Background Blocks

Resource Library - Whitepaper

Brand Arrow

Protect Thousands of VMs and Your IT Budget

BEST PRACTICES FOR OPTIMIZING VMWARE DATA PROTECTION

As organizations update their existing infrastructure – and design new infrastructure – to support on-premises or cloud data in Software Defined Data Centers (SDDC), their approach to data protection needs to be modernized. By deploying more virtual machines (VMs) as part of their virtualized software-defined environment, they’re able to achieve greater agility, flexibility and market speed. However, with this proliferation of new VMs comes potential for impact on production workloads and data protection operations.

In these new, hybrid environments, having a data protection approach that covers everything – on-premises, cloud, and hybrid – becomes crucial. But providing adequate protection and recovery for these mixed environments can also add unnecessary complexity. With the right tools and some best practices, you can optimize protection for your virtualized environment, and provide a solid foundation that will accommodate future growth and evolving technologies.

Need Info Here

As your VM environment expands, you need to minimize the impact of your protection operations on production workloads, while still meeting your SLAs and protection window. We will look at some best practices that can help you to optimize protection for your virtualized environment. This will allow you to offload data movement away from your production resources – reducing cost, accelerating recovery times, and increasing your capabilities – without introducing the typical risks associated with having unnecessary data exposure. Let’s look at some now:

  • Selecting the right data transport mode
  • Reducing impact on production workloads
  • Mitigating unnecessary risk
  • Reducing backup and restore windows
  • Maintaining SLAs while controlling costs
Brand Arrow

TRANSPORT MODES

To meet your required protection and recovery SLAs, it’s critical that you select the correct transport mode. For example, SAN and Hot Add transport modes can enable LAN-free data movement and are typically faster than local area network (LAN) operations using a network block device (NBD) or secure NBD (NBD-SSL). However, in many cases – because VMs can be categorized into different types based on their size, change rate, and business criticality – utilizing a mixture of techniques can ultimately be more effective.

  • Storage Area Network (SAN) Mode – This mode is supported for direct-connect storage using Fibre Channel (FC) or Internet SCSI (iSCSI) protocols. In SAN mode, a Virtual Server Agent and Media Agent are installed on the same physical computer, with a direct connection. This eliminates the need for data transfer over the LAN, and also provides the best performance for both backups and restores.
  • HotAdd Mode – Here the Virtual Server Agent is installed directly onto a virtual machine. HotAdd refers to the ability to dynamically add virtual or physical hardware to a running system without downtime. This mode can also eliminate the need for data transfer over the network during backup and restore processes.
  • Local Area Network (LAN) NBD and NBDSSL – When no other transport mode is available, storage applications can use LAN transport for data access, either NBD (network block device) or NBDSSL (encrypted). The Virtual Server Agent and MediaAgent are installed on different computers and communicate with each other over the network. The NBD transport method is built into the virtual disk library, is always available, and transmits data over TCP/IP.

Because VMs can be categorized into different types based on their size, change rate, and business criticality – utilizing a mixture of transport mode techniques can ultimately be more effective.

Brand Arrow

REDUCING IMPACT ON PRODUCTION WORKLOADS

vSphere Storage APIs for Data Protection (VADP) is VMware’s data protection framework that enables backup products to do centralized, efficient, off-host LAN-free backup of vSphere virtual machines. This offloads backup processing from ESXi hosts and reduces costs by allowing each ESXi host to run more virtual machines.

Data protection for VADP leverages the snapshot capabilities of VMware vSphere to enable backup across SAN without requiring downtime for virtual machines. As a result, backups can be performed non-disruptively at any time of the day without requiring extended backup windows or the downtime to applications and users typically associated with backup windows.

Ongoing management and integration of snapshots can present the most significant challenge. If you’re like most, you’ll want to be able to convert as many manual, labor-intensive tasks as possible into automated, consistent, and repeatable procedures. When evaluating solutions, you’ll want to look for ways to reduce complexity and enable the management of your hardware and VMware snapshots as part of your overall strategy. With virtualization, you may still require some level of application integration to enable rapid, automated recovery of critical workloads within the virtual machine. In those scenarios, being able to incorporate database and application agents for granular item recovery is highly desirable.

Another key consideration is the ability to recover directly from hardware snapshots. This provides flexibility and enables aggressive Recovery Point Objectives (RPO) and Recovery Time Objectives (RTO). So, for flexibility and cost containment, a snapshot management solution that can support multiple vendors and multiple hypervisors – including cloud hypervisors and snapshots – is critical.

Brand Arrow

MITIGATING RISK

Risk of data exposure is one of the primary reasons that SAN Mode transport might not be your best choice. SAN transport involves exposing your virtual machines to external servers. It also requires management and configuration changes to your storage array and network, which can, of course, introduce additional risk.

When evaluating any solution, you’ll certainly want to consider if it will be cost-effective, but you’ll also want to know that it will be effective across all your environments, and provide full integration with existing systems. Ideally, it should:

  • Easily extend your data protection policies and schedules across mixed virtual environments – including both on-premises and in the cloud – simplifying the ongoing management of your environment
  • Offer a single, scalable – and proven – data platform, one that can eliminate the risks associated with stand-alone products
  • Have minimal impact on your protected virtual machines, causing no disruption to your applications or end-users
Brand Arrow

REDUCING BACKUP AND RESTORE WINDOWS

Many organizations have multiple virtual environments; perhaps yours is one of them. Point products that might work well within a single environment will typically be much less effective in larger, more complex environments, and – more importantly – they aren’t scalable. To improve performance for your VM backup windows, you’ll want to:

  • Eliminate redundant data silos, and minimize the kind of expensive infrastructure that’s typically associated with stand-alone backup and recovery products
  • Consider a platform approach that allows you to scale easily, and can provide backup for as many VMs as required by your enterprise, both now and into the future
  • Automate the application of policy-based rules, in which you define your rules once, and have the platform automatically detect new workloads and assign the appropriate protection policies
  • Enable access to granular recovery options that allow you to easily recover both files and folders
  • Power-up virtual machines directly from your backup to virtually any disk location, without needing to restore it, minimizing impact on your workloads, while providing efficiency and cost-effectiveness
  • Provide user-friendly self-service options that allows users to back-up and recover their own VMs as needed, freeing-up your IT resources for other, higher value activities
  • Automate replication from source VMs to destination VMs, applying incremental changes since the last sync point. This minimizes impact on production workloads and enables point-in-time recovery from older backups, ideal if a source VM has failed or been corrupted

Performance improvements such as these are most easily achievable when utilizing a unified, integrated platform that supports multiple virtual environments. The upside potential is significant savings in IT administrative time; faster VM backup and recovery; and a solid foundation for future growth, not typically available through rudimentary point products.

Brand Arrow

MAINTAINING SERVICE LEVEL AGREEMENTS WHILE CONTROLLING COSTS

Meeting your SLAs requires data portability, along with the ability to interface with several platforms, including multiple clouds and other storage types. You need to be able to move and protect your data seamlessly across multiple destinations, regardless of format.

With vendors such as VMware, Microsoft, Amazon, Google, and others all providing virtual machine services, having the ability to easily integrate with them provides you with choice, flexibility, and control, along with the potential for reduced cost and quicker results.

  • Self-service virtual machine recovery can significantly reduce the demands on staff, freeing your technical experts to focus on more strategic activities and priorities
  • Provide on-demand access to data without waiting for a full restore operation to complete
  • Enable spinning-up virtual machines using protected copy data as the source
  • Support diverse customer environments with flexible VM data protection options. These include guest agents, consistent image VSA, application-integrated VSA, and hardware-assisted data protection
  • Deliver fast VM data recovery from virtually any copy, including disk, on-premises, cloud, tape, or off-premises copy
Brand Arrow

THOUSANDS OF VMS AND COUNTING

As VMware continues to innovate the ways in which enterprises move to a cloud-based digital business model, the proliferation of new virtual machines continues to grow. When looking to protect thousands of VMs – or even more – there are many steps that administrators can take to help them meet their data protection SLAs while avoiding questionable investments in costly, one-dimensional point solutions. By starting with a single, unified platform, you’re well positioned to gain tighter control over your workloads and costs. When evaluating solutions, look for those that can support a diverse VM environment and multiple vendors, and one that enables automated processes for fast, more cost-efficient recovery.

Protect, Monitor and Manage Your Virtualized Data

Read about the six ways you can improve data management across physical, virtual and cloud environments.

  1. Jason Buffington, New Agility Requirements in Data Protection, Enterprise Strategy Group, November 2015
  2. Gartner, Predicts 2017: IT Services Market Opportunities Expand in the Digital Era, October 20, 2016
  3. 2017 State of the Cloud Report, RightScale, 2017
  4. Jason Buffington, New Agility Requirements in Data Protection, Enterprise Strategy Group, November 2015
  5. Gartner, Predicts 2017: IT Services Market Opportunities Expand in the Digital Era, October 20, 2016
Brand Arrow

Protect, monitor and manage your virtualized VMware data.