all as-a-service companies should practice.

Consider this: have you seen an as-a-Service company succeed that isn’t built on transparency and trust?

About Online Tech

Online Tech owns and manages SAS-70 secure and reliable multi-tenant data centers across the Midwest.  With a full range of colocation and managed dedicated server offerings for SaaS hosting services, Online Tech reduces IT data center costs, operational risks and downtime and insures their SaaS clients’ servers are always on, always online, and always safe.

oosing a colocation data center, a company should consider a location that is both cost effective and minimizes the risk of natural disasters. A Michigan data center offers the perfect environment to keep a data center running with minimal down time and lower operating costs.

An area with a low risk for natural disasters and has a cool, mild climate is the best location for a data center. A Michigan data center uniquely meets all of these requirements.  A Michigan data center can deliver colocation more cost effectively, more reliably and with a Midwest work ethic that is rarely available at other data centers across the country.

What makes a Michigan data center more stable than those in other locations and the best choice for companies?

1. A low risk of earthquakes.

The U.S. Geological Survey’s (USGS) National Seismic Hazard Mapping project in 2008 shows that the entire state of Michigan has an extremely low peak ground acceleration (PGA). This is a measurement conducted by precise instruments and reflects how hard the earth shakes in a given area. This indicates that the disturbance of the operation of a Michigan data center by an earthquake is extremely unlikely.

Peak Ground Acceleration Map

2. A low risk of tornados and hurricanes.

A Michigan data center faces a significantly lower risk of experiencing tornados and hurricanes in comparison to the rest of the U.S. According to an American Red Cross map detailing national hurricane hazard ar

eas, Michigan data centers sit in the lowest risk zone. The National Oceanic and Atmosphere Administration (NOAA) Tornado Hazard Map depicts the number of recorded tornadoes per 1,000 square mile area between 1953 and 2004 in the U.S.  Michigan is among the states with the lowest tornado risk experiencing only one to five tornados for the entire 50 year span. A Michigan data center is at a much lower risk for a severe outage due to the very low occurrence of hurricanes and tornados compared to the rest of the country.

Tornado Risk Map

3. A history of the lowest number of major natural disasters in the United States.

According to NOAA, from 1980 to 2009, Michigan experienced the lowest number of natural disasters causing $1 billion or more in damage.  In fact, only Alaska has fewer major natural disasters than Michigan.

4. A favorable climate.

A Michigan data center runs more cost effectively in the state’s cool, mild climate. The average outdoor temperature a Michigan data center faces across the year ranges from 25 degrees Fahrenheit to 75 degrees Fahrenheit, lowering the overall cost of cooling the data center as compared to much of the country.  This provides a significant cost advantage for every Michigan data center with free cooling for all and only four months of the year with temperatures that average over 60 degrees Fahrenheit.

5. A Michigan data center has lower power costs and more inexpensive property costs.

A Michigan data center can translate the lower cost it takes to keep the facility cool due to the cold climate as savings for clients. Inexpensive real estate and reasonable power costs in Michigan can also drive significant colocation cost savings.  However, when companies are deciding which Michigan data center to choose they should not pick based on cost alone. Companies should ask enough questions to clarify all of the services offered by each provider and compare those to their current and potential future needs.

According to the Federal Emergency Management Agency (FEMA) around 70 percent of all power outages are due to weather. Minimal extreme weather conditions equal less data center and operational downtime for companies. A Michigan data center will experience fewer natural disasters, has a more favorable climate, and typically incurs lower facility costs that can be translated into lower costs for their clients. A Michigan data center is the best choice for businesses looking to outsource.

About Online Tech

Online Tech owns and manages two SAS-70 secure and reliable multi-tenant Michigan data centers.  With a full range of colocation and managed dedicated server offerings, Online Tech reduces IT data center costs, operational risks and downtime and insures their clients’ servers are always on, always online, and always safe. For more about our Michigan data center, visit http://www.onlinetech.com/managed_hosting/data_centers/

One of the challenges in communicating the value of managed colocation is demonstrating that the operating costs of a server far outweighs its purchase price. While servers are relatively inexpensive, the cost to operate them can be significant. There are many factors that contribute to server operating costs that typically fall into one of four categories:

• Housing and Protecting the Server – Including server location, power and the internet connection; data center capabilities (i.e. raised floors, fire suppression); and firewall or threat management systems
• Monitoring and Tracking the Server – Including any monitoring of operating characteristics (disk space, RAM usage, etc) and the alert messaging system
• Responding and Repairing the Server – Covering any necessary server error services (i.e. troubleshooting and remediation services, part replacement and warranty)
• Preventative Measures – include disaster recovery solutions such as on-site or off-site backup to ensure that the data on the server is always recoverable

What is Colocation?

Colocation is a service that hosting operators offer to house and power servers in their multi-tenant data centers. This can have several benefits depending on the services included by the data center operator. At a minimum, colocation involves a place to house the server, the power to run it, the HVAC to keep it cool, and the network connection to operate it online. The colocation services can then be expanded upon to include everything required to ensure that it is working 100% of the time without any effort from the company.

How is Colocation different from a Managed Dedicated Server?

To sum it up, a managed dedicated server is owned by the hosting provider whereas with colocation, the servers are owned by the client. Managed dedicated servers are fully managed by the hosting provider with a number of server-side features and capabilities than basic colocation. Managed colocation, on the other hand, applies all of the same services of a managed server to the colocated server.

What is Basic Colocation?

Basic colocation typically focuses on just the first category of server costs: housing and protecting. Basic Colocation is a hands-off type of service that allows clients to take advantage of a data center’s infrastructure, but leaves the client solely responsible for managing the server. If there is a server malfunction (ex. server runs out of memory), the client is responsible for monitoring, managing and mitigating any server problem.

Basic colocation services tend to vary between data center operators, but typically provide the following features that vary depending on the quality or tier classification of each data center:

• Physical housing for the servers
• Uninterruptible power delivery
• Internet connections
• Environmental controls (ex. air conditioning and fire suppression)
• Physical security for the data center and servers

The more redundancy that a data center has in its power and network infrastructure, the higher uptime it delivers and the higher data center rating or tier it achieves (tier IV being the highest). Depending on the Service Level Agreement (SLA), colocation providers can often guarantee a power and network availability upwards of 99.99% of the time .

Besides the infrastructure of the data center, the main operating costs (and therefore price motivators) for colocation are:

• Power (in Kilowatts consumed)
• Bandwidth (in Mbps)
• Rack space (measured in “U”s)
• Setup Costs (Labor and Equipment)

What is Managed Colocation?

There is a wide variety of services that fall under “managed colocation” which starts with basic colocation as the underlying service platform. The client provides the server hardware typically loaded with the operating system and applications. However, the managed colocation provider picks up from there – monitoring and managing the server, mitigating any issues that come up and maintaining the hardware for the client. Rather than leaving the client responsible, managed colocation outsources day-to-day server management operations such as:

• Threat management
• Technical support
• Multi-probe around-the-clock monitoring, alerts, and logging
• Asset tracking
• Patch management
• Capacity planning
• First responder to any problems or issues that arise, troubleshooting, and remediation
• On-site backup and/or off-site backup
• Part replacement and service

Managed colocation provides the flexibility to decide what services and procedures the client wants to outsource versus which they want to have more control over. It transfers the day-to-day hassle of managing servers from the client to the managed colocation provider.

Questions to Ask Managed Colocation Providers

1. Basic colocation questions – What tier data center will the servers be hosted at? Is there 24×7 support offered by the managed colocation provider? What kind of physical and network security is in place to protect the servers? What is the service level agreement (SLA)? Is the data center SAS 70 Audited?
2. What does the managed colocation provider monitor? Know what metrics are being monitored by the managed colocation provider and how notification of abnormalities, outages and errors will be made. Will the client be notified when they are close to running out of disk space? Will the alert from the managed colocation provider come via e-mail, text message, or telephone?
3. Is data being backed up and how? Is it on-site or off-site backup, online or tape? How often is it backed up and for how long? What software is used to backup the data and what procedures does the managed colocation provider have in place to ensure that none is lost?
4. What happens if the server malfunctions? This may be the most important question to ask a managed colocation provider. Who is going to troubleshoot and identify the problem? Is the client responsible for the parts or the labor or both? Who will coordinate the resolution to the problem?
5. Are changes to the server being tracked and logged? Are patches being applied to the server? Why? How? When? How long does it take for the managed colocation provider to resolve a support ticket? Is that logged? Are changes to the server’s configuration tracked and logged by the managed colocation provider?

Managed colocation delivers a new middle ground between managed dedicated servers and basic colocation- fully managed, client owned servers in a colocation data center. IT departments that understand their true server operating costs are driving the trend to managed colocation as they move past the basic “ping, power, and pipe” and outsource the day-to-day management of their colocated servers. Managed colocation lets them focus on their users’ experience and applications where they add their greatest value.

About Mike Klein

Mike Klein is the President and Chief Operating Officer at Online Tech, a leading managed data center operator in the Midwest. Online Tech offers a full range of managed colocation and managed server hosting in their SAS-70 secure and reliable multi-tenant data centers across the Midwest. Visit http://www.onlinetech.com for more information.

How do they continue to deliver excellent service without making major capital investments in essential IT and facilities infrastructure?

How do they minimize the ongoing operating expenses and tech personnel costs of being at the leading edge of a very competitive environment?

Below are 7 essential questions any company should ask before selecting a SaaS hosting provider.

1) Is a 100% service guarantee the same thing as an SLA?

SaaS hosting providers offer a certain level of functionality to their customers, with the level of service being specified in a Service Level Agreement (SLA).

Many SaaS hosting providers will offer a “guaranteed 100%” uptime. However, this doesn’t mean they deliver 100% uptime. Instead, the SaaS hosting providers receive a service credit if 100% uptime isn’t delivered. Therefore companies should carefully review the SaaS hosting provider’s uptime statistics and SLAs to make sure it is consistent with the SLA being offered to customers. Also ask for written details on how they deliver their electrical power, network and server SLAs.

2) Do they get the big picture?

Want to know a SaaS company’s worst hosting provider nightmare?  Signing a contract only to discover that their SaaS hosting provider does little more than deliver ‘ping, power & pipe’.

What’s saved in low monthly costs is lost many times over by not being able to expand the infrastructure in order to support rapid growth, or recover quickly from operational problems. Be sure to gain a full, written understanding of the technical and business relationship that will be in place before signing a contract with a SaaS hosting provider.

3) Does the SaaS hosting provider offer ‘high availability’ managed dedicated servers?

Business-class data centers require significant capital investments to provide and maintain space, power, cooling and network functionality in a way that is cost-effective for SaaS companies to take advantage of. At the server level this translates into being able to deliver a range of affordable, high-end managed server solutions that match business requirements.

Contrast that approach with one where a SaaS hosting provider assigns an in-house developer the often hurried task of deploying poorly configured servers for testing, staging and even production. That infrastructure quickly becomes a growth bottleneck due to security, reliability or scalability issues.

4) How do they balance real-time maintenance windows with always-on service?

If a SaaS hosting provider’s hardware stack is designed properly and redundant hardware is used correctly, full production availability can be maintained even during a maintenance window.

Real-time maintenance allows new features to roll out quickly since maintenance windows do not have to be scheduled and announced to customers.  Developers will appreciate it because they’re able to work more normal business hours, but in a real-time maintenance environment. Users like it too because their confidence in the reliability and stability of the entire application solution increases.

5) How exactly will the managed server hosting solutions keep pace with your business growth?

A significant challenge for a SaaS hosting provider is to properly size the “store.” It’s very important to select a hosting provider who has the infrastructure to support future growth.

For example, a company begins with dedicated servers and a managed backup server in the same data center. Then, as clients and data traffic increase, the business model will require that backup data is securely replicated offsite on a daily basis.

But if the managed server hosting provider has only one data center, how will this be achieved? Such unplanned limitations can become major roadblocks for expanding SaaS companies.

6) Does the SaaS hosting provider offer professional incident management and escalation procedures?

Nothing is more frustrating as a SaaS hosting provider than getting little to no information about a service issue from the hosting provider.

The best SaaS hosting providers have a genuine and very visible culture of service that prevents this from happening to clients. Expect to see systems such as port-level network monitoring plus automated trouble ticket notification in the event of a service issue.

Any reputable SaaS hosting provider should be following clearly documented guidelines for all aspects of configuration and change management. If necessary, ask to see these documents.

The best SaaS hosting providers will also show a willingness to help with application upgrades by involving their network engineers and Sys Admin staff to minimize the impact of any service disruptions.

7) How can a business be sure that their customers’ application data is securely and reliably hosted by the SaaS provider?

The Internet has given a whole new meaning to the idea of a ‘data bank’. For SaaS hosting providers, it’s becoming critical that a trusted business partner securely and reliably hosts their customers’ data.
To some extent, managed dedicated servers hosted in an outsourced data center have now become a trusted data bank on which the SaaS business model depends.

SaaS businesses that address these 7 questions before choosing a SaaS hosting provider will increase their uptime and minimize their frustrations with their server infrastructure. This lets them focus on the highest value-add portion of the SaaS business model – the application and the users’ experience.

About Mike Klein

Mike Klein is the President and Chief Operating Officer at Online Tech, a leading managed data center operator in the Midwest. Online Tech offers a full range of colocation and managed server hosting in their SAS-70 secure and reliable multi-tenant data centers across the Midwest. Visit Online Tech’s website for more information about Michigan SaaS hosting.