Explanation of disaster recovery in cloud computing

Disaster recovery in cloud computing is a relatively new concept, and like many technology trends, there is a lot of hype and wrong information. In the "Storage" magazine article from Jacob Gsoedl, one can understand the top disaster recovery issues of cloud computing, such as security and data recovery, and whether disaster recovery in the cloud is a good choice for the organization.

Cloud computing, as well as mobile and tablet devices, have brought people into the era of high technology. But when it comes to hype, cloud computing is more than people think, which has unintended consequences, sometimes exceeding its actual utility.

While the concept of cloud-based disaster recovery (and some products and services) is still emerging, some companies, especially small and medium-sized businesses, are discovering and starting to take advantage of disaster recovery cloud services. It may be an attractive alternative to companies that bundle IT resources because the cost of using cloud services is ideal for businesses that adopt disaster recovery, where infrastructure servers are idle most of the time. Deploying disaster recovery sites in the cloud reduces the need for data center space, IT infrastructure, and IT resources, dramatically reducing costs and enabling small businesses to deploy disaster recovery options that can only be used in large enterprises. Lauren Whitehouse, senior analyst at Enterprise Strategy Group (ESG) in Milford, Mass., said: "Cloud-based disaster recovery will discuss moving from data center space and hardware to cloud capacity planning."

But disaster recovery in cloud computing is not a perfect solution, and companies need to clearly understand its shortcomings and challenges before investing. Security is usually at the forefront of the watchlist:

• Is the data securely transmitted and stored in the cloud?

• How to authenticate users?

• Is the password the only option, or is the cloud provider providing some type of two-factor authentication?

• Does the cloud computing provider meet regulatory requirements?

• And because cloud computing needs to be accessed over the Internet, bandwidth needs need to be clearly understood. Only planning bandwidth requirements can migrate data to the cloud without the risk of adequate analysis of how data is accessible in the event of a disaster:

• Does the organization have the bandwidth and network capacity to redirect all users to the cloud?

• How long does it take to recover if you plan to recover from cloud computing to an on-premises infrastructure?

Chander Kant, CEO and founder of Zmanda, a cloud backup service and open source backup application provider, said: "If users use cloud-based backup as part of disaster recovery, they need to design a backup set for recovery."

Other key considerations include the reliability of cloud computing providers, availability, and the ability to serve users during a disaster. Choosing a cloud service provider or management service provider (MSP), it is critical to provide services under agreed conditions, and making the wrong choices may not put users in IT hell, but it is easy to get into trouble. .

Design disaster recovery in the cloud blueprint

As with traditional disaster recovery, there is no single disaster recovery blueprint in cloud computing. Each enterprise is unique in its applications and the relevance of its applications to its business and its industry. Therefore, cloud computing disaster recovery plans (also known as cloud computing disaster recovery blueprints) are very specific and unique to each organization.

Classification is the first principle for deriving traditional and cloud-based disaster recovery plans. The process of designing a disaster recovery plan begins by identifying and prioritizing applications, services, and data, and determining acceptable downtime for each computer before significant business impact occurs. Then, the priority and the required recovery time objective (RTO) will determine the disaster recovery method.

In this process, identifying key resources and recovery methods is the most relevant aspect because organizations need to ensure that all critical applications and data are included in their blueprints. Similarly, in order to control costs and ensure rapid and centralized recovery as the plan needs to be implemented, organizations need to ensure that irrelevant applications and data are ignored. The more concentrated the disaster recovery plan is, the more likely it is that the company will periodically test and execute within the defined goals.

By identifying and prioritizing applications and defining RTOs (Recovery Time Objectives), organizations can determine the best and most cost-effective way to implement RTO, which needs to be done through applications and services. In rare cases, organizations will have a disaster recovery approach that works for all applications and data; more likely, there are several ways to protect applications and data with similar RTO clusters. Seth Goodling, Acronis Virtualization Practice Manager, said: "The combination of cost and recovery goals drives different levels of disaster recovery."

Disaster recovery in cloud options

Hosted apps and hosting and disaster recovery. An increasingly popular option is to bring major production and disaster recovery instances into the cloud and be handled by a managed service provider (MSP). By doing so, companies will gain all the benefits of cloud computing, from the cost of use to the elimination of on-premise infrastructure. Instead of implementing it yourself, companies postpone disaster recovery to cloud or managed service providers. The choice of service providers and the process of negotiating appropriate service level agreements (SLAs) are critical. By passing control to the service provider, the organization needs to ensure that it provides uninterrupted service in the definition SLAs of the primary instance and the disaster recovery instance. Greg Schulz, founder and senior analyst at StorageIO Group, said: "The relevance of service level agreements to cloud computing providers cannot be overstated, and SLAs are negotiating access to enterprise applications."

For email and some other business applications such as customer relationship management, pure cloud gaming is becoming more and more popular, and Salesforce.com has been the leading cloud services CRM market leader.

Back up to the cloud and recover from disasters in the cloud. In this approach, applications and data remain internal, data is backed up to the cloud, and restored to local hardware in the event of a disaster. In other words, backups in cloud computing are an alternative to tape-based offsite backups.

When considering cloud backup and recovery, it's important to have a clear understanding of backups and more problematic recovery aspects. Backup to cloud computing is relatively straightforward, and backup application vendors have expanded their backup suites to choose to back up directly to popular cloud service providers such as AT&T, Amazon, Microsoft, Nirvanix and Rackspace. David Ngo, director of engineering alliances at CommVault Systems, said, "Our cloud connector migrates deduplicated, compressed and encrypted data to the cloud and allows for the retention of data in cloud computing. They properly summarize the enterprise should The search function moves data to products in the cloud. Similarly, cloud gateways such as Cirtas Bluejet Cloud Storage Controller, F5ARX Cloud Extender, NasuniFiler, Riverbed Whitewater and Twin Strata Cloud Array can be used to move data into the cloud. They span internal deployments and cloud storage and keep internal data and data synchronized in cloud computing."

The challenging aspect of disaster recovery using cloud-based backup is recovery. Since the bandwidth to be recovered is limited and may reach several terabytes, restoring data back to the local within the defined RTO can be challenging. Some cloud backup service providers offer the option to restore data to disk and then send it to the customer for local local recovery. Another option is the large local cache that was recently backed up for local disaster recovery.

"I firmly believe that backup needs are local and sent from there to the cloud; in other words, cloud computing backups become a secondary offsite backup for the enterprise," said Jim Avazpour, president of infrastructure at OS33.

On the other hand, depending on the data to be recovered, functions such as compression and, more importantly, deduplication can make recovery from data in the cloud to the on-premises infrastructure a viable option. A good example is Rockford Construction in Michigan, which uses Stor Simple appliances for cloud-based protection of its Exchange and Share Point infrastructure. "In the event of a disaster, we will extract virtual machines from the cloud; with Stor Simple's deduplication technology, we can only download a full copy of the virtual machine, and the differences between the other versions are different." Rockford Building Shaun Partridge, vice president of corporate IT, said.

Back up and restore to the cloud. In this approach, the data is not restored back to the on-premises infrastructure; instead, it is restored to the virtual machine in the cloud. This requires cloud storage and cloud computing resources, such as Amazon's Elastic Compute Cloud (EC2). Recovery can occur when a disaster occurs in a statement or in a continuous (pre-stage). Keeping the pre-staged DR virtual machine relatively up-to-date by scheduled recovery is critical to situations where a positive RTO needs to be met. Some cloud service providers are promoting the provision of cloud virtual machines as part of their DR products. Chris Poelker, vice president of enterprise solutions at FalconStor Software, said: "Many cloud service providers use our products for secure deduplication replication and virtualize servers into the cloud.

Copy to a virtual machine in the cloud. For applications that require aggressive recovery time and recovery point objectives (RPO) and application awareness, replication is the data migration option of choice. Virtual machines replicated to the cloud can be used to protect the cloud and internal production instances.

In other words, copy virtual machines for cloud computing to cloud computing virtual machines and internal deployment to cloud computing virtual machine data protection. Replication products are based on continuous data protection (CDP) such as CommVault continuous data replicators, snapshots or object-based cloud storage such as EMC Atmos or Hitachi Content Platform (HCP). Robert Primmer, senior technical expert and senior director of content services at Hitachi Data Systems, said: "Cloud computing service provider Peak Web Hosting enables local HCP instances to be replicated to Peak Web HCP instances instead of another internal HCP instance."