by Pat Hanavan
IT organizations have been caught between a rock and a hard place. Charged with protecting their company's information, IT organizations have established aggressive service level agreements (SLAs) that impact the manner in which they implement data protection by setting recovery point objectives (RPO) and recovery time objectives (RTO).
Organizations struggle with shrinking or non-existent backup windows, the need to recover quickly, often to a specific point in time, and even meeting compliance or regulatory guidelines. Backing up to tape is no longer adequate; not only is it difficult to administer for backups and recoveries, but it lacks the speed, reliability, flexibility and simplicity IT needs to meet stringent SLAs. Backing up to disk using virtual tape emulation or virtual tape libraries also falls short as the administration of the solution is tape-centric and schedule driven. Add in the explosion of data, along with the challenge of protecting remote offices, and you have the challenge facing many of today's business--with IT sitting on the front lines of aligning business needs with today's technology.
As a result, a growing number of IT organizations are augmenting their traditional backup and recovery strategies with continuous data protection (CDP) solutions. CDP dramatically improves RPOs and RTOs while eliminating backup windows. What's more, CDP not only reduces the need for tape in the backup and recovery process but it also makes recovery easy enough that users can often recover their own files, without help from IT.
What is CDP?
CDP is a process that lets organizations continuously capture or track data modifications and stores changes independent of the primary data, enabling recovery points from specific points in the past. CDP systems may be block, file-, or application-based and can provide fine granularities of restorable objects to infinitely variable recovery points in time.
CDP reduces the complexity of the data protection system and eliminates the classic challenge of theing backup window because it eliminates the need for full, incremental, or differential backups by protecting data immediately and then continuously backing it up to disk. CDP is not a complete replacement for traditional backup but rather an important component of a well-rounded backup and recovery strategy.
Can CDP be leveraged for backing up and recovering email? As the predominant form of communication for business transactions, email is an application that is mission-critical to organizations of all sizes. It generates a huge amount of information that must be immediately available and protected. The loss of a single message may generate hours of unnecessary and frustrating labor for administrators and/or users and can lower productivity or affect business operations. And with the introduction of Exchange 2007, organizations need protective solutions that can support the latest offering from Microsoft.
Not surprisingly, the amount of email data requiring protection and availability is growing exponentially. IT, in turn, is faced with the challenge of backing up this critical data within the existing backup window and recovering it quickly. Moreover, they must not only be able to back up and recover whole email databases but they also require a system which enables recovery of individual mailboxes or emails. However, if administrators want to back up email databases for complete disaster recovery purposes and be able to recover individual email, folders, or mailboxes, they typically have had to do separate backups.
New granular recovery technologies have emerged that enable mail messages, mailboxes, and folders to be restored individually without having to restore an entire email database, and without separate and redundant mailbox backups. In an Exchange environment, for example, only a single-pass full or incremental backup of Exchange is required, which dramatically decreases the time required to protect all mailboxes while also reducing the backup storage requirement.
CDP significantly streamlines backup and recovery of email by completely eliminating the need to perform scheduled daily email backups, and speeding recovery, thereby delivering email continuity for businesses.
How does CDP enable end users to recover their own data?
Because CDP is a disk-based protection and recovery solution, it is possible to enable end users to retrieve their own data. Some CDP solutions provide this type of functionality; some utilizing a simple Web interface that requires no training and enables end users to retrieve previous versions of files without contacting IT. Empowering end users to retrieve their own data frees up IT to focus on other business-critical needs of the organization.
With these self-service recovery solutions, retrieving lost, corrupted, or overwritten data is as easy as searching for and downloading a file from the Internet. There is no backup tape to locate or load and no additional information to restore to find the correct file. Best of all, these solutions do not require the installation of client software or agents on individual desktops laptops, and a familiar web paradigm requires no additional training. Users need only a standard Web browser, making data retrieval easier than ever.
Does CDP help protect against data loss or corruption resulting from security threats?
DP offers inherent security against threats by protecting data from the impact of destructive attacks from viruses, worms, Trojan horses and the like. By continuously backing up data, CDP enables organizations to retrieve the last "clean" version of data that was stored before the attack. Being able to "dial back" to just before a loss is a huge win for today's businesses and IT.
What role does traditional tape backup play in data protection today?
For decades, tape has proven to be an effective means for data protection and recovery. Tape is an inexpensive medium for storing data, and its portable format can easily be moved offsite. IT administrators are also familiar with tape and tape backup procedures, making it a relatively simple task to perform-although only trained administrators can recover data from tape.
Tape based backup systems also lacks the flexibility and simplicity that many organizations need today in a data protection solution. Speed of recovery, most especially for individual files or emails, is also an issue with tape solutions. Proper tape media management and rotation, combined with the need to monitor and manage the health of the tape hardware are complex systems which require IT knowledgeable personnel.
Today, more and more organizations are moving to a disk-to-disk-to-tape strategy. They are leveraging the inherent speeds and reliability of disk-based backup--including continuous protection- and then migrating backed up data to tape for secondary data protection and longer term data protection or offsite storage.
Utilizing both disk-based and tape-based data protection systems and methods in the right manner will bring the greatest protection for businesses, and will help IT align its strategies to best meet business needs.
Pat Hanavan is Vice President, Product Management, Data & Systems Management Group at Symantec.
www.symantec.com
COPYRIGHT 2007 West World Productions, Inc.COPYRIGHT 2007 Gale Group
Rabu, 23 Juli 2008
An effective SME disaster recovery strategy for branch offices
by Mark Phillippi
The 2007 Atlantic hurricane season started early this year, the second time in five years a storm has formed before June 1. As a result, experts are again predicting an above average hurricane year, estimating that there is a 74 percent chance a category three or higher storm will hit the U.S. mainland.
While the chance your business will be affected by a major natural disaster or terrorist attack is slim, other causes of unplanned downtime such as a hardware component failure, fire, power outage or surge, network attack or even human error are more prevalent. It is important to ensure your company has a reliable strategy for getting the business up and resuming normal business operations, no matter what the cause. Without a disaster recovery strategy the consequences to your business could be paramount. According to the Wall Street Journal, 92 percent of small businesses that experience significant data loss due to a major disaster go out of business within five years. You don't want to become part of this statistic.
With mission-critical data growth exploding in the small to midsized enterprise (SME) market-much of it on the edge of the network in branch offices-traditional disaster recovery solutions may not be enough to ensure business continuity throughout your entire organization.
Complete Restoration Disaster Recovery
In order to protect all company data regardless of physical location and to ensure business continuity across the entire organization, SMEs need to deploy a complete disaster recovery solution that integrates local backups and remote off-site vaulting in a single solution. Using a complete and integrated solution allows your organization to restore entire systems quickly, and get them back in the production environment as soon as possible after experiencing downtime. A complete bare metal restoration solution restores systems from the ground up-including the operating system (Windows, Linux, Solaris, etc.), business applications (Exchange, SQL, Oracle, Novell, etc.), user profiles and data files, in addition to reformatting drives.
The following details what SMEs should look for in a complete disaster recovery solution that protects their data center as well as data and systems in remote offices.
Quick Recovery
When it comes to quick recoveries and an SME's ability to resume normal business operations after experiencing downtime, the stakes couldn't be higher. Simply put, if customers are inconvenienced by not being able to visit a retail location, access a Website, email a sales rep, or receive requested information in a timely manner-regardless of the reason-they will move on to a competitor. The availability of services and products at headquarters and in branch offices is absolutely vital to the health of the company, making quick restoration of the IT systems that run the business a top priority. Minutes do matter.
Most SMEs do not realize that quick restores not only rely on recovering lost data, but also rely on rebuilding damaged or lost servers and workstations. Operating systems need to be restored, drives need to be reformatted, business applications need to reinstalled, data needs to be recovered and user profiles need to be restored. End users cannot do their job unless workstations and servers are returned to their normal state before downtime occurred. This takes time.
The most effective disaster recovery solutions speed up the systems recovery process by automating this process through scripting. Automatic bare metal restore allows administrators to perform a complete systems restoration to an existing system or to a new system with different hardware. While it typically takes a seasoned IT veteran a full day or longer to completely restore a system manually, there are disaster recovery solutions that allow staffers with little IT training to recover a complete system in less than an hour.
Off-Site Backups and Vaulting
In an effort to protect against data center outages and regional natural disasters, it is essential that SMEs implement off-site backups and a remote vaulting system that store data and system information at least 150 miles away. Tape rotation at a warehouse across town, a common approach of SMEs, does not provide the same level of off-site protection since the facility will most likely be impacted by the same disruption. The best backup and recovery solutions incorporate both local and off-site backups in one package. After first performing a local backup in the branch office, the solution replicates data to an off-site location where it is stored in a data vault and is easily recoverable.
Because bandwidth and server requirements could affect performance for end users, it is essential that SMEs deploy a solution that incorporates electronic data synchronization to off-site vaults. By only sending data that is new or has changed since the last backup, the solution requires less resources and dramatically reduces the amount of data that has to be sent over the WAN. This allows backups to be completed more quickly, efficiently and reliably. In the same vein, synchronization technology enables administrators to pause and re-engage in the middle of a backup if a connection is disrupted during the data transfer, preventing administrators from ever having to manually restart a transfer.
Vendor Neutral
When a disaster hits and remote offices are pulled off-line, SMEs don't always have the luxury of being able to replace their servers with the exact model or even the same brand. Often, they must make a quick replacement with what is available. Because of this, it is important that the company's disaster recovery solution is able to restore systems to different hardware. For example, an HP server needs to be able to be restored on a Dell box if that is what is available. This also gives SMEs flexibility with their purchasing decisions, allowing them to choose hardware that is the best suited for specific business needs and budget rather than letting interoperability inhibit choice.
Point-in-Time Recovery
Giving customers the ability to restore systems to a set point in time is an essential element of a complete disaster recovery solution. This is especially useful if a single system crashes and a database needs to be synched up with other end users. Email is another example. Point-in-time recovery allows administrators to recover deleted emails or files as needed or restore systems to a point in time within the past 30 days.
Simple Management
Most SMEs do not have the IT and staffing resources to dedicate to backup management and monitoring, making it essential that any disaster recovery solution be easy to use and automate as much manual process as possible. This is especially true on the edge of the network in branch offices where often there is no full-time IT administrator and a business staff member is typically tasked with starting the backup process each night.
It doesn't have to be this way. SMEs can make sure their disaster recovery solution is managed consistently from a central management console, giving a trained and dedicated administrator at headquarters complete access and visibility into the backup cycle. In addition, the solution needs to automate basic tasks, including the recovery process. This can help streamline IT management, prevent human errors and take business continuity responsibility off the shoulders of untrained business staff.
Scalability
Complete restoration solutions also need to be scalable to allow SMEs to expand their disaster recovery strategy in line with the business. It seems strange to think that a company can over-protect its IT systems, but it is nearly impossible to protect every server all of the time. SMEs need to determine what systems need to be protected and how fast they need to be recovered. Deploying a scalable disaster recovery solution that is flexible is a good start, as growing companies' needs can change quickly.
Some providers price their solutions on a per gigabyte of user data basis in conjunction with time-to-recovery needs. This ensures that SMEs are getting exactly what they require at an affordable price and that the IT budget is directly tied to the needs of the business.
Security
Another feature that SMEs may need is encryption technology that secures data as it is in transit and while being backed up and restored. Industry and government regulations are getting tighter every year, forcing companies-especially growing businesses-to plan for future compliance requirements, making it essential that all sensitive data is properly encrypted from the outset.
It is vital that SMEs take into account remote office computing when implementing their business continuity strategy, relying on a robust and complete disaster recovery solution that enables efficient backups and quick restores. Administrators need to ensure that they are able to manage the backup and recovery process from a central location, conduct complete restorations and recover systems to different hardware. Only then can SMEs be truly prepared for whatever Mother Nature throws their way.
Revenue Opportunity for Systems Integrators
While the need to back up remote data and branch office systems off site exists, SMEs don't always have the resources to deploy and maintain a remote disaster recovery site that is distant enough to provide protection from regional disasters. Systems integrators who already maintain large, centralized data centers can fill this niche. Given the right tools, systems integrators can establish a disaster recovery service that backs up data locally each night and replicates that data to an off-site disaster recovery site managed by a knowledgeable, dedicated IT staff.
The key to implementing a successful off-site vaulting managed service is finding the right solution to make remote backups efficient enough that bandwidth and storage costs don't cancel the margins. As stated in this article there are reliable disaster recovery solutions specifically designed for remote SME environments. In addition to the requirements discussed in the main article, make sure the solution leverages block-level backups, only backing up data that has been changed or created since the last backup. In this way, you can eliminate redundant data going over the WAN, saving bandwidth and shortening the backup cycle.
Managing customers' off-site vaulting in an efficient and reliable manner is a service that could prove highly profitable for systems integrators, especially when they already possess most of the resources required to offer the service. It not only generates additional revenue and increases margins but positions them as a full-service disaster recovery service provider and closely ties their business with the needs of their customers.
Mark Phillippi is vice president of technology and product management for Unitrends.
www.unitrends.com
COPYRIGHT 2007 West World Productions, Inc.COPYRIGHT 2007 Gale Group
The 2007 Atlantic hurricane season started early this year, the second time in five years a storm has formed before June 1. As a result, experts are again predicting an above average hurricane year, estimating that there is a 74 percent chance a category three or higher storm will hit the U.S. mainland.
While the chance your business will be affected by a major natural disaster or terrorist attack is slim, other causes of unplanned downtime such as a hardware component failure, fire, power outage or surge, network attack or even human error are more prevalent. It is important to ensure your company has a reliable strategy for getting the business up and resuming normal business operations, no matter what the cause. Without a disaster recovery strategy the consequences to your business could be paramount. According to the Wall Street Journal, 92 percent of small businesses that experience significant data loss due to a major disaster go out of business within five years. You don't want to become part of this statistic.
With mission-critical data growth exploding in the small to midsized enterprise (SME) market-much of it on the edge of the network in branch offices-traditional disaster recovery solutions may not be enough to ensure business continuity throughout your entire organization.
Complete Restoration Disaster Recovery
In order to protect all company data regardless of physical location and to ensure business continuity across the entire organization, SMEs need to deploy a complete disaster recovery solution that integrates local backups and remote off-site vaulting in a single solution. Using a complete and integrated solution allows your organization to restore entire systems quickly, and get them back in the production environment as soon as possible after experiencing downtime. A complete bare metal restoration solution restores systems from the ground up-including the operating system (Windows, Linux, Solaris, etc.), business applications (Exchange, SQL, Oracle, Novell, etc.), user profiles and data files, in addition to reformatting drives.
The following details what SMEs should look for in a complete disaster recovery solution that protects their data center as well as data and systems in remote offices.
Quick Recovery
When it comes to quick recoveries and an SME's ability to resume normal business operations after experiencing downtime, the stakes couldn't be higher. Simply put, if customers are inconvenienced by not being able to visit a retail location, access a Website, email a sales rep, or receive requested information in a timely manner-regardless of the reason-they will move on to a competitor. The availability of services and products at headquarters and in branch offices is absolutely vital to the health of the company, making quick restoration of the IT systems that run the business a top priority. Minutes do matter.
Most SMEs do not realize that quick restores not only rely on recovering lost data, but also rely on rebuilding damaged or lost servers and workstations. Operating systems need to be restored, drives need to be reformatted, business applications need to reinstalled, data needs to be recovered and user profiles need to be restored. End users cannot do their job unless workstations and servers are returned to their normal state before downtime occurred. This takes time.
The most effective disaster recovery solutions speed up the systems recovery process by automating this process through scripting. Automatic bare metal restore allows administrators to perform a complete systems restoration to an existing system or to a new system with different hardware. While it typically takes a seasoned IT veteran a full day or longer to completely restore a system manually, there are disaster recovery solutions that allow staffers with little IT training to recover a complete system in less than an hour.
Off-Site Backups and Vaulting
In an effort to protect against data center outages and regional natural disasters, it is essential that SMEs implement off-site backups and a remote vaulting system that store data and system information at least 150 miles away. Tape rotation at a warehouse across town, a common approach of SMEs, does not provide the same level of off-site protection since the facility will most likely be impacted by the same disruption. The best backup and recovery solutions incorporate both local and off-site backups in one package. After first performing a local backup in the branch office, the solution replicates data to an off-site location where it is stored in a data vault and is easily recoverable.
Because bandwidth and server requirements could affect performance for end users, it is essential that SMEs deploy a solution that incorporates electronic data synchronization to off-site vaults. By only sending data that is new or has changed since the last backup, the solution requires less resources and dramatically reduces the amount of data that has to be sent over the WAN. This allows backups to be completed more quickly, efficiently and reliably. In the same vein, synchronization technology enables administrators to pause and re-engage in the middle of a backup if a connection is disrupted during the data transfer, preventing administrators from ever having to manually restart a transfer.
Vendor Neutral
When a disaster hits and remote offices are pulled off-line, SMEs don't always have the luxury of being able to replace their servers with the exact model or even the same brand. Often, they must make a quick replacement with what is available. Because of this, it is important that the company's disaster recovery solution is able to restore systems to different hardware. For example, an HP server needs to be able to be restored on a Dell box if that is what is available. This also gives SMEs flexibility with their purchasing decisions, allowing them to choose hardware that is the best suited for specific business needs and budget rather than letting interoperability inhibit choice.
Point-in-Time Recovery
Giving customers the ability to restore systems to a set point in time is an essential element of a complete disaster recovery solution. This is especially useful if a single system crashes and a database needs to be synched up with other end users. Email is another example. Point-in-time recovery allows administrators to recover deleted emails or files as needed or restore systems to a point in time within the past 30 days.
Simple Management
Most SMEs do not have the IT and staffing resources to dedicate to backup management and monitoring, making it essential that any disaster recovery solution be easy to use and automate as much manual process as possible. This is especially true on the edge of the network in branch offices where often there is no full-time IT administrator and a business staff member is typically tasked with starting the backup process each night.
It doesn't have to be this way. SMEs can make sure their disaster recovery solution is managed consistently from a central management console, giving a trained and dedicated administrator at headquarters complete access and visibility into the backup cycle. In addition, the solution needs to automate basic tasks, including the recovery process. This can help streamline IT management, prevent human errors and take business continuity responsibility off the shoulders of untrained business staff.
Scalability
Complete restoration solutions also need to be scalable to allow SMEs to expand their disaster recovery strategy in line with the business. It seems strange to think that a company can over-protect its IT systems, but it is nearly impossible to protect every server all of the time. SMEs need to determine what systems need to be protected and how fast they need to be recovered. Deploying a scalable disaster recovery solution that is flexible is a good start, as growing companies' needs can change quickly.
Some providers price their solutions on a per gigabyte of user data basis in conjunction with time-to-recovery needs. This ensures that SMEs are getting exactly what they require at an affordable price and that the IT budget is directly tied to the needs of the business.
Security
Another feature that SMEs may need is encryption technology that secures data as it is in transit and while being backed up and restored. Industry and government regulations are getting tighter every year, forcing companies-especially growing businesses-to plan for future compliance requirements, making it essential that all sensitive data is properly encrypted from the outset.
It is vital that SMEs take into account remote office computing when implementing their business continuity strategy, relying on a robust and complete disaster recovery solution that enables efficient backups and quick restores. Administrators need to ensure that they are able to manage the backup and recovery process from a central location, conduct complete restorations and recover systems to different hardware. Only then can SMEs be truly prepared for whatever Mother Nature throws their way.
Revenue Opportunity for Systems Integrators
While the need to back up remote data and branch office systems off site exists, SMEs don't always have the resources to deploy and maintain a remote disaster recovery site that is distant enough to provide protection from regional disasters. Systems integrators who already maintain large, centralized data centers can fill this niche. Given the right tools, systems integrators can establish a disaster recovery service that backs up data locally each night and replicates that data to an off-site disaster recovery site managed by a knowledgeable, dedicated IT staff.
The key to implementing a successful off-site vaulting managed service is finding the right solution to make remote backups efficient enough that bandwidth and storage costs don't cancel the margins. As stated in this article there are reliable disaster recovery solutions specifically designed for remote SME environments. In addition to the requirements discussed in the main article, make sure the solution leverages block-level backups, only backing up data that has been changed or created since the last backup. In this way, you can eliminate redundant data going over the WAN, saving bandwidth and shortening the backup cycle.
Managing customers' off-site vaulting in an efficient and reliable manner is a service that could prove highly profitable for systems integrators, especially when they already possess most of the resources required to offer the service. It not only generates additional revenue and increases margins but positions them as a full-service disaster recovery service provider and closely ties their business with the needs of their customers.
Mark Phillippi is vice president of technology and product management for Unitrends.
www.unitrends.com
COPYRIGHT 2007 West World Productions, Inc.COPYRIGHT 2007 Gale Group
Reflections on virtualization for the SMB
by Mark Ferelli
In both the server and storage industries, the potential for business in serving small to medium businesses can be substantial. For many vendors, it is uncharted territory and an appreciation of the pain points of the SMB is a key action item.
It is all too tempting to look at the SMB as just a smaller version of the enterprise data center. But the yawning difference between the SMB and the enterprise in terms of the delta in budgetary resources, staffing resources, legacy resources and the IT mission makes this vision difficult to sustain.
Where commonality does exist is in the requirement for the efficient and cost-effective use of servers and storage. Both the SMB and the enterprise are increasingly committed to cost reduction and consolidation in server and storage acquisition and operation. Consolidation in particular is popular, since it represents the use of less space, less electricity, and fewer hours and personnel committed to management of the assets.
To attain these lofty goals, the SMB can now look to virtualization as a strategy. Properly applied, virtualization reduces total cost of operation and boosts performance of the assets that have been virtualized.
Server Virtualization
For purposes of this discussion, virtualization comes in two flavors: server and storage. Server virtualization creates a level of abstraction permitting a single computer to act as several computers. It gives the user the impression that multiple servers exist on a single box. Each server session enjoys its own dedicated memory space, dedicated storage space and emulated processor.
There are two major applications offering server virtualization support. One is Microsoft Virtual Server 2005 R2 (For the desktop, Microsoft offers Virtual PC, which came to the Redmond giant with the acquisition of Connectix. This was the same implementation that allowed Mac systems to run DOS type sessions to achieve cross platform compatibility.) The other is VMware, the godfather of server virtualization applications running a broad series of operating systems.
Other major virtualization applications are available from companies such as SWsoft and XenSource, although they represent a small piece of the virtualization pie. There also are tools that can be used for deploying virtual servers, migrating servers and protecting servers aside from those offered by the virtualization vendors themselves, such as Acronis True Image and Acronis Full Circle. In addition to migrating from the physical to virtual environments, these tools also can be used to migrate from a virtual server to a dissimilar hardware-based virtual server or from a virtual server back to a physical server. Working together, the Acronis tools also can be used for backup and disaster recovery of virtual servers in an SMB and enterprise environment.
Most recently, VMware has taken aggressive steps to bring server virtualization to the SMB, focusing on that key SMB requirement: Ease of use. VMware Server, VMware's low-end virtual generator, has been free since July 2006. Some 70% of its 1.2 million downloads is reported to have gone to small or medium-sized businesses. To satisfy that essential ease of use requirement, VMware is offering VirtualCenter for VMware Server.
VirtualCenter is a single management console for configuring, provisioning, and managing virtual machines. It includes a high-availability module and a VMotion virtual machine module that migrates from one physical server box to another. It also includes a dynamic resource allocation module for assigning memory, CPU duty cycles, and network bandwidth between virtual machines on a pool of physical servers. The SMB version of VirtualCenter is restricted to the VMware Server, an older form of virtual server technology that translates virtual machine needs through the physical server's operating system.
Storage Virtualization
Storage virtualization is well defined in Tom Clark's definitive work Storage Virtualization: "the logical abstraction of physical storage systems and thus, when well implemented, hides the complexity of physical storage devices and their specific requirements from management view." Even a basic understanding of virtualization delivers both data center manager and SMB multiple benefits: a better handle on downtime, better general storage performance, fewer problems of lost data, and the ability to aggregate storage devices, delivering best use of storage assets.
The need for storage virtualization appears, as a rule, when it is necessary for the SMB to implement a storage area network (SAN). This determination is often a function of the amount of data to be stored. If storage requirements require more than 2 TB, a SAN is widely considered by IT professionals to be the preferred solution to direct- or network-attached storage. This is because SANs allow storage scaling over time and provides such benefits as high availability, high performance, scalability and ease of management.
Virtualization in a SAN involves abstracting the logical (or virtual) presentation of stored data from the physical disk asset where the data actually rests. The result is a more efficient use of storage assets. Data could migrate from one device to another in the background while the virtual presentation of the storage remains constant
Storage virtualization is anything but new, even if the phrase is new to the SMB. The reorganization of storage in a RAID array with data striping onto multiple drives is a type of virtualization. Virtual tape was pioneered by IBM many years ago. In fact, the most basic type of virtualization is disk partitioning, where multiple partitions appear as separate disks to the operating system.
Why Virtualize?
The SMB can look at virtualization as a route to achieve a variety of benefits in terms of storage management, many of which have been discussed here. But taking a long view, virtualization is a reliable route to the development of a utility infrastructure, where server processing and storage are managed as resources similar to the way electricity and water are managed for homes and businesses.
Utility IT infrastructure is highly responsive and makes the best use of available hardware investments. Take, for example, the need to provide IT support for a new business process. In a conventional storage networking environment, satisfying this need will likely require the purchase of servers and either internal disks or an attached array to house the applications and data that this new process will require.
In a utility infrastructure, the new application is provisioned with processing power, and storage space is allocated from a pool of resources within the structure. The application is housed in a virtual machine and the data in support of the new business process is stored on a virtual storage volume. In cases like this one, any new capital expenditure for servers and disk storage will be commodity purchases made to increase the size of the processing and storage pool, not specifically to support the new business process.
Virtualization is an enabling technology for a utility infrastructure. Virtualization is the means by which the processing and storage are abstracted and managed separately from the underlying hardware layer.
From the point of view of server and storage management, resource allocation and IT support for SMB applications, virtualization makes a great deal of sense. Once the SMB gets over the intimidation of the V-word, the only challenge will be to choose whose utility infrastructure will best solve prevailing IT/business problems.
Mark Ferelli, formerly the chief editor of Computer Technology Review, is a freelance writer and storage specialist based in North Hollywood, CA. He can be reached at mcferelli@yahoo.com.
COPYRIGHT 2007 West World Productions, Inc.COPYRIGHT 2007 Gale Group
In both the server and storage industries, the potential for business in serving small to medium businesses can be substantial. For many vendors, it is uncharted territory and an appreciation of the pain points of the SMB is a key action item.
It is all too tempting to look at the SMB as just a smaller version of the enterprise data center. But the yawning difference between the SMB and the enterprise in terms of the delta in budgetary resources, staffing resources, legacy resources and the IT mission makes this vision difficult to sustain.
Where commonality does exist is in the requirement for the efficient and cost-effective use of servers and storage. Both the SMB and the enterprise are increasingly committed to cost reduction and consolidation in server and storage acquisition and operation. Consolidation in particular is popular, since it represents the use of less space, less electricity, and fewer hours and personnel committed to management of the assets.
To attain these lofty goals, the SMB can now look to virtualization as a strategy. Properly applied, virtualization reduces total cost of operation and boosts performance of the assets that have been virtualized.
Server Virtualization
For purposes of this discussion, virtualization comes in two flavors: server and storage. Server virtualization creates a level of abstraction permitting a single computer to act as several computers. It gives the user the impression that multiple servers exist on a single box. Each server session enjoys its own dedicated memory space, dedicated storage space and emulated processor.
There are two major applications offering server virtualization support. One is Microsoft Virtual Server 2005 R2 (For the desktop, Microsoft offers Virtual PC, which came to the Redmond giant with the acquisition of Connectix. This was the same implementation that allowed Mac systems to run DOS type sessions to achieve cross platform compatibility.) The other is VMware, the godfather of server virtualization applications running a broad series of operating systems.
Other major virtualization applications are available from companies such as SWsoft and XenSource, although they represent a small piece of the virtualization pie. There also are tools that can be used for deploying virtual servers, migrating servers and protecting servers aside from those offered by the virtualization vendors themselves, such as Acronis True Image and Acronis Full Circle. In addition to migrating from the physical to virtual environments, these tools also can be used to migrate from a virtual server to a dissimilar hardware-based virtual server or from a virtual server back to a physical server. Working together, the Acronis tools also can be used for backup and disaster recovery of virtual servers in an SMB and enterprise environment.
Most recently, VMware has taken aggressive steps to bring server virtualization to the SMB, focusing on that key SMB requirement: Ease of use. VMware Server, VMware's low-end virtual generator, has been free since July 2006. Some 70% of its 1.2 million downloads is reported to have gone to small or medium-sized businesses. To satisfy that essential ease of use requirement, VMware is offering VirtualCenter for VMware Server.
VirtualCenter is a single management console for configuring, provisioning, and managing virtual machines. It includes a high-availability module and a VMotion virtual machine module that migrates from one physical server box to another. It also includes a dynamic resource allocation module for assigning memory, CPU duty cycles, and network bandwidth between virtual machines on a pool of physical servers. The SMB version of VirtualCenter is restricted to the VMware Server, an older form of virtual server technology that translates virtual machine needs through the physical server's operating system.
Storage Virtualization
Storage virtualization is well defined in Tom Clark's definitive work Storage Virtualization: "the logical abstraction of physical storage systems and thus, when well implemented, hides the complexity of physical storage devices and their specific requirements from management view." Even a basic understanding of virtualization delivers both data center manager and SMB multiple benefits: a better handle on downtime, better general storage performance, fewer problems of lost data, and the ability to aggregate storage devices, delivering best use of storage assets.
The need for storage virtualization appears, as a rule, when it is necessary for the SMB to implement a storage area network (SAN). This determination is often a function of the amount of data to be stored. If storage requirements require more than 2 TB, a SAN is widely considered by IT professionals to be the preferred solution to direct- or network-attached storage. This is because SANs allow storage scaling over time and provides such benefits as high availability, high performance, scalability and ease of management.
Virtualization in a SAN involves abstracting the logical (or virtual) presentation of stored data from the physical disk asset where the data actually rests. The result is a more efficient use of storage assets. Data could migrate from one device to another in the background while the virtual presentation of the storage remains constant
Storage virtualization is anything but new, even if the phrase is new to the SMB. The reorganization of storage in a RAID array with data striping onto multiple drives is a type of virtualization. Virtual tape was pioneered by IBM many years ago. In fact, the most basic type of virtualization is disk partitioning, where multiple partitions appear as separate disks to the operating system.
Why Virtualize?
The SMB can look at virtualization as a route to achieve a variety of benefits in terms of storage management, many of which have been discussed here. But taking a long view, virtualization is a reliable route to the development of a utility infrastructure, where server processing and storage are managed as resources similar to the way electricity and water are managed for homes and businesses.
Utility IT infrastructure is highly responsive and makes the best use of available hardware investments. Take, for example, the need to provide IT support for a new business process. In a conventional storage networking environment, satisfying this need will likely require the purchase of servers and either internal disks or an attached array to house the applications and data that this new process will require.
In a utility infrastructure, the new application is provisioned with processing power, and storage space is allocated from a pool of resources within the structure. The application is housed in a virtual machine and the data in support of the new business process is stored on a virtual storage volume. In cases like this one, any new capital expenditure for servers and disk storage will be commodity purchases made to increase the size of the processing and storage pool, not specifically to support the new business process.
Virtualization is an enabling technology for a utility infrastructure. Virtualization is the means by which the processing and storage are abstracted and managed separately from the underlying hardware layer.
From the point of view of server and storage management, resource allocation and IT support for SMB applications, virtualization makes a great deal of sense. Once the SMB gets over the intimidation of the V-word, the only challenge will be to choose whose utility infrastructure will best solve prevailing IT/business problems.
Mark Ferelli, formerly the chief editor of Computer Technology Review, is a freelance writer and storage specialist based in North Hollywood, CA. He can be reached at mcferelli@yahoo.com.
COPYRIGHT 2007 West World Productions, Inc.COPYRIGHT 2007 Gale Group
Secure edge messaging appliances with embedded LDAP maximize email protection by Mike Dodson
It's no wonder that email is such a big thorn in IT's side. Mail volume doubles every six to nine months; an astonishing 80-90 percent of it unwanted and/or abusive. Four hundred and fifty new viruses attack companies every month. Eighty-five percent of abusive mail is sent by a zombie computer, with 500,000 new zombies launched daily according to Commtouch. Directory Harvest Attacks (DHAs), meanwhile, are increasing by as much as 30 percent annually. Not surprisingly, managing email and email security is getting more complex and more costly. Or is it?
The typical organization uses messaging software that runs on dedicated servers. In companies that use anti-virus and anti-spam software, IT managers often spend hours or days trying to get email server and email security software to work together. Upgrades are incremental and expensive and can cause server-to-server and other incompatibilities that disrupt email service. To continue protecting the growing message volume, companies must add new servers regularly in a virtually endless cycle of budget-and-build.
More importantly perhaps, standalone email security taxes bandwidth and does not provide adequate protection. Typically, email security is architected as a separate layer in front of the messaging firewall. While this method simplifies deployment, it often requires additional connections through the firewall, which can burden the messaging infrastructure with unanticipated load and vulnerabilities. In this architecture, recipient validation and per-recipient policy enforcement requires reaching through the firewall to directly query the corporate directory. Thus a breach in email security would leave organizations vulnerable to theft of user IDs and passwords and even HR and financial data.
Every surge in email traffic from spam or virus attacks places load demands on the corporate firewall and corporate directory. This in turn causes a loss in quality of service of the corporate network and mission-critical applications that depend on the corporate directory. In the worst case, if the directory is not sized to handle email traffic spikes, a spam attack could halt user access to valuable internal applications such as a CRM program.
To maintain quality of service, companies must continually add capacity to the corporate directory, firewall and other network components to accommodate the ever-increasing volume of spam. Or they can find an alternative architecture.
[FIGURE 1 OMITTED]
Integrated Protection at the Edge
According to IDC ("Messaging Solutions with 'Baked In' Security Delivers for IT and Messaging Users"), appliances that integrate messaging and security make better sense. A secure appliance is inexpensive, requires minimal administration, and blocks up to 80 percent of unwanted email at the firewall, reserving network bandwidth for email you do want. All essential messaging and security features are combined in a single box-from antivirus and anti-spam protection to integrated LDAP for recipient verification at the edge, to reputation filtering and user policy controls. Finally, appliance system upgrades can be made locally without removing the appliance, with no risk of incompatibility because security and messaging features reside on one platform.
Reputation filters add another layer of protection. Most reputation filters are a global one-size-fits-all, allowing for little or no customization to bar unwanted traffic from entering the network. But many organizations find these static filters impractical. Flexible reputation filters, on the other hand, not only take into account the sender's reputation, but also the recipient's level of sensitivity to missing a legitimate message that may have come from a sender with a bad reputation.
Flexible-reputation filters based on SMTP etiquette and RFC compliance are uniquely effective at blocking zombies and other sources of spam because they can block new abusers with no reputation history. It also provides a clear path for legitimate senders.
Spammers and hackers will continue finding new, more devious ways to use email to wreak havoc on corporations. A secure messaging appliance with an embedded LDAP server and flexible reputation filters at the network edge is our best defense.
Mike Dodson is the director of security strategy for Mirapoint. To access the IDC whitepaper, "Messaging Solutions with 'Baked In' Security Delivers for IT and Messaging Users," visit Mirapoint's Web site at http://www.mirapoint.com/idc.
COPYRIGHT 2007 West World Productions, Inc.COPYRIGHT 2007 Gale Group
The typical organization uses messaging software that runs on dedicated servers. In companies that use anti-virus and anti-spam software, IT managers often spend hours or days trying to get email server and email security software to work together. Upgrades are incremental and expensive and can cause server-to-server and other incompatibilities that disrupt email service. To continue protecting the growing message volume, companies must add new servers regularly in a virtually endless cycle of budget-and-build.
More importantly perhaps, standalone email security taxes bandwidth and does not provide adequate protection. Typically, email security is architected as a separate layer in front of the messaging firewall. While this method simplifies deployment, it often requires additional connections through the firewall, which can burden the messaging infrastructure with unanticipated load and vulnerabilities. In this architecture, recipient validation and per-recipient policy enforcement requires reaching through the firewall to directly query the corporate directory. Thus a breach in email security would leave organizations vulnerable to theft of user IDs and passwords and even HR and financial data.
Every surge in email traffic from spam or virus attacks places load demands on the corporate firewall and corporate directory. This in turn causes a loss in quality of service of the corporate network and mission-critical applications that depend on the corporate directory. In the worst case, if the directory is not sized to handle email traffic spikes, a spam attack could halt user access to valuable internal applications such as a CRM program.
To maintain quality of service, companies must continually add capacity to the corporate directory, firewall and other network components to accommodate the ever-increasing volume of spam. Or they can find an alternative architecture.
[FIGURE 1 OMITTED]
Integrated Protection at the Edge
According to IDC ("Messaging Solutions with 'Baked In' Security Delivers for IT and Messaging Users"), appliances that integrate messaging and security make better sense. A secure appliance is inexpensive, requires minimal administration, and blocks up to 80 percent of unwanted email at the firewall, reserving network bandwidth for email you do want. All essential messaging and security features are combined in a single box-from antivirus and anti-spam protection to integrated LDAP for recipient verification at the edge, to reputation filtering and user policy controls. Finally, appliance system upgrades can be made locally without removing the appliance, with no risk of incompatibility because security and messaging features reside on one platform.
Reputation filters add another layer of protection. Most reputation filters are a global one-size-fits-all, allowing for little or no customization to bar unwanted traffic from entering the network. But many organizations find these static filters impractical. Flexible reputation filters, on the other hand, not only take into account the sender's reputation, but also the recipient's level of sensitivity to missing a legitimate message that may have come from a sender with a bad reputation.
Flexible-reputation filters based on SMTP etiquette and RFC compliance are uniquely effective at blocking zombies and other sources of spam because they can block new abusers with no reputation history. It also provides a clear path for legitimate senders.
Spammers and hackers will continue finding new, more devious ways to use email to wreak havoc on corporations. A secure messaging appliance with an embedded LDAP server and flexible reputation filters at the network edge is our best defense.
Mike Dodson is the director of security strategy for Mirapoint. To access the IDC whitepaper, "Messaging Solutions with 'Baked In' Security Delivers for IT and Messaging Users," visit Mirapoint's Web site at http://www.mirapoint.com/idc.
COPYRIGHT 2007 West World Productions, Inc.COPYRIGHT 2007 Gale Group
Langganan:
Postingan (Atom)