Most people would agree on a definition of what constitutes a disaster. Most would also agree on what “recovery” means. Put those two words together, however, and people start speaking different languages.
Depending on whom you talk to, “disaster recovery” might be software. Then again, it might online backup or backup tapes stored offsite. To some it is replication to a remote data center. For others it is active-active data centers so there is never any service interruption. Some wouldn’t consider a disaster recovery solution complete unless it included mobile workspaces and a shipping container of racked-and-wired servers that could be rolled to the site.
The good part of all this is that, whatever your company considers necessary to recover its own systems from a disaster, some vendor is offering that service. Here are eight firms offering disaster recovery/business continuity (DR/BC) services.http://o1.qnsr.com/log/p.gif?;n=203;c=204655439;s=10655;x=7936;f=201806121855330;u=j;z=TIMESTAMP;a=20400368;e=i
Peak 10, Inc. of Charlotte, N.C., targets mid-market businesses with its Recovery Cloud service. It doesn’t have offerings for PCs or mobile devices, nor does it just replicate customers’ storage, choosing instead to focus on business-critical infrastructure.
Peak 10 has four services customers can use to protect their data and ensure application availability. Recovery Cloud is a fully managed service that is an add-on to a Peak 10 Cloud subscription or used to protect a customer-owned private VMware environment. Replication is done at the hypervisor level, agnostic of the underlying storage. There is complete isolation between customers, and the company never exposes a management interface to the cloud that may put all customers at risk. Peak 10 includes a scheduled testing, runbook maintenance, and ongoing monitoring and management of the protected environment.
“Most companies look at BC/DR preparation as one-time events,” says Ken Seitz, Peak 10’s director of product strategy. “They configure data replication, create a process, and may even do an initial test. Beyond that, it falls into the ‘set it and forget it category.’”
Regular testing very rarely occurs and the process document almost immediately becomes outdated as aspects of the environment continue to evolve and change, so at the time of a disaster there are a number of problems with the failover. Peak 10 avoids this by performing a minimum of two tests at the point of provisioning. The initial is done by the individual(s) drafting the customer-specific runbook and configuring the technology. The second test is done with the customer and other Peak 10 operational team members to validate the runbook. If significant modifications are made during the second test, yet another test is done. After the provisioning period, tests are run a minimum of twice per year.
If a customer isn’t on a VMware based environment, Peak 10 offers a Managed Replication service based on an agent at the operating system or non-VMware hypervisor level that can protect servers and applications. It is managed on a customer-by-customer basis, so it requires a custom SLA dependent on the scope of the environment protected.
Finally, protection can be done at the application level, such as the SQL-as-a-service offering packaged around Microsoft SQL Server 2012, that has business continuity (local, automated failover) and disaster recovery (geographically diverse, manual failover). Additionally, Peak 10 can manage application level BC/DR for dedicated Microsoft application environments including SQL, Exchange and SharePoint.
The Recovery Cloud service provides a 30 minute Recovery Point Objective (RPO), and a one hour Recovery Time Objective (RTO). SQL as a Service includes a 100% availability SLA. Service Level Agreements around Managed Application and Managed Replication offerings are dependent on choices the customer makes about their environments and are thus negotiated individually. Recovery Cloud has three key pricing components: the number of servers being protected, the disk foot print required in the disaster recovery environment, and the cloud compute resources that are used during testing and declaration.
SunGard Availability Services
SunGard Availability Services has been in the DR business since the 1970s and supports over 8,000 customers worldwide from small to medium enterprises all the way up to Fortune 500 companies.
John Paul Blaho, SunGard’s director of product marketing, says that companies are facing three main challenges in the DR/BC area. “Recovery teams are experiencing difficulty in synchronizing their recovery environment to their production environment due to the sheer number of product environment changes,” he says. “Organizations are limited in resources to properly test and manage their recovery program, or lack expertise to support the needs to develop a proper recovery plan, and companies are failing to recover against their mandated recovery objectives during tests or actual declared disasters.”
SunGard has more than 5 million square feet of data center and operations space with advanced HVAC controls, security systems, hardware and software, as well as reliable tape and disk storage and sophisticated networking services. With more than 80 facilities worldwide and more than 30 mobile units, the company provides customers with a range of options for DR/BC.
“Our tiered approach to recovery of applications based on business process criticality provides the flexibility to shift solution mix as business requirements change,” says Blaho. “Customer options include fully-managed backup and recovery, dedicated and cloud-based recovery options, shared recovery options, as well as workforce continuity and software solutions.”
SunGard’s Managed Recovery Program (MRP) provides customized solutions structured around specific business needs and application requirements with SunGard assuming full accountability for the data recovery program and overall disaster recovery management. SunGard assigns a dedicated Service Delivery Manager who has expert knowledge of the customer’s recovery program, including the workflow, timeline and communication procedures. The manager conducts test planning, test execution, test monitoring, post-test reporting, and the post-test reviews.
SunGard’s Recover2Cloud for Server Replication is designed for the most critical applications. Data is replicated from selected production servers to secure cloud infrastructure with near-zero recovery point objectives and recovery time objectives of less than four hours. The service also incorporates Continuous Data Protection to enable recovery to any point in time within several days prior to a failure.
For organizations with large-scale virtual environments, Recover2Cloud for vCenter SRM is a fully managed solution that leverages a customer’s existing storage devices and VMware skill set and so has reduced management costs and pay-as-you-go pricing based on usage.
Windstream Hosted Solutions
Windstream Hosted Solutions has about twenty data centers currently operating, primarily east of the Mississippi, with more under construction.
“IT is increasingly critical to operations for all businesses, and acceptable downtime for key applications has shrunk dramatically in recent years,” says Jaclyn Mispagel, Windstream Hosted Solutions, sr. consultant - product marketing. “Yet many companies have no disaster recovery plan or have serious gaps in their plans that limit their ability to quickly restore operations in the event of a disaster.”
Windstream’s Disaster Recovery as a Service (DRaaS) is geared to mid to large enterprises in all verticals. Mispagel says that what distinguishes Windstream’s DRaaS from other vendors is that it is a complete packaged service customized to the needs of the particular customer.http://o1.qnsr.com/log/p.gif?;n=203;c=204655439;s=10655;x=7936;f=201806121855330;u=j;z=TIMESTAMP;a=20400368;e=i“[Other companies] are piecing together disparate services to make a solution for the customer,” she says. “We have already done the initial legwork on this and have created the solution base packaging to help make it easier for the customer to understand the benefits.”
All the DR services are provided from Windstream’s SSAE 16-compliant data centers equipped with 2N power and cooling, and 80 percent of the customer base is also using the data centers for hosting production environments. Customers can use Internet (VPN recommended) and/or private MPLS connectivity to access the DR environment—either Windstream provided or third party. Windstream monitors the replication processes to ensure they are successful. Customers also have access to a cloud portal to monitor the status of their DR environment in real time.
There is an initial DR drill (test) immediately after implementation, and the packages standardly include one additional DR drill annually. The customer has the option to purchase additional DR drills. The DR drill is set up to be non-impacting to production and not break replication.
Windstream offers multiple packages to meet various customer environments, physical or virtual. Each of these packages has a basic pricing set up of a base package fee (includes virtual firewalls, VLANs, DR drills, etc.) and variables such as the number of servers, filers being protected, total storage price per GB, and the various recovery VMs or recovery pools that may be needed by the customer. The SLAs call for 100 percent uptime on the infrastructure, 99.9 percent uptime on cloud services and an RTO of four hours (up to 50 VMs).
Acronis offers a software-only solution that includes compression and deduplication, only sending block level changes to the cloud for backup. The company serves SMBs through enterprises, with a strong presence in the government and education sectors. A single console is used to manage backup and recovery for physical, virtual and cloud environments.
“Acronis allows for the creation of live images of physical image,” says Seth Goodling, director of strategic technology at Acronis. “Not only does it protect the physical infrastructure but it also allows you to move to different infrastructure in case of physical hardware failure.”
The software supports Windows and Linux servers and desktops, and provides granular recovery for Microsoft databases (Exchange, SQL, Active Directory and Sharepoint.) The company also provides disk space for disk-to-disk-to-cloud backup. Services are priced per physical host.
HP offers enterprise clients Continuity Services from recovery centers located in the Americas, Europe, Africa, Asia and Australia.
“All enterprises today are reliant on their IT environment to run their business processes,” says Annabelle Baxter, manager, worldwide media relations, for HP Enterprise Services. “HP offers Continuity Services to protect and minimize disruptions to IT infrastructures and associated business processes caused by site outages, disasters, and environmental events and to assist with general system/application availability.”
HP’s midrange recovery services come in four levels. Active-active comes with dedicated, fully managed equipment and near-zero recovery time objective and near-zero recovery point objective. Hot site service is cloud-based disaster recovery with a 1-4 hour RTO and 15 minute RPO. Warm sites have shared, managed services with RTOs of 8 to 24 hours and RPOs based on the last backup. Finally, there are cold sites with shared equipment and no managed services. HP also offers active-active, hot site and warm site recovery services for mainframes, with approximately 240,000 MIPS (million instructions per second) of IBM System z mainframe capacity.
HP provides recovery services for most types of servers, storage, databases, applications and mobile devices. It has 10,000 seats of mobile and fixed workspaces for emergency use, both general workspaces as well as areas tailored for call centers and the financial services industry. The prices depend on the services desired. For HP Enterprise Cloud Services, continuity is priced primarily based on the size and number of servers protected and the amount of storage protected.
As one would expect from Big Blue, IBM’s Business Continuity and Resiliency Services (BCRS) consultancy can provide recovery services for any type of server, database, application or storage its customers have.
“We utilize all parts of IBM, including the IBM Research group, to provide services that include infrastructure recovery, cloud based, managed and consulting,” says Richard Cocchiara, IBM distinguished engineer, Chief Technology Officer (CTO) and the managing partner of consulting for BCRS.
IBM BCRS has 160 recovery centers around the world and offers mobile emergency computing resources in some areas. Backup can be done by tape or through deduplicated online replication. Service levels are dependent upon a client’s environment and requirements, with prices based on the services needed. But, although IBM is a Fortune Global 100 firm, its recovery services are not just for large enterprises and multinationals.
“We find that our cloud based recovery services have brought recovery capabilities that were once only available to large companies within reach of small and medium sized companies,” says Cocchiara.
nScaled, Inc.’s Recovery as a Service (RaaS) is geared toward mid-market firms in the US and UK running up to 550 physical or virtual servers and needing up to 100TB of cloud storage.
“Our offering has never been based on or rooted in public cloud architecture, and we built it originally and specifically for DR as a service,” says Scott Reynolds, nScaled’s director of marketing. “We started out focused on cloud-based data center, server and file recovery with RPO and RTO within minutes.”
RaaS is designed to recover anything from a single file all the way to a complete data center. It supports Windows, Linux and UNIX servers; Windows and Linux PCs; and MS SQL, MySQL and Oracle databases. Clients can protect data, physical and virtual servers, applications, and entire data centers with complete infrastructure stacks: pre-built, pre-configured virtual data centers priced at less than what it would cost them to build it themselves.
The administrative console is designed to be simple enough for business managers to use, eliminating the requirement for system administration skills. RaaS includes process automation, monitoring, testing and alerts. The underlying multi-tenant architecture allows both local and remote server spin-ups. nScaled’s resource allocation model can service multiple disaster declarations simultaneously, rather than recovering clients on a first come, first served basis. Pricing is by storage and compute resource required.
NTT Cloud Recovery
NTT Communications targets multinational and enterprise customers with its Cloud Recovery, which is part of its Recovery as a Service Suite. Cloud Recovery allows business operations to continue as usual during unexpected server failures or planned downtime through auto-failover and redirection of traffic to a real-time, replica server in a secure NTT America cloud environment.
“The key challenge that companies are facing is not if they need BC/DR capabilities, but how to adapt their current environment architecture so that there is a good overall fit with the cloud architecture,” says Nayan Naik, NTT Communications, director product strategy, cloud recovery. “Companies tend to overlook the fact that fundamentally they are protecting the same applications, users and access that they had yesterday but in a totally different infrastructure model. They no longer have direct access to the physical level or ownership of the equipment which changes the way in which they need to plan, educate and react.”
Cloud Recovery is a storage-agnostic platform that works with Microsoft and Oracle Databases and Microsoft and Blackberry applications running on Windows, Linux (Red Hat, CentOS, SUSE and Oracle) and UNIX (Solaris and AIX) servers. It also has a host-based replication model allowing for support of 100 percent of the PC (Windows and Linux) based desktops for data replication capabilities.
NTT provides a DR planning program that allows customers to establish a completely protected environment for their systems in the exact same manner that they would their production systems. Customers can then test their recovery in a sandbox without impacting production systems. Services are priced per protected VM/System.