Using traditional methods to migrate applications and data to the cloud is typically time-consuming because production servers may have to be brought to a standstill in order for the migration to be executed.  You also have to keep in mind your storage footprint - the larger the storage footprint, the longer the migration process will take.  In fact, large system migrations can typically take hours or days to complete.

The cloud has become increasingly popular as a platform to host business applications because it offers a very cost-effective and flexible IT platform for many organizations.  But how do you get your existing applications and data into the cloud?  That’s a question we are frequently asked here at iland.

One way is to migrate your data from a file server using traditional data copying tools but you need to ensure that users cannot change any data while the migration is taking place otherwise changes to files that have already been copied will be lost.

You may also need to create a plan for recreating the file shares, permissions, compression, encryption and other settings on the new storage. This can translate into further hours of downtime even if everything works as planned the first time—and much longer if it doesn't.

At first glance it may seem daunting but don’t worry - there are simple ways you can overcome these challenges and perform data migrations without disrupting production applications and shutting down their servers.

Double-Take Move and Double-Take Availability are migration and data replication tools that allow you to migrate physical and virtual workloads with complete data protection.  Leveraging Doubletake replication not only allows you to address the migration process for moving applications and data into the cloud. It also offers a way to automatically avoid or correct any migration errors that may occur.

Doubletake replication offers exceptional application and data protection/migration in virtual and cloud environments.  Thanks to a single unified management console it is easier to use than ever before.

Six Reasons Why Double-Take Availability Provides Superior High Availability

1. Simplicity - Set It and Forget It data protection includes advanced features that automate the initial setup and ongoing management of your availability environment.
2. Plug and Play Protection - Double-Take’s pre-configured Universal Virtual Recovery Appliance for physical and virtual environments gives you a plug and play option for extending the protection of your virtualized environment.
3. Real-time Replication – Offers you superior protection compared to snapshots because it ensures the uninterrupted availability of critical application systems and virtually eliminates the potential for data loss.
4. Zero Downtime during Migrations - Replication technology allows users to remain active while Double-Take Availability is recreating the production environment on the new system.
5. Platform Independence - Double Take Availability allows you to cost-effectively protect more of your environment.
6. Simplified Management - A unified console makes it easier to implement, manage and extract value from a powerful High Availability environment.

The following steps detail how to configure the Doubletake replication/migration process:

Step 1
Doubletake Replication Process

Data Sync Tabs
Always select these settings when setting up a replication set:

• Start, Programs, Double-Take, Double-Take replication Console
• Double-click over SOURCE server and login with local admin accounts
• Right-click over SOURCE server and select, New, Replication Set from the menu
• Name the replication set “sourceservername_newservername-rep”
• Double-Click replication set and select files and folders stated in the Sync/move package

Right-click over replication set and select “Save”

Step 2
Right-click over Replication Set and select “Connection Manager”

Step 3
Select the “Mirroring” tabs and make sure the following is selected

Step 4

• Click the “Servers” tab and select the target
• If a direct sync job do a one-to-one mapping
• If the move package/sync job states to copy to another directory, select and “All to One”

(NOTE: Start Mirror on connect and Start replication on connection are marked off by default)

Step 5

• Select “Connect” and the replication job will start
• The job can be viewed from the SOURCE server

When job the Mirror status states “Idle”, the Disk Queue states “0” and the Replication Status reads “Ready” the job is complete

Disconnect the replication set to break the connection/ complete the job

Doubletake replication through the use of Double-Take Move and Double-Take Availability offers a simple and convenient means to migrate data and applications to the cloud.

Historically, disaster recovery (DR) solutions have been difficult and costly to implement. DR has been a pain point for many IT administrators who have to maintain consistency across duplicate hardware in various locations, document and maintain detailed run books, react to changes in the environment, and schedule testing that is classically disruptive to the production environment.  And replication solutions are often time-consuming.  They require you to back up data and manually transport the back-ups to a different location.

iland offers a number of cloud-based DR solutions that dispel  many of these traditional DR issues.  One of these solutions leverages the Dell EqualLogic PS Series built-in Auto-Replication feature to make disaster recovery a rapid, manageable, reliable, and affordable process. Through EqualLogic’s Snapshot and Auto-Replication capability in the iland cloud, end-to-end data protection is now possible.

At the customer site EqualLogic SAN snapshots offer quick recovery based on:

• Volume changes
• Snapshot schedule
• Risk
• Bandwidth
• Need to recover

In the iland cloud:

• Critical volumes or possibly all volumes are replicated
• A secondary site is available for operation if the customer site fails.

The EqualLogic replication service is performed between the primary customer site and the iland cloud.

To set up an EqualLogic replication service in the iland cloud, iland sets up a site-to-site VPN tunnel between your site and the iland cloud. These two PS Series groups are then configured as replication partners and iland provides you with your delegated space for replication. The volumes designated for DR are selected allowing you to push your data to iland’s SAN infrastructure.

As shown in the diagram above, once the volumes’ replication is completed, the volumes become replicas in the remote site.  Replicas are similar to snapshots in that they represent the contents of the volume at a specific point in time.  And replicas ensure that the iland group always preserves a complete and stable copy of the volume data.

The initial push of data to the iland cloud can be challenging if the replication partnership occurs over a slow data link.  EqualLogic provides its Manual Transfer Utility to save the initial replica image to external media, physically ship it to iland and restore the image from the external media.

Now that we have discussed how EqualLogic replication is configured, what happens when a real DR event occurs? iland provides two  alternatives for that possibility:

EqualLogic Array-Based Replication only:
This option provides a cost effective offsite data storage with an option to have VMware-based images (4.x or higher) powered on in the event of a disaster.

EqualLogic Array-Based Replication with Standby Managed Cloud Resources:
This option provides cost effective offsite data storage with guaranteed resources for VMware-based images (4.x or higher).

In disaster recovery situations, iland takes over from the primary site.  When the primary site is ready to resume its original role, iland uses replica failback to efficiently synchronize the changed data on the secondary site with the primary site.   Unlike many cloud providers, iland gives you the ability to test your DR solution once a year for free.  You also have the ability to test more than once a year if you require.

An EqualLogic replication service in the iland cloud offers a very effective and low-cost solution for your disaster recovery strategy.

Disaster recovery has become, well… a disaster, depending on who you talk to. There are numerous details that have to be considered prior to DR implementation: RTO (recovery time objective), RPO (recovery point objective), what to back up, retention policies, and perhaps most importantly, how the DR environment will be accessed by end users.  In other words, how will your end users be affected?

Even after all of this has been determined, a company must make yet another important decision - how do we make it happen?  Some organizations have the luxury of multiple datacenters at their disposal, large IT departments, and the expertise to implement their strategies in-house. But the majority of companies simply don’t have the staff, facilities, or experience required to attain their DR goals.

By leveraging VMware vCloud Director, iland provides a cloud-based DR solution that makes implementation, deployment, and the management of the DR plan easier—IT staff are able to configure their security layers and deploy virtual servers in minutes. vCloud Director is essentially a Web-based, self-service view of the cloud. It includes 10 Gigabit Ethernet connections to multiple Tier 1 providers, enterprise firewall services, and role-based access control through the integration of your directory services.

Because you have control over your DR environment, you can fine tune it to make sure it meets your business’s standards and test it as frequently as you want.  Ultimately, you control your end users’ experience.

The takeaway here?  The implementation and management of your DR plan doesn’t have to be a strain, nor do your customers have to take the brunt of unnecessary or unplanned downtime. vCloud Director is an effective tool that’s already preparing businesses (of all sizes) for natural and manmade disasters without having their customers feel the effects. iland and vCloud Director are changing things for the better.

This week I want to cover the three major hurdles companies face when planning, implementing and executing a disaster recovery (DR )plan.

• How do I get the initial data and subsequent differentials over to iland?

• How do my end-users access the environment at time of failover?

• How do I get my data back after failover?

There are many variables to consider for each of these questions but for the sake of this blog we will try to focus on common elements.  There are three main factors to consider when determining the feasibility and success of a long term DR plan:

• The actual amount of bandwidth (specifically upload) available for the replication of data between sites.    

• The change rate that occurs on guests you wish to replicate.

• The frequency of replication you want to achieve.

Bandwidth – this is of huge importance when considering a disaster recovery site.  The amount of available upload speed can show you the maximum amount of data that you can push across the WAN in a single day.  (Note: a quick and easy calculation is to take the amount of upload speed available in Mbps and multiply by 9; this will show you your maximum data transfer per day in GB i.e. 10 Mbps would allow for about 90GB of data transferred in a day).

Change rate - the change rate in your environment can typically be captured if you are already performing backups or snapshots in your local environment.  This change rate can help you determine the actual amount of data you will need to replicate between sites to keep your systems synchronized. 

Replication frequency - the frequency of replication can help you factor in the multiplier you need to use on your change rate.  (If you are measuring your change rate per day but plan on replicating every hour, you will need to factor in the increase in replication this will trigger because the same blocks on a server may change between every replication period rather than once daily).

Even with sufficient replication speed to accommodate your change rate, you may have an initial amount of data that seems unfeasible to replicate across the WAN.  To overcome this for all iland managed DR products, iland ships you an encrypted drive for the initial seeding of data. Once iland imports this data, most customers can simply begin differential replication which greatly expedites the deployment process.

Access to data at the iland site during a failover test or actual DR event can vary according to customer needs.  For some customers the majority of their environment is internal facing, while others may rely on WAN-based connectivity for all their servers.  Whatever the customers’ requirements, iland provides solutions that support IPsec Site-to-site VPN, SSL and Client VPN, Terminal services, app streaming, NAT and access-list configuration, and the ability for a customer to tie in an existing MPLS or point-to-point circuit. Many iland managed DR solutions allow customers to keep their existing internal subnet assigned to the machines replicated, allowing for faster failover and seamless integration with secondary sites.

Getting data back after a DR event can vary based on the severity of the customer outage.  For customers that have a site failure but all their data is still intact, iland DR solutions support failback options designed to simply replicate back the changes that have occurred between failover and cutback.  For customers that lose all local data, iland can support the initial seeding of data back to their site using the method outlined for initial seeding above.

Regardless of a customer’s DR goals, iland has a variety of flexible, cost-effective options available to ensure our customers are protected.

If you’re considering implementing a cloud-based disaster recovery plan for your organization, there are a number of items you need to think about with respect to what you currently have in place and what you may need to add. Employing a cloud environment for your DR can’t be done on a whim. It requires careful thought, so here’s a checklist to help get you started and hopefully make your cloud-based discussions with both your internal teams and cloud-based providers a more productive experience:

Identify Your Current Infrastructure Environment
How your current IT infrastructure environment is configured is an important piece of your DR plan and you’ll need to ask yourself these questions:
* What kind of storage do we currently have?
* What systems do we currently have in place?
* What is our mix of physical and virtual machines?
* Have we already tiered our applications?  If not, do we need to? 
* In the event of a disaster, do we want instantaneous failover or are we comfortable with some downtime?  If the latter, how much downtime?

Identifying these areas of your IT environment gives you a solid base from which to start planning your cloud-based DR implementation. 

Determine Your Bandwidth
Bandwidth costs have seen considerable reductions over the past few years but the amount of bandwidth you currently have will have a direct effect on the speed, recovery time and amount of data you want stored in a cloud environment.  You need to ask yourself:
* What bandwidth do we currently have available for upload?
* What RPO do we want? 

And for those that already have a DR plan in place, what sort of change rate are you seeing in your environment?  If your company has recently adopted an aggressive growth strategy you may be looking at much larger bandwidth requirements.

Other Considerations
Executive Team Buy-In - do you have buy-in from your executive management team?  To encourage their support of your DR planning, ask them what your company would do with none of its data and systems available.  Would it survive?  Customers today are used to instant gratification - would they really stay with you if you’re down for any length of time?  Or would they simply switch to your closest competitor?

Budget – be realistic with your DR plan when it comes to budget.  In our experience, most organizations start their DR discussions wanting immediate failover and recovery – until they find out the true cost.  At that point they’ll determine what they really need and what they can manage without depending on the budget available.  iland is adept at providing appropriate cloud-based DR solutions that fit within budgets  for our customers.

Testing – how are you going to verify that your DR plan works?  The only surefire way is to test it.  Cloud-based DR has made testing much easier to execute so we strongly recommend an annual DR test to find out what works and what doesn’t.  iland offers a complementary annual DR test.

Unfortunately, a well-conceived, cloud-based DR plan comes too late for the majority of companies that don’t have one when disaster strikes.  There are a few lucky businesses that survive downtime during a disaster but they usually conclude that an effective DR plan isn’t just a nice-to-have.  It’s a necessity.  And that’s when iland gets the call…..
 

Flexibility and agility—they’re the two attributes of a virtual environment on the VMware® platform that keep businesses functioning at peak capacity. If you’re already using VMware you’ll be pleased to know that iland offers a VMware hybrid cloud, making it possible for you to move your data into the cloud and manage it from the same VMware platform you’re already using. You might be wondering why anyone would want to transition data to a new cloud environment, other than for the simple management of virtual environments. Well, instead of having to buy expensive hardware, licensing, and storage to account for anticipated business growth, you can quickly spin up virtual machines whenever you need them.  

Think about it. Overprovisioning for seasonal demands, special projects and even testing and development has slowed down business growth for years. With VMware-based, vCloud Powered services, the time and money you would have spent setting up new servers, SANs, licensing, cooling equipment, etc. could be used for future innovation or other initiatives that could propel your business forward. Better still, you can access all your resources from one, unified platform. 

With iland Cloud Services, you can leverage the same security and performance you’ve been using versus having to convert your machines to a proprietary public cloud platform. These attributes are essential to a reliable disaster recovery solution—additionally customers can subscribe to backup offerings, making the recovery from an unplanned disaster easier and more efficient. 

With a hybrid cloud, you stay in complete control while enabling a self-service provisioning model for end users. You can set up administrative permissions for others within your organization or integrate your directory services and if you house a portion of your data in the iland cloud, you’ll have greater control over your security and control the properties of the environment locally.  

In a nutshell, IT departments and solution providers who are utilizing VMware can feel secure about extending their virtual environments into the cloud with vCloud hosting provider iland, maximizing their existing investments, and even extend their brand by white labeling the iland cloud.

Flexibility and agility. They’re the two attributes essential to a streamlined, cost-conscious, and progressive business.

I speak to people on a daily basis about the importance of Disaster Recovery; not simply backup or replication, but an actual plan that will fully cut over end users from one environment to another in a way that allows the data being protected to be used in a production like manner. 

The copying of data itself is normally one of the more straight-forward parts of a replication or backup plan.  There are hundreds of vendors that offer image level, agent, array, and host based options along with numerous features to reduce bandwidth or utilize storage space. iland has partnerships with a number of vendors we consider to be the best in these fields, but simply having the data at a secondary location is not a disaster recovery plan.  

An important consideration with all disaster recovery planning is end user access and usage patterns in production environments and how this translates to disaster recovery. For example, a company with a high percentage of web applications may need to consider their goals with respect to public IP utilization, DNS cutover, geographic location of endusers, load balancers, and bandwidth usage.  A different company with primarily internal systems may have completely different needs that are more focused on IPsec VPN connectivity for secondary sites, terminal services, Virtual Desktops, and WAN accelleration.  

Each customer environment is unique when it comes to the challenge of creating a seamless end user experience, but sometimes the best way to improve is to find out firsthand what you don't know. 

Every iland Standby or Live Disaster Recovery product comes with a Recovery Action Plan and a complimentary annual test. 

The Recovery Action Plan is the living document between you and iland that allows you to document key parts of your failover process and have a clear understanding of the responsibilities of iland, your company, and your end users for a successful failover.  

Of equal importance, an annual test (which can be performed non-intrusively if desired) can help your team determine the parts of your Recovery Action Plan that need changing to better enhance the user experience, or better protect the data being replicated.

Disaster Recovery requires planning for success. The team at iland knows that successful DR is a moving target that requires fine tuning on a regular basis. With enterprise class facilities and staff, and expertise in a number of disaster recovery methods we are here to help.

Location, Location, Location.  What holds true for real estate is also true for technology.  Access to data is the top priority of IT staff, and whether you look at production resources with low latency performance or disaster recovery with geographic diversity, location matters.

Production and development sites can benefit from having resources in a specific geographic region for a number of reasons.

• Lower latency: having cloud resources in a region with close proximity to your end users can reduce network latency and provide a favorable experience for remote desktop, IPsec VPN, and of course IP telephony.
• Carrier availability: having cloud resources in a specific region gives organizations more flexibility in regards to network carriers.  For example, iland leverages top tier providers such as Level 3 and AboveNet (in addition to others) for backbone connectivity, but it is not uncommon for an organization to desire point-to-point or MPLS for WAN connectivity.  This can be especially helpful if your organization leverages a regional carrier and you wish to tie them directly into iland Cloud resources.
• Mixed cloud: an experienced cloud provider should be able to offer customers the ability to co-locate physical servers/network gear and leverage them on the same LAN segment as their cloud resources.  Having the option of multiple datacenter locations allows greater flexibility for the implementation of non-virtual resources into an existing or new cloud deployment.

Disaster recovery goals also benefit from different cloud locations. The primary benefit is the ability to select a location to use as a disaster recovery target that meets your organization’s goals with respect to how far away the data is kept in a second site, while still allowing you to select a secondary site that will ensure production performance capabilities in terms of latency and cloud capacities should you need to failover. 

This benefit applies to production customers as well as customers leveraging a private cloud looking for disaster recovery options.  For example, customers that leverage iland for both production and disaster recovery can benefit from multiple cloud offerings with managed disaster recovery options that allow the customer to have their data replicated to multiple iland sites with reserved workload capacity available.

iland has seven datacenters throughout the US and Europe (with future expansion into Asia Pacific) and whether an organization is leveraging them for production, development, or disaster recovery every user deserves enterprise capabilities.   All iland cloud services datacenters are built with this concept in mind as are our locations.

Backup plans are meant to give us a level of comfort so that if something disastrous happens we have a plan in place to deal with the consequences.

When it comes to preparing a business for a potential disaster the same principles apply – except that there’s a whole lot more to think about when it comes to protecting your business’s assets and IT infrastructure.  For example have you thought about your data and how it would be protected and accessed should a disaster occur?

Cloud computing offers a realistic and cost effective option for disaster recovery and business continuity whatever the size of your business.  Being able to house your company’s data in multiple, high availability and geographically redundant datacenters means that your data remains intact and accessible in the event of a disaster.  And for businesses in California that may have to face an earthquake, those in the Midwest tackling tornadoes, companies in the north dealing with ice storms in winter and folks in the south facing hurricanes and tropical storms it’s nice to know that you can continue in business should you ever be affected.

Overall, backup plans just make sense and geographical redundancy is what matters most when developing your disaster recovery plan. The alternative is that if your business operations stall and data can’t be replicated to a secondary site, it could cost you thousands of dollars in lost revenue, not to mention the loss in customer confidence.