ESG Validation

ESG Lab Validation: Veeam Availability Suite and Cisco UCS

Introduction

Veeam and Cisco make the claim that Veeam’s purpose-built availability solution, which combines Veeam Availability Suite with Cisco Storage Servers, is a unique solution that enables users to meet the highly stringent recovery time and point objectives for tier-1 applications. As organizations struggle to realize the full potential of virtualization and converged infrastructure, it becomes ever clearer that IT administrators need a powerful, easy-to-use, and affordable availability solution to today’s IT challenges.

Background

Although backup and recovery is a longstanding IT discipline, it remains challenging and problematic. When asked about their primary data protection challenge, organizations’ top responses, including cost(s), keeping pace with the capacity of data to protect, and managing complexity, were closely associated with data growth (see Figure 1). Examining these challenges alongside mandates from IT leadership to increase the reliability and performance of backups and recoveries while reducing the hardware, software, and operational costs of data protection puts IT organizations in a tight spot.[1]

Figure 1. Top Challenges with Organization’s Current Data Protection Processes and Technologies

Source: Enterprise Strategy Group, 2017

In another survey, current and potential backup appliance users were asked why they have deployed or are considering the deployment of a backup appliance. More than half cited simplified management (51%), with nearly half citing streamlined deployments (47%).[2]

Organizations are enthusiastically modernizing their data centers, leveraging server virtualization and converged infrastructures, and continuing to expand their virtualized workloads to include business- and mission-critical applications. Modernization will increase operational agility, facilitate higher availability, and reduce equipment, power, cooling, facilities, and management costs through consolidation. However, many existing backup solutions have not kept pace with these new environments and often negate the cost reduction and management simplification benefits. Traditional backup methods with agents in each virtual machine (VM) increase the load on hosts and the operational burden on IT staff to maintain hundreds to thousands of agents. In addition, bolted-on image-based backup adds significant complexity and often little capability.

Modernizing the production environment without also modernizing protection capabilities will not work. A modernization journey should include both production and protection within the same strategic plan to ensure both a lack of hindrance and the ability to accommodate and rely on the resiliency and recovery measures provided by the improved protection technologies.

  • For production, that means reliability is priority one, with ease of use, performance, and agility being key corollaries.
  • For protection, efforts must be about the same four principles: reliability, ease of use, performance, and agility.

What is needed is a pre-validated, purpose-built backup solution that leverages the core attributes of virtualization and converged infrastructures—including cost, agility, and availability benefits—to accelerate the modernization of the data center to support tier-1 business- and mission-critical application workloads, which require high availability and full protection.

Veeam Availability Solution with Cisco UCS

Veeam Availability Suite with Cisco UCS bridges the gap by providing customers with a differentiating solution that offers more than just backup. Using virtualization, storage, and cloud technologies, Cisco and Veeam deliver high-speed recovery, data-loss avoidance, and verified protection with complete visibility via Cisco’s proven UCS platform for applications requiring high availability and scalability. The solution is designed to work for architectures of any size, from small, single-site organizations to large multi-site environments incorporating hybrid and public cloud data repositories. With the Cisco Storage Server’s combination of computing power, memory, and high-capacity storage, deployment can be accomplished on a single Cisco UCS S3260 server, or it can accommodate more complex architectures.

Veeam Backup & Replication, a core component of Veeam Availability Suite, is a powerful, easy-to-use, and cost-effective backup and availability solution. It provides fast, flexible, and reliable recovery of virtualized applications and data, bringing VM backup and replication together in a single software solution. Veeam Backup & Replication delivers award-winning support for VMware vSphere and Microsoft Hyper-V virtual environments.

Figure 2. Solution Overview

Source: Enterprise Strategy Group, 2017

Veeam Availability Solution features include:

  • Recovery of entire VMs, individual files, and application items.
  • Fast, agentless, item recovery and e-discovery for Microsoft Exchange, SharePoint, and Active Directory and Transaction-level restore of Oracle databases and SQL Server databases.
  • Automatic recoverability testing of every backup and every replica, every time.
  • Built-in WAN acceleration to get backups and replicas offsite up to 50x faster than standard file copy while reducing traffic across the network.
  • Unlimited scale-out backup repository, direct storage access, and built-in deduplication and compression.

The Cisco UCS S3260 Storage Server is a modular, high-density storage server that’s ideal for modern data protection. The Cisco UCS S3260 addresses the need for highly scalable computing with high-capacity local storage. Designed for a new class of applications, it is simple to deploy and excellent for backup repositories. Together, Veeam and Cisco UCS S3260 create a robust platform for backups, reducing backup ingest bottlenecks and providing faster backups using parallel processing.

Figure 3. Solution Overview (Storage)

Source: Enterprise Strategy Group, 2017

Key Cisco UCS Storage Server features include:

  • Dual two-socket server nodes based on Intel Xeon processor E5-2600 v4 CPUs with up to 36 cores per server with up to 512 GB of DDR4 memory per server node and support for high-performance Non-Volatile Memory Express (NVMe) and flash memory.
  • Massive 600TB data storage capacity that easily scales to petabytes.
  • Integration with Cisco UCS Manager and standalone Cisco Integrated Management Controller (CIMC).
  • Dual-port 40Gbps system I/O controllers with Cisco UCS Virtual Interface Card (VIC) 1300 embedded chip and Unified I/O for Ethernet or Fibre Channel connected to existing NAS or SAN storage environments.

ESG Lab Validation

ESG Lab performed hands-on evaluation of the Veeam/Cisco data protection solution at a Cisco facility in San Jose, California. Testing was designed to highlight the differentiated capabilities of the combined solution, with a focus on ease of deployment and management, operational efficiency, and price/performance of the solution compared to other purpose-built solutions on the market.

Price/Performance

ESG Lab evaluated the price/performance of the Veeam and Cisco solution and modeled cost/capacity and price/performance compared with leading purpose-built backup appliances (PBBAs) on the market as of this writing.

ESG Lab Validated

ESG Lab tested performance of the Veeam and Cisco solution in protecting a VMware vSphere environment with 36 Windows Server 2016 server VMs running on six Cisco UCS C220 Rack Servers. Multiple full and incremental backups were run to obtain a baseline measurement of backup performance throughput, which was used to model price/performance. A series of tests were run, scaling the number of jobs and number of available drives after each test, with the final tests using all 36 virtual machines and all 56 drives in the Cisco UCS S3260.

As seen in Figure 4, throughput scaled as workload was increased, and the fully provisioned S3260 sustained backup throughput of 10.44 TB per hour. It’s important to note that this is the actual data transferred across the wire.

Figure 4. Veeam and Cisco UCS S3260 Performance Throughput

Source: Enterprise Strategy Group, 2017

Next, ESG Lab looked at the cost per terabyte of the Veeam and Cisco solution compared with other PBBA vendors in the market. Mid-level offerings from each vendor were compared and pricing data was gathered from publicly available sources. Quotes were provided to ESG Lab by one reseller as of December 8, 2016.

As seen in Figure 5, the Veeam/Cisco solution offers higher capacity compared to other offerings at a significantly lower cost per terabyte of storage.

Figure 5. Comparison of Cost/Capacity between Veeam/Cisco Solution and PBBAs

Source: Enterprise Strategy Group, 2017

Next, ESG Lab looked at price/performance as compared with one vendor’s solution that is as similarly configured to the Veeam/Cisco configuration as possible and used vendor-published pricing and performance data for comparison.

Figure 6. Cisco and Veeam Price/Performance Comparison

Source: Enterprise Strategy Group, 2016

The performance of the Veeam and Cisco solution was 28.9% higher than the alternate solution (10.44 TB per hour versus 8.1 TB/hour), but as shown in Figure 6, the price/performance of Veeam and Cisco was 49% lower ($37.26 per GB per hour versus $73.16 for the other solution).

Why This Matters

Data center modernization has accelerated the pace of business by enabling such feats as nearly instant infrastructure provisioning with greater flexibility and availability. Realizing the benefits, organizations are working to virtualize more business- and mission-critical applications and leverage more converged infrastructure, while struggling to adhere to mandates from IT leadership to increase the reliability and performance of backups and recoveries in addition to reducing the hardware, software, and operational costs of data protection.[3]

ESG Lab has validated that the Veeam and Cisco solution provides not only great performance that shrinks backup windows and reduces risk, but also compelling price/performance, delivering 10.44 TB per hour of ingest performance at 49% lower cost than other vendors evaluated by ESG Lab.

The ability to quickly return to normal operations after data loss or data corruption is critical to an organization’s success. Extended downtime is extremely costly and can lead to the demise of a business. ESG Lab validated Instant VM Recovery, which enables IT administrators to easily boot VMs directly from compressed, deduplicated backup files—which enabled restart of a VM from a backup in less than two minutes. This powerful feature minimizes downtime and disruption for not only business users, but also IT. Users don’t have to wait to resume working until the entire backup file is recovered and presented back to production storage. Built for virtualization and delivered in a pre-validated configuration on the powerful Cisco UCS platform, Veeam can leverage the virtual environment to reduce recovery time and streamline the recovery process.

 

Ease of Deployment and Management

This section of the validation report explores how Veeam Availability Suite 9.5 and the Cisco S3260 Storage Server have been combined to create an easy-to-manage purpose-built data protection solution. The preconfigured solution includes Windows OS and Veeam application installation along with backup repository storage configuration to ensure easy deployment into the customer’s environment.

ESG Lab Validated

ESG Lab began validation of ease of deployment and management by logging into the Cisco Integrated Management Controller (CIMC) and reviewing the storage configuration. As shown in Figure 7, the Cisco S3260 was configured with two virtual drives. Virtual drive one was configured as a RAID1 drive set with 460 GB of space used to host the Windows OS. Virtual drive number two was configured as a RAID60 drive set with 160 TB of space allocated for the Veeam backup repository. From the Windows OS running on the Cisco S3260, the RAID60 drive set was formatted with the Resilient File System (ReFS), assigned a drive letter, and used to create a directory. The resulting configuration, E:\Backup, was subsequently used for the Veeam backup repository. The Veeam management interface was then used to define and add the new space to the Veeam backup configuration as a new Microsoft Windows server type backup repository.

Figure 7. Backup Repository Configuration

Source: Enterprise Strategy Group, 2017

Next, using the Veeam UI, ESG Lab navigated through the process of configuring an automated backup job. The process consists of five simple steps:

  1. Naming the job.
  2. Selecting the virtual machines to protect.
  3. Defining the location to store backup images.
  4. Setting guest processing attributes.
  5. Scheduling the job.

As shown in Figure 8, ESG Lab used the HOME tab in the Veeam UI to select the backup job icon, which presented options for a VMware or Hyper-V type backup job. We chose to add a new VMware backup and used the configuration wizard to complete the process. First, we assigned a name and description for the new job. Then, the Add Object screen presented us with a list of virtual machines from the test VMware environment. We added a single VM with Microsoft SQL for our test configuration. Next, but not shown in Figure 8, we selected the previously configured backup repository, E:\Backup, on the Cisco S3260 as the storage location. We used the Guest Processing screen to set application-aware processing for the SQL instance. Veeam 9.5 supports application-aware backups for Microsoft Active Directory, Exchange, SharePoint, SQL, and Oracle Databases. Next, the schedule screen, also not shown in Figure 8, was used to configure when we wanted the backup to run and how long we wanted to retain the data. To complete the process, we reviewed the settings and committed the changes.

Figure 8. Backup Job Creation

Source: Enterprise Strategy Group, 2017

Finally, as shown in Figure 9, ESG Lab reviewed the Veeam data protection workflow. Veeam is an image-level data protection solution that leverages hypervisor-based snapshots in VMware and Microsoft Hyper-V environments. It communicates with a host system at the hypervisor level to efficiently transfer and store only the data needed to fully protect the production environment’s virtual machines and applications.

For VMware environments, Veeam leverages the built-in change block tracking (CBT) on the source to efficiently transfer data. For legacy Microsoft Hyper-V environments, Veeam developed its own filter driver to enable change block tracking efficiency. Now, in Hyper-V 2016, Veeam leverages the built-in resilient change tracking (RCT) feature. Once the protected data has been transferred, the Veeam application handles deduplication and compression at the backup repository to efficiently store backup images.

When configuring backup jobs, Veeam users can choose between three types of incremental backups, Forever Forward—the default—creates a full backup in the first session, then copies only data blocks that have changed in subsequent sessions. Forward Incremental backups use a traditional incremental schema that leverages periodic Synthetic or Active Full backups. Forward Incremental also works well for data sets that require additional copies such as those that require an offsite copy. The Reverse Incremental Backup is ideal for disk-to-disk backups. It uses a rolling full type schema where incremental changes are injected into the latest backup to create a full image and changed data is stored as an incremental backup file for point in time recoveries. As shown in Figure 9, the backup copy job is created by Veeam from data in the backup repository. No resources are needed from the production environment to create a copy.

Figure 9. Data Protection Workflow Process

Source: Enterprise Strategy Group, 2017

Why This Matters

When it comes to justifying IT investment, reducing operational expenses is an important consideration.[4] And ease of management and deployment can play a major role in reducing operational expenses. This is epically true in data protection environments where a multitude of very different operations need to be conducted. Because these operations do not always occur on a regular basis, it’s easy to get rusty in their execution. This can quickly add time and effort and, in the worst-case, lead to failures.

ESG Lab confirmed that the Veeam/Cisco backup solution with its familiar, easy-to-use management interface and preconfigured deployment makes it easy to meet these challenges. With Veeam and Cisco, configuring a backup job, backup copy job, replication, or even instant recovery looks and feels the same, and scaling that solution is just as easy as the preconfigured initial deployment process.

 

Operational Efficiency

From its inception, Veeam Backup & Replication was designed to seamlessly integrate with and enhance the native data management features of virtual environments to deliver advanced data protection. In this section of the report, ESG Lab demonstrates how the Veeam/Cisco data protection solution helps improve operational efficiency by expanding its capabilities, beyond simple backup and restore, to handle more advanced data protection workflows and even secondary workloads such as test and development.

ESG Lab Validated

ESG Lab started validating the operational efficiency capabilities of the solution by reviewing the Instant VM Recovery process, a feature added to Veeam Backup & Restore in 2010 with the release of version 5.0. As shown in Figure 10, ESG Lab used the Veeam Restore Wizard to initiate the process. The wizard walks the user through all the recovery steps, from selecting the VM to be restored all the way through to the desired recovery location.

Veeam Instant VM Recovery uses the backup images stored in the repository on the backup server to instantly recover a VM to operational status, minimizing RTO. Here, the backup images highlighted in the green callout box are presented to the virtual machine host as a datastore from which VM3, also highlighted in green above the hypervisor, boots up and runs. While the virtual machine is running in instant recovery mode, all reads come from the backup images located in the backup repository on the Veeam backup server—in this case the Cisco S3260. All the writes while in instant recovery mode go to a redo log on the repository. To return the VM to full production, data is migrated—in the background— from the backup image and redo log to the production datastore. ESG Lab performed an instant restore and the virtual machine was completely recovered, booted, and running in less than two minutes.

Figure 10. Instant Recovery

Source: Enterprise Strategy Group, 2017

Next, ESG Lab explored the use cases for a Veeam Virtual Lab. A Veeam Virtual Lab allows users to power up VMs in the production environment in much the same way Instant VM Recovery works. However, the Virtual Lab environment keeps the VMs on an isolated virtual network. This enables identical VMs or groups of VMs to run on the same host system without interfering with one another.

As shown on the right side of Figure 11, ESG Lab used a Veeam configuration wizard to set up a Virtual Lab named vLAB. The wizard walked us through all the required steps including assigning a name, selecting the host, creating datastores, creating a proxy, and setting up all the networking. A critical component to a Virtual Lab is the proxy server. This is a Linux VM that runs from the Virtual Lab datastore and provides all the network fencing required to isolate the Virtual Lab environment. The proxy also provides a network tunnel to the virtual lab environment so that users outside of the lab environment can gain access. The proxy is connected to both the production network and the virtual lab network.

The left side of Figure 11 shows the VMware environment used for the validation report from the vSphere client view. The black callout box highlights two virtual machines, Srv41-VM01 in the production environment that have been previously backed up by Veeam and the vLAB virtual machine Srv41-VM01_3c45… At the bottom of the figure is the Cisco S3260 Storage Server with a repository that contains backup images for virtual machines in the vSphere environment. The green callout box highlights the Virtual Lab proxy server that isolates the vLAB virtual machines from the production network.

Figure 11. Virtual Lab Configuration

Source: Enterprise Strategy Group, 2017

It should be noted that Srv41-VM01 and Srv41-VM01_3c45… are, from a configuration standpoint, identical. The reason they can run in the same VMware environment is due to the network fencing provided by the proxy server in the Virtual Lab environment. This exercise demonstrates how a Veeam Virtual Lab could be used in a test scenario to instantly provide a sandbox identical to a production system to test for ransomware, or reduce the impact of application updates and patches.

Veeam SureBackup relies on Virtual Labs to perform backup validation. For test and development, the Virtual Lab can be powered on indefinitely. It should be noted that the backup images are not altered, and they remain in the backup repository, for their defined retention period, ready to be used for recovery tasks.

Finally, ESG Lab reviewed the Veeam/Cisco solution features that contribute to efficient, scalable data protection storage. The solution’s storage efficiency starts at the beginning of the protection process with the incremental forever backup schema and change block tracking. Once the data is transferred, Veeam applies deduplication and compression to the backup images stored on the high-density Cisco S3260 backup repository. Now, as shown in Figure 12, the solution supports the ReFS file system for the backup repository. This feature helps deliver more capacity optimization by allowing multiple Synthetic Full backups to share the same data blocks. For utilization monitoring, Veeam ONE provides detailed visibility of the backup repository to help with capacity planning.

Figure 12. Efficient Data Storage

Source: Enterprise Strategy Group, 2017

Why This Matters

IT organizations are constantly being asked to do more with less. Data protection specialists and the applications they support are not exempt from this mandate.

ESG Lab confirmed that when it comes to improving operational efficiency, the Veeam/Cisco backup solution is ready to deliver. The combined solution helps data protection professionals deliver multiple workloads from their data protection solution with the right resources in the right location. These workloads include true backup image validation, support for test and development, disaster recovery, business continuity and sandboxing for security. Patches and updates can be tested on an exact replica of production providing a key CAPEX benefit: eliminate the need to purchase a separate test environment, and an equally important OPEX benefit: eliminate the need to maintain a test environment to match production. This eliminates the risk of differences between production and test that invariably show up as issues when changes are made to production.

Finally, even though the solution efficiently leverages existing backup images for these workloads and environments, it does not alter the images. The images therefore remain ready for recovery when needed.

 

 

ESG Lab Validation Highlights

  • ESG Lab is pleased to see an integrated data protection solution from Veeam and Cisco. The solution makes it easier to deploy and scale Veeam Availability Suite environments. ESG Lab believes the Cisco UCS S3260 Storage Server was a wise choice. It enables better use of advanced Veeam features by delivering the right compute, memory, and storage resources at the right place in the protection infrastructure.
  • ESG Lab also found the combined solution was well suited to handle multiple workloads. The Cisco S3260 had the resources to effortlessly spin up a Virtual Lab environment during our validation project. Virtual Lab environments easily handled workloads such as test and development and backup validation, and backup copy jobs without putting any load on the production environment.
  • Veeam and Cisco delivered economic value on multiple levels, delivering a significantly lower cost per TB than all other vendors examined by ESG Lab. Price/performance was also excellent, delivering 10.44 TB per hour ingest performance and a 49% lower $/GB/hour cost.
  • ESG Lab has repeatedly tested Veeam recovery capabilities and found them to be fast, easy to use, and reliable. Organizations can recover at an extremely granular level, from whole VMs down to individual files and application objects, very quickly.

Issues to Consider

  • Veeam software and the Cisco UCS S3260 Storage Server solutions both have a lot of great standalone features and functions that any data protection environment can benefit from. Simply adding a jointly validated reference architecture can give customers and their business partners a completely customizable kit for their data protection environment. However, those that want the simplicity of a turnkey solution could benefit from pre-configured and optimized bundles that are ready to deploy from channel partners.
  • The test results presented in this report are based on applications and benchmarks deployed in a controlled environment with industry-standard testing tools. Due to the many variables in each production data center environment, capacity planning and testing in your own environment are recommended.

The Bigger Truth

Modernizing the data center—from leveraging more application virtualization to implementing converged infrastructures and expanding implementations to include business- and mission-critical applications—will increase operational agility, facilitate higher availability, and reduce equipment, power, cooling, facilities, and management costs through consolidation. Organizations put themselves at risk if they expect their traditional data protection solutions to be able to keep pace. Traditional methods are slow to evolve, leaving open an opportunity for new solutions that are built from the ground up to support the new way of doing things, such as highly virtualized data centers (of all sizes, including private and public cloud providers).

Veeam Availability Suite with Cisco UCS is designed to bridge the gap by providing customers with a differentiated solution that offers more than just backup and recovery. Using virtualization, storage, and cloud technologies, Cisco and Veeam deliver high-speed recovery, data-loss avoidance, and verified protection with complete visibility via Cisco’s proven UCS platform for applications requiring high availability and scalability.

ESG Lab validated that the Veeam and Cisco solution with its familiar, easy-to-use management interface and preconfigured deployment made configuring a backup job, backup copy job, replication, or even instant recovery fast and easy, and scaling that solution is just as easy as the preconfigured initial deployment process.

ESG Lab also confirmed that when it comes to improving operational efficiency, the Veeam and Cisco combined solution helps data protection professionals deliver multiple workloads from their data protection solution with the right resources in the right location. These workloads include true backup image validation, support for test and development, disaster recovery, and business continuance. While the solution efficiently leverages existing backup images for these workloads, it does not alter the images, so they remain ready for recovery when needed.

Finally, ESG Lab validated that the Veeam and Cisco solution provides not just great performance that shrinks backup windows and reduces risk, but compelling price/performance, delivering 10.44 TB per hour of ingest performance at 49% lower cost than other vendors evaluated by ESG Lab.

When truly modernizing your IT strategy, protection and production must evolve within a single plan. The key criteria of this plan are reliability, manageability, and agility. Cisco and Veeam are delivering agility that is powered by the innovative “plumbing,” including the UCS fabric interconnects and Veeam’s approach for backing up VMs. The combined solution represents significant investments in well-received management interfaces and the reputation of delivering a more reliable solution than alternatives.

Through hands-on testing, ESG Lab can state unequivocally that the Veeam and Cisco solution lives up to these claims. The features we tested demonstrate clearly that Veeam Availability Suite addresses the key challenges of IT managers and the needs of users, and when combined with Cisco UCS, adds compelling performance and price/performance.

Appendix

Table 1. ESG Lab Test Bed

  1. Source: ESG Research Report, 2015 Trends in Data Protection Modernization, September 2015.
  2. Source: ESG Research Report: The Shift Toward Data Protection Appliances, March 2015.
  3. Source: ESG Research Report, 2015 Trends in Data Protection Modernization, September 2015.
  4. Source: ESG Research Report, 2016 IT Spending Intentions Survey, February 2016.

Download the PDF version of this report

Topics: Data Protection