Month: November 2010

Virtualization For Disaster Recovery

Virtualization As It Relates to Data Recovery

In making the decision to utilize such virtualization technologies, Microsoft points out some important points to consider and understand regarding a virtualized infrastructure.

1. How do I protect data in the virtualized infrastructure?

2. How do I limit my risk to my virtual servers assuming they are compromised?

Every business considering virtualization should be able to answer these questions on their own or through the services of a third party provider. (Microsoft Corp, 2009).

A reputable outsourced IT organization considers these inquiries when designing and building a data center (so that the small business owner does not have to); they will have a structured system for protecting your data in case of a system failure.

Disaster recovery is a advantage of virtualization over traditional systems.  Particularly since, despite its critical importance to businesses, disaster recovery plans are often an afterthought due to equal parts cost, and wishful thinking.  Virtualization, manifested through a reputable hosted IT provider’s services, automates the process to the extent that the burden to foresee, plan, and have a data recovery strategy in place is effortlessly shouldered by the underlying virtual system.

Hosted IT Service providers and their virtual systems utilize several techniques to abolish the adverse impacts of system failures.

Take, for instance, the typical Microsoft SQL Server an accounting department may use to manage business-critical financial data for an organization – – clearly a priority for high-level protection.  A hosted IT provider can virtualize the SQL Server infrastructure so that it runs on two different physical computers and all data is stored on a separate storage area network (SAN).

By deploying the SQL Server in a virtual platform, multiple physical servers can host the application data simultaneously.  If one fails, the other instantaneously services the users the other would normally handle.

Compare, a single server model that leaves users/customers un-served after a crash. A scenario that is always unexpected, always inconvenient, and in the worst cases can be potentially crippling if the timing is bad.  A virtualized system, on the other hand, can immediately transition to accommodate users/customers from the failed system, without them even knowing any crash occurred.(Microsoft Corp., 2008).

The simple crashes described above occur frequently enough during a server lifespan that they should be addressed. Crashes in any system are a near-inevitability, they are in many ways unpreventable and can result in not only the data loss business owners dread, but also temporary interruptions of service.  Because such crashes are not completely preventable, it is paramount to minimize their impact, a principle aim of virtualization.

In fact, while virtualization can minimize the impact of potential losses of service, it can effectively eviscerate the aforementioned dread of data loss though its data management.

Virtualization: Data Recovery and Failover Systems

As stated earlier, data fuels a company’s applications.  It is the often the key resource for a business regardless of size.  Thus there is a concomitant duty for whoever manages a company’s data to zealously safeguard it.

To meet this need virtualization rapidly restores data from backups. For example, consider backing up email. Symantec and Gardner found that 75% of a company’s intellectual property resides in e-mail and other messaging applications. Traditionally, businesses safeguard and back up their data on-site through tape backup or other procedures performed by local IT staff. Some small businesses even attempt to have one or two employees with limited IT experience perform this function.  Traditional disaster recovery involved restoration of data from tape backup or similar media.  This involved laboriously reinstalling all system software so the server could function and begin serving clients again.

In a hosted environment data backup and recovery is orchestrated by a highly trained administrator.  Using virtualization technologies the administrator automates the backup process.  to a virtual backup system.  Tapes or other physical backup media are not necessary. In a virtualized environment outsourced IT providers can offer independent data storage. This type of storage is often offered through a storage area network (SAN) which uses computers operating separately from web and/or file servers.  The SAN thereby does not succumb to the same crash that affected the web and/or file servers.  In other words, separate data storage preserves the data in the virtualized environment when the primary operating system fails. In this environment recovery of the server can be accomplished by cloning the server to a backup, thereby creating a new virtual server. The clone can be immediately reconnected to data on the SAN, allowing for rapid recovery from the crash. (CDW, 2008)

The virtual environment has added advantage of various redundancies, and exponentially quicker recovery.  CDW notes that simplifying the server in order to make it easier to recover after a disaster is a principal goal of  application server virtualization and consolidation.  (CDW Business Continuity Reference Guide).  In fact, server virtualization is now the leading technology used for disaster recovery. (CDW Business Continuity Reference Guide).  This is, in part, due to the elimination of unpredictable recovery time.  A good outsourced IT provider will typically provide maximum downtime guarantees up-front for various contingencies.  For example, being able to depend on a maximum of four hours downtime and a twelve hour recovery time, allows a business to pass those time-frames on to a fickle client base that would consider taking their business elsewhere, rather than dealing with the unknown or unpredictable.

Virtualized systems also offer the additional benefit of replication; a use which can go beyond recovery.  With several simple commands, an administrator can replicate an entire data center, backed up in online storage, at will.  Such virtualized systems are accessible at any time. (Microsoft Corporation, 2009).

Virtualization also offers a “failover” or redundant/standby system that can seamlessly switch over if there is some manner of abnormal termination of an application or network.  This is just another nuance of virtualization that facilitates “business as usual” despite computer problems as usual.

Learn more about virtualization on

Why Hosted Makes Sense

To the same extent that we rely on the data which fuels our businesses, those business rely on the Information Technology (IT) systems which control that data.  Perhaps more significantly, business absolutely depend on an all-important caretaker of these IT systems.  As the cornerstone of many businesses, this trinity of (1) data, (2) systems, and (3) IT management must be available every single moment they are needed, in other words, seven days a week.  (CDW 2010).  In sum, where these three tools (data, systems, IT management) are a company’s lifeblood, those companies need access to them without excuse, and without exception. (CDW 2010).

However, achieving a comfortable level of reliability can be at odds with achieving an affordable level of reliability.  This is especially a significant hurdle for smaller businesses.  The cost of maintaining an IT staff can be oppressive.  The spectrum can range from a single employee to manage all IT needs, to a part-time independent contractor limited to upkeep and problem-solving.  Neither presents the ideal solution as there is too often a compromise of either reliability or affordability.  Many enterprises are given the unenviable decision of not hiring the internal, skilled IT staff that can manage and repair a robust computing system, or hiring an expensive full-time employee to keep their systems up and running.

 Cloud-computing and virtualization are at the heart of an outsourced IT infrastructure that can provide the solution to this perennial business dilemma.  We will consider outsourced IT and its effect on cost control, hosted applications, data virtualization management, scalability, and disaster recovery, when considered against more a more traditional IT infrastructure.