The documentation set for this product strives to use bias-free language. For the purposes of this documentation set, bias-free is defined as language that does not imply discrimination based on age, disability, gender, racial identity, ethnic identity, sexual orientation, socioeconomic status, and intersectionality. Exceptions may be present in the documentation due to language that is hardcoded in the user interfaces of the product software, language used based on RFP documentation, or language that is used by a referenced third-party product. Learn more about how Cisco is using Inclusive Language.
Cisco offers the Disaster Recovery as a Service (DRaaS) Solution architecture. This architecture enables Cloud Service Providers (CSPs) to offer Disaster Recovery (DR) services to workloads outside of the service provider's management domain that are in customer premise environments or in colocated environments. Service providers can also offer data protection and data survivability services to workloads within the cloud provider's Virtual Port Cloud (VPC) environment and management domain.
This chapter includes the following major topics:
Previous releases and white papers for the DRaaS Solution validated DR solutions from Cisco partners (InMage and Zerto) overlaid on Virtual Multiservice Data Center (VMDC) Version 2.3. This allowed VMDC-enabled CSPs to enhance their addressable market, improve financial performance, and differentiate from commodity/public cloud solutions.
Release 2.0 of the DRaaS Solution architecture, which is described in this document, is designed to provide a new set of DR-related capabilities by leveraging new features of the recently released Cisco VMDC Virtual Services Architecture (VSA) Version 1.0 system. This release of the DRaaS solution increases VMDC-enabled CSP differentiation by adding new, advanced network and operations features to the solution, including the following:
For CSPs, the traditional DR system constitutes a substantial portion of expenses annually. A cloud-based DR system uses a "pay as you go" model and minimizes the impact of downtime through continuous data replication to the CSP cloud. Protected machines can be recovered in the CSP cloud in a matter of minutes rather than hours, enabling business continuity when a disaster event is identified. See Figure 1-1.
The most important end user consumable service being enabled by this system architecture is enabling service providers to offer disaster recovery for both physical and virtual servers from a customer data center to a service provider VPC. The key requirements for DRaaS are Recovery Point Objective (RPO), Recovery Time Objective (RTO), performance, consistency, and geographic separation. RPO is the maximum amount of data loss tolerated during disaster recovery and RTO is the maximum amount of time that can be used to restore services.
CSPs that deploy the DRaaS Solution architecture detailed in this guide can offer the following end user-consumable services for both physical and virtual servers on an aaS (as-a-Service) basis:
Figure 1-2 In-Cloud Disaster Recovery (ICDR)
The global DRaaS and cloud-based business continuity market is expected to grow from $640.84 million in 2013 to $5.77 billion by 2018, at a CAGR of 55.20%. The market presents a strong opportunity for the CSPs to take advantage of the demand for DRaaS services, as illustrated by Figure 1-3.
Figure 1-3 Strong Market Demand for DRaaS
Further investigation of the global demand patterns for DRaaS indicates that the market opportunity and interest is equally spread across the enterprise, mid-market, and SMB segments, as summarized in Figure 1-4.
Figure 1-4 Global DRaaS Demand by Segment
The Forrester studies indicate that barriers exist that need to be addressed to achieve wide-scale adoption of disaster recovery at the enterprise level and from a service provider level.
Traditional DR solutions require matching hardware at both the source side and the target side with the replication being performed by a hardware device, usually the storage array. This created a capital cost barrier for the equipment purchased and significantly increased the administrative overhead to the point that the Forrester survey shows the majority of the respondents had no plan of implementing disaster recovery.
From a service provider perspective, the lack of similar equipment at each customer site made offering a DRaaS solution so expensive that it was not pursued as a feasible service offering.
Even if the hardware cost barrier can be overcome, traditional DR solutions require large administrative efforts to implement. Implementation usually has an extended professional services engagement and a significant learning curve for the administrators. For the service provider, building the core DR infrastructure is only part of the challenge. Creating a multi-tenant capable service offering has traditionally required a significant application development and programming effort.
Cisco's Disaster Recovery as a Service (DRaaS) solution architecture is based on Virtualized Multiservice Data Center (VMDC) and Cisco Unified Computing System (UCS). VMDC is a reference architecture for building a fabric-based infrastructure providing design guidelines that demonstrate how customers can integrate key Cisco and partner technologies, such as networking, computing, integrated compute stacks, security, load balancing, and system management. Cisco UCS is a next-generation data center platform that unites compute, network, storage access, and virtualization into a cohesive system designed to reduce total cost of ownership (TCO) and increase business agility.
Cisco VMDC and UCS reduce infrastructure expenditures (CAPEX) and operational expenses (OPEX) to increase profitability by reducing the number of devices that must be purchased, cabled, configured, powered, cooled, and secured. The unified architecture uses industry-standard technologies to provide interoperability and investment protection.
Cisco VMDC and UCS help businesses adapt rapidly and cost efficiently in response to changes in the business environment by enabling the fast provisioning of IT infrastructure and delivery of IT as a service. Deployment time and cost is more predictable using an end-to-end, validated, scalable and modular architecture. The unified architecture supports multiple applications, services, and tenants.
Cisco VMDC and UCS simplify IT management to support scalability, further control costs, and facilitate automation—key to delivering IT as a service and cloud applications. The architecture enhances the portability of both physical and virtual machines with server identity, LAN and SAN addressing, I/O configurations, firmware, and network connectivity profiles that dynamically provision and integrate server and network resources.
Increased regulatory pressure drives the need for DR and business continuity plans and presents a hierarchy of requirements for the implementation of these solutions (geographic restrictions, regulatory compliance, etc.). Enterprises are constantly faced with budget constraints that prevent infrastructure duplication. Building disaster recovery infrastructure is a contextual business activity that requires a degree of specialization with IT skillsets or resources that are significantly harder to build without sufficient scale. Under these circumstances, a growing desire exists to consume DR as a service, allowing incremental deployment and growth as budget becomes available. See Figure 1-5.
Figure 1-5 Cisco's DRaaS Blueprint Solution
The selection of a specific technology and implementation for the implementation of DRaaS is a highly complex decision with technology challenges that need to be adequately explored and analyzed. See Figure 1-6.
The following questions arise in the choice of the DRaaS implementation:
Figure 1-6 DRaaS Technical Challenges
The use of traditional storage-based replication requires an identical storage unit on the disaster recovery site from the same vendor. The storage array-based replication software is not application aware and needs additional intervention at the host level to achieve application consistency. Multiple points of management are required while performing disaster recovery and this introduces complexity in protecting and recovering workloads. The traditional storage-based replication approaches lack granularity and can replicate either all VMs or none that are residing on a logical unit number (LUN). Replication of data happens between LUN pairs that need to be identical and this restricts the ability to failover a single VM residing on the LUN. See Figure 1-7.
Figure 1-7 Storage-based Replication
Traditional storage replication approaches need additional functionality to take snapshots or clones of the target LUN to perform disaster recovery drills without interrupting data replication. Otherwise, replication has to be stopped for disaster recovery drills. Storage array-based replication does not support continuous data protection natively and data cannot be protected from logical failures.
DRaaS offers the following value to service providers:
One commonly encountered question is how the backup-based disaster recovery approaches compare to Cisco's recommendation for DRaaS architecture for SPs. Table 1-1 shows the key considerations and a comparison of the approaches.
Cisco DRaaS presents a model (see Table 1-2 ) that clearly delineates the responsibilities of the service providers that provide the DRaaS services and the end customer guidance on the ownership and expectations in the system offering.
Figure 1-8 is a financial model that presents the monetization opportunity for service providers associated with the deployment of the Cisco DRaaS solution architecture.
Figure 1-8 Monetization Opportunity for SPs
DRaaS provides the following value for Enterprises: