1. Backup the entirety of the VMware ESX server and all of its guest operating systems at one time
2. Load an agent on every guest and treat each of them like a physical machine
3. Skirt the issue entirely by using storage array-based snapshots.

• Leverage information gathered from multiple different organizations to create a common repository of information that everyone can benefit from
• More readily disseminate information throughout multiple organizations to include Symantec's own support team
• Proactively notify companies when they are running configurations that have resulted in problems in other companies and under what circumstances
• Creating a collaborative, cloud-based service offering so companies can confidently outsource their function to Symantec and yet know their needs will be met

The general economic malaise of the past few months is not going unnoticed by anyone as it seems every day more companies are cutting back and tightening their belts in anticipation of a lean 2009. Just in the last months, numerous companies including 3M, Dow Chemical, and Hewlett-Packard, just to name a few, have announced cutbacks in staffing. But for those individuals that remain, the task does not get any easier. Most if not all end-users that I talk to are getting a hard push by their IT executives to cut costs as the days of simply purchasing more infrastructure is an unacceptable solution.

The need to maximize one's infrastructure investment is becoming a requirement for IT departments in 2009. To deliver on this, some are not hesitating to lean on their IT vendors to help them get the most benefit out of what they already own or plan to buy. In response to this, IT vendors are stepping up to the plate to provide these remaining individuals some guidance on how to best maximize what they still own. For instance, at Gartner's recent Data Center conference, Symantec offered some key insights on just how to cut costs by providing five actionable points:

1. Stop Buying Storage
2. Adopt Thin Provisioning
3. Think Like a Web Company
4. Leverage Deduplication
5. Migrate Production Applications to VMware

Stop Buying Storage and Start Managing it

Storage resource management (SRM) tools have gotten a bad rap over the last few years but the effective use of SRM tools like Veritas Command Central Storage can provide companies with a solid understanding of what storage is actually in use versus what is allocated to the hosts. In doing so, companies can obtain accurate capacity utilization metrics and then can make informed determinations about what storage is over-provisioned, what storage is under-provisioned, what storage is misused, what storage is orphaned and what is unclaimed or unused storage capacity. Once companies quantify their storage environment and how it is used, companies can then make more informed and accurate predictions as to future purchases and then only buy exactly what they need.

Adopt Thin Provisioning

Most companies have too much capacity in their storage environment that not only goes unused but consumes extra power and rack space. Eliminate some of that overhead by not only implementing storage systems that support thin provisioning but a thin aware volume manager and file system as well. Using storage systems that support thin provisioning in conjunctions with Veritas Storage Foundation and the Veritas Thin Reclamation API and Smart Move features can help to significantly drive up storage utilization (the 80% range is realistic) while driving down costs across your storage environment by only maintaining the amount of storage that your environment actually needs.

Think like a Web Company

Companies in the web business focus on three areas when building their infrastructure: massive scalability; cost-effective infrastructures and providing just enough in disaster recovery resources to maintain their business. Using basic and readily available software tools in your environment, you can accomplish tasks and avoid the need to purchase solutions that run on purpose built hardware. As an example, using your already paid for Veritas Storage Foundation for Windows, you can convert a readily-available Windows server into a file server which can eliminate the need to spend extra capital on a specialized NAS storage subsystem. The same principle can be applied to DR. Taking the time to understand your applications and their associated disaster recovery (DR) processes, you can prioritize what applications to recover and only have to perform DR on the ones required to return your business to service.

Leverage Deduplication

Symantec now provides deduplication features in all layers of NetBackup through its integration of Pure Disk. Companies can now realize space reductions at every level in their enterprise including the data center, remote offices and VMware environment where the amount of redundancy just screams for space savings. In doing so, companies can see significant cost savings in data storage, WAN/LAN bandwidth utilization, backup administrative costs and the elimination of tape devices in remote locations.

Migrate Production Applications to VMware

VMware is not a new concept or technology but using it for production applications is a place where organizations still tread lightly. Symantec's introduction of Veritas Cluster Server (VCS) into VMware and using it to protect virtual machines puts most, if not all, of these fears to bed. VCS for VMware provides application layer aware high availability, something that currently does not exist inside of VMware. It also includes other features like global protection, automated DR testing, and quick failover. So applications that probably were not great candidates for VMware in the past due to strict SLA requirements may now be virtualized with no impact to the customer's expected availability. The good news is that in taking this step, it adds another layer of cost savings as many of the production applications that once resided on dedicated physical hardware can now reside in the virtual world.

Each of these approaches has its pros and cons so it really depends on what you are trying to accomplish with your backup and recovery environment. In fact, that's really the point. It depends upon the backup goals and constraints for each server, and the problems you are trying to resolve so using any and all of these three deduplication options in your backup environment might make sense.

Now that Symantec has integrated PureDisk into its NetBackup product suite, companies can deduplicate data at any of these levels. In the past, PureDisk primarily solved the problem of backing up remote offices by deduplicating data at these locations but this new integration ratchets up the features available for your backup and recovery environment.

NetBackup has always given the companies the freedom to pick and choose whatever disk storage system they desire to use as a backup target. Now with the incorporation of PureDisk into NetBackup, that same freedom is extended to now place deduplicated data on whatever storage device that companies so desire without the requirement to choose a specific deduplication appliance. Further, in NetBackup 6.5, Symantec modified PureDisk so it does more than just deduplicate data on clients but it can deduplicate data anywhere in the backup stream according to where it makes the most sense in your backup requirements, all the way down to the application level.

For example, it makes sense to keep deduplication as a client side process for host systems that have either limited bandwidth to the media server or applications that consume minimal resources on the server during backup times, such as a departmental file or database server. The client approach is also attractive for heavily-consolidated virtual machine environments, to reduce the backup load on the consolidated physical servers.

After determining which backup clients require client-side data reduction, the next decision to make is which ones will benefit from deduplication at the media server layer. These clients are characterized as systems that have adequate bandwidth to the media server but have insufficient processing power to efficiently deduplicate backup data on their own in the time required. By directing the client's backup stream to the media server, deduplication is off-loaded to the media server which deduplicates the data in-line as the backup occurs.

However, even this option may add too much time to the backup window of certain servers so PureDisk also includes an option to deduplicate data after the backup job is complete. Using this option, the backup to disk occurs to high-speed disk without interruption (i.e. - no slow down to deduplicate the data during the backup window) and PureDisk deduplicates the data using a post-processing algorithm. Regardless of which option or options that enterprise companies need, NetBackup allows companies to pick and choose which data reduction model that best meets a specific application server's needs, be that host side, in-line or post-processing.

Incorporating PureDisk into NetBackup also compliments its existing capabilities to recover your application server data remotely without spending extra money for hardware and software to accomplish this. Using this solution, companies can now replicate deduplicated PureDisk backup data stores to a remote location without needing to change their existing backup processes or reconfigure their DR facility. Equally important during these tough economic times, replicating deduplicated data requires less bandwidth so companies many need smaller WAN pipes to handle data movement which could potentially lower recurring monthly costs.

Finally, lest we forgot, VMware is also supported by NetBackup PureDisk which is an environment that just screams for data reduction at the guest level, ESX Server level or both. Whether you are using NetBackup at the guest level, or incorporating it with VCB, the PureDisk option offers a huge reduction in space on the disk target, just based on the redundancy of files inside of VMware. Going with a guest approach offers the best of both worlds, as you enable guest based clients, you will greatly limit the amount of data and bandwidth you are using across your network, as well as saving critical bandwidth resources for the ESX server itself. This approach typically makes a lot of sense based on your deployment of VMs: the more standard images they have the more reduction you can achieve.

Companies can debate all day long about which deduplication approach is best but at the end of the day, in this economic climate, companies are looking to debate less and cut costs more. NetBackup PureDisk 6.5 takes the deduplication debate off the table and gives them the flexibility to select the most appropriate deduplication option regardless of the application. Symantec's decision to incorporate PureDisk into NetBackup 6.5 capitalizes on NetBackup's support of heterogeneous storage infrastructure while giving companies newfound abilities to deduplicate data at any level in their backup infrastructure.

An area that is often overlooked in an IT infrastructure, at least until it's needed, is the backup and recovery environment. Then when the realization hits the company that it needs backup software, it's typically complex to install, configure and maintain, even in small environments, because of the fact that backup consists of so many moving parts (backup servers, tape robots, disk-based arrays, SAN networks, etc.). The good news is that more hardware and software vendors are stepping up to the plate and partnering to take some of the complexity out of installing and configuring backup software in these size environments. The most recent announcement between Dell and Symantec is the latest in the growing number of symbiotic relationships between hardware and software vendors in the backup space.

The recent joint launch between Dell and Symantec of the Dell PowerVault DL2000 Powered by Backup Exec backup and recovery appliance is just one of the fruits of a relationship that goes back more than twelve years. This appliance is much more than just Dell installing Backup Exec 12.5 onto a Dell Server and then selling it with some storage or tape. The integration it offers between the backup software and the underlying hardware should make even a seasoned sys-admin take notice.

One of the first features that caught our eye is that the PowerVault DL2000 can be up and running with a fully operational version of Backup Exec in less than 20 minutes. Equally notable, Backup Exec now also fully controls and manages all the resources contained within the appliance. For example, Backup Exec software now controls the backup and recovery process as you would expect, but also manages and maintains the storage resources within the DL2000 as well.

That is worth repeating. It manages and maintains all of the DL2000's storage resources. That means all of those pesky administrative tasks associated with array management (RAID configuration, discovering new LUNs, provisioning storage, etc.) that administrators normally perform thru a separate interface are now accomplished inside the Backup Exec console. Now that's a backup appliance we can all get behind.

From hardware perspective, the DL2000 provides disk-to-disk-to-tape (D-D-T) right out of the box. The disk layer is Dell's MD1000 shelf packs that allow a customer to scale up to 190 drives and 190 TBs. The beauty of this new solution is that Backup Exec now assumes the management of these components so that disk can be seamlessly added to the system and managed using Backup Exec. Companies can then expand disk storage with whatever capacity or performance characteristics that they need since the MD1000 supports SAS or SATA drives with varying capacity. Connecting to tape is also part of this new integrated solution as it also supports multiple types of connectivity (SAS, FC, SCSI) so companies can manage existing or new tape devices as part of the DL2000 solution.

This addition of the Veritas Thin Reclamation API to Veritas Storage Foundation brings an in-depth and very attractive feature set to those companies that are now using subsystems with thin provisioning or who plan to move to them in the future. The Thin Reclamation API allows the file system and volume manager to effectively interact with a Thin-Aware storage subsystem and provide the capabilities such that when data is removed (e.g., user deletes a file), these deleted files are understood by the storage array as free and unutilized blocks that can be re-absorbed into the available pool of storage.

Now think about that for a minute, an end-to-end solution that ensures your disk subsystem is running at peak utilization levels at all times. This is a significant development especially when you consider the entire process is automated and transparent to the application stack. That is a really awesome concept for those of us who have spent anytime managing storage. Considering that utilization within most storage shops hovers at best around 20 - 40% most of the time, this is something you will definitely want to add to your management tool set to help make you shine during budget planning cycles.

SmartMove is another new addition to Storage Foundation. Unlike the Thin Reclamation API there is no certification involved, it simply works from any server to any thin-storage. This feature enables the movement of data from existing fat LUNs to thin LUNs that reside inside any thin-aware storage subsystem via an online migration process.

As an example, assume you decide to move data to a thin-aware storage array or your existing storage vendor has now added thin provisioning to its existing system. You can simply create those new filesystems on your thin-array(s), perform a dynamic SmartMove migration (with no impact to the applications) from your legacy array(s) and "Bingo!", you now have thin-aware LUNs (and reclaimed the 60% wasted storage capacity from the legacy array). Also during the migration process, SmartMove ensures that it will only send the specific blocks of the volume that are being used by the application of what has changed and not the entire volume. Further, it supports all major enterprise operating systems including Linux, Windows and UNIX as well as file systems such as NTFS or VxFS.

However it is still important to note that the Veritas Thin Reclamation API has just been released and requires your storage system vendor to take part in an integration/certification process with Symantec. Also the Thin Reclamation API does not alone enable you to create a thin volume on a standard array. Rather the Thin Reclamation API provides the linkage needed to take advantage of thin provisioning on storage subsystems that support it.

Now that we have seen the Thin Provisioning API feature from Symantec, it's clear that Symantec is continuing to extend its lead over its competition. Providing real integration with storage systems that offer thin provisioning was sorely needed to harness the power of thin provisioning and enable end-users to take full advantage of this advancement in storage technology. By layering the Thin Provisioning API into the host volume management layer of the Veritas Storage Foundation will only help further spur the widespread corporate adoption of thin provisioning. Already we see storage vendors like 3PAR, HDS, HP and others lining up to take advantage of these latest features within the Veritas Storage Foundation and we see no reason for this trend of storage vendors adopting this technology to do anything but accelerate in the coming years.

Marty Ward, Symantec's Director of Product Marketing for the Data Protection Group, already points to some of the recent successes as a result of this new team. Shortly after they were combined, the two groups started to leverage technologies that already exist across its data protection portfolio so companies of all sizes can look to Symantec to provide a complete solution.

For example, NetBackup's technology for protecting VMware was ported to Backup Exec so small and midsize businesses (SMBs) could achieve the same reduction in backup and recovery times that were previously only found in NetBackup. On the flip side, Backup Exec's patent-pending Granular Recovery Technology that enables it to do a single backup of a Microsoft application like Exchange and recover anything from single message to the entire database was ported to NetBackup. This kind of internal cooperation is helping to expand the addressable market for these technologies and drive key customer solutions to market at a faster pace.

But at higher level, Symantec is looking to accomplish much more by merging these teams into one product group than just playing nice and sharing code. Instead it is redefining its data protection strategy to provide complete enterprise data protection across virtual environments, remote offices, desktops, laptops, servers, applications and databases.

As part of this, the Data Protection Group is collectively enhancing its focus on disk-based data protection. Right now the worst kept secret in the industry is that every company, regardless of its size, is evaluating using disk in some way as part of their data protection scheme and Symantec is obviously not ignoring this trend. However, when it comes to selecting a disk-based data protection solution, the choice depends on the service levels of your applications, the goals of your organization, and the size of your budget. As a result, companies may end up with an assortment of disk arrays ranging from commodity disk arrays to intelligent disk appliances like virtual tape libraries.

The objective of Symantec's Data Protection Group is to become the management platform no matter what hardware platform you choose. For instance, using Symantec's integrated deduplication and continuous data protection technologies, you can protect data on existing, commodity disk arrays. Conversely, if you have already made an investment in a third party appliance, you can take advantage of Symantec's OpenStorage interface to better integrate and manage the extended features that they offer from the same Symantec software. As an example, both Backup Exec and NetBackup integrate tightly with NetApp's Filers to efficiently backup and recover data from an NDMP Filer snapshot.

Of course, the best data protection strategy in the world is pretty much worthless if you can't recover any of the data you backup. As the final rung in its comprehensive and singular data protection strategy, Symantec centers its approach to recovery on three broad categories: granular application recoveries; local and remote disaster recoveries; and, mission critical application recoveries. While other providers can arguably claim efficient recovery technologies, Symantec does distinguish itself in its patent-pending Granular Recovery technology, which allows for quick recovery of an entire system or individual files, emails, documents and virtual machines from a single-pass image backup of Microsoft Windows, Exchange, SharePoint and VMware servers. This eliminates redundant brick-level backups and shortens recovery times by orders of magnitude.

The Symantec Data Protection Group is creating common strategies, sharing code between products and better positioning its products for its customers. Early indications are that this team has created some new synergies internally while improving the features of the products that it is currently delivering. All it all, it sounds like it is breaking new ground while avoiding the turf wars that such mergers sometimes create.

Legacy clustering methods for high availability (HA) and disaster recovery (DR) have not really kept up with the times. As corporations have started their move into next-generation data center models that take into account virtualization, massive scale and tiered application environments, they are left with incomplete, or at best cobbled together, HA/DR scenarios.

That begins to change with last week's release of VCS One. Though Symantec already has a long running and very successful history in clustering, its current products are still largely rooted in the existing clustering model. With the introduction of VCS One, Symantec takes a very new and unique approach to clustering. VCS One removes the barriers of a traditional clustering engine and creates a new environment where the operating system and the number of systems (physical or virtual) is no longer a factor. In so doing, companies are no longer tied to a single OS clustering solution anymore that only protects the application database in their most critical environment.

VCS One acts as a light weight glue that enables multiple systems to act in a coordinated way to support end-to-end applications - be they physical, virtual or even different operating systems. It's already extremely scalable: in just its first rendition it scales out to 256 nodes versus the standard 32 nodes using normal clustering technology. Another key differentiator is that it no longer is integrated into the OS Kernel so the days of those laborious cluster upgrades are over. All organizations need on their host is a VCS One client that coordinates and reports to the central clustering repository, called the VCS One Policy Master.

Right out of the gate, Symantec provides two very slick use cases for VCS One: Multi-Tier HA and Scale-Out Control. Multi-Tier HA, in our view, is a real need for any customer with Tiered-Applications sitting across multiple OSes. For example, say a customer has an order-entry system where the web front-end is on Windows and the backend database is running on Solaris. Multi-Tier HA provides the integration so that if any portion of the application goes down, it can be restarted in the proper order ensuring quicker return to service.

In some of Tim's past experiences in trying to navigate the boundaries of a Windows team, a Unix team, etc, it was extremely difficult to coordinate the shutdown and proper restart of a tiered application. So instead of wasting time trying to place help desk calls and trying to find the proper people to bring up the application, it can now be driven by an automated policy and initiated by your operations crew.

Scale-Out Control provides the ability of routine or scheduled restarts of applications and servers from a single-pane of glass that again takes into account the transparency of physical, virtual, mixed OS'es, etc. For example, during the evening hours a customer may want to perform some extensive mining on a database that was used during the day, but come morning they want to disable the mining routines to bring back the levels of performance needed for daily operations. Scale-Out Control provides an automated way of performing these functions, and those functions are delivered in a single-management console, allowing the Sys-Admins and Architects to design and implement the configuration, but has the Operations crew perform the function. This saves significant expenses from an operational perspective.

Symantec's long history and extensive experience with clustering is made evident in this release of VCS One. Symantec provides undeniable evidence that they have a firm grasp on what customers need to continue to survive and thrive in the next-generation data center environments. By eliminating the traditional boundaries of clustering and creating an entirely new approach to data center infrastructure management, companies can not only rethink how they manage clustering, but also how they need to think about managing their data center. Yet this is only the beginning and the future roadmap of VCS One has some very attractive features so stay tuned because this product is just getting started.

Now enter VMware. Customers can and should plan to use the advanced features offered by VMware such as VMotion, Storage VMotion, and DRS (Distributed Resource Scheduler). However these features lack a couple of critical components that are needed for application high availability: application awareness as well as the inability to account for unplanned downtime (e.g., user errors).

While VMware's offerings certainly provide a streamlined process for planned downtime, their awareness is limited to the vm layer. This is problematic when one considers that most failovers do not occur at the hardware level but instead occur at the application level and are too often unplanned. In this respect, VCS compliments VMware's offerings since they work together to increase application availability by accounting for both hardware and software mishaps.

VCS accomplishes this by taking the availability of applications running on VMware virtual machines (VMs) to a new, application-aware level. First off, everything you have grown accustomed to with VCS on physical machines still exists when running in a VMware environment, such as local, metro, and long-haul clustering as well as failover. VCS also works with storage controller replication technologies, such as EMC's SRDF and HDS's Universal Replicator, to coordinate movement/ownership of data between different systems and then failover the application to the alternative site. Because it is application aware, it also carries forward its existing extremely deep application integration with applications such as Exchange, Oracle, SAP, SQL Server and others that companies deem necessary when configuring these application servers for HA.

The VCS software puts an application agent on each VM on the physical ESX server that takes away little or none of the performance (CPU, Memory, Network, Storage I/O) the VMs need. In return VCS provides a deeper understanding of how the application uses the VM's and physical machine's resources and is integrated with the ESX service console. If VCS detects that an application is beginning to encounter problems, it can failover to an alternative physical or virtual machine. VCS also removes the requirement for stand-by or idle machine as you can run VCS in an active-active mode of operation. In this state, you can take advantage of the hardware investment at the remote location.

One very exciting feature available in VCS for VMware is the ability to cluster and failover VMware's VirtualCenter server. VirtualCenter acts as the fulcrum of your VMware management environment but it comes with some dependencies in order for it operate. Specifically, the VirtualCenter platform is dependent on Windows and needs a functional VM Guest. In these environments VirtualCenter failover and clustering is an imperative since VirtualCenter runs on a physical machine. Only VCS makes it a reality.

Now let's look at a failure scenario. Assume you have three physical VMware servers; two at a primary site and one at a remote location for DR and Business Continuity with three VMs running on Server 1 at the primary site with VCS configured on all of them. Assume only one application running in VM #2 on Server 1 crashes due to a faulty patch, virus, or worm. At this point, VCS can take over and ensure that the VM guest on the second physical VMware ESX server is in recovered state before bringing it online. It also can ensure the application has not suffered any loss of data either. Now the VM can be seamlessly failed over to the other physical ESX server at the primary location.

Take this a step further and you lose both ESX Servers at the primary site, the shared storage and/or both. Since the data was replicated via EMC SRDF (or other major storage array vendor), VCS will now automatically take over and fail some or all of the VMs and their hosted applications and storage to the remote location--with a single click.

VCS also has a very innovative feature called Fire Drill which lets an end-user test the failover scenarios we've looked at above. This feature ensures that all the resources are operating within their functional means (Storage, Network, Replication, application(s) and VM dependencies). Fire-drill allows you to test every aspect of the cluster failover, with the exception of DNS changes, all in real-time without impacting the availability of your production applications to the business. This can be very useful for validating you disaster recovery and business continuity plans.

Adding VCS into your VMware environment allows companies to move the reliability and availability of their VMware infrastructure to a more complete local and global level. These techniques for clustering VMware will give you and your organization the confidence it needs to deploy critical business applications in virtualized environment without sacrificing the application availability and integration to which they have become accustomed.