You are currently browsing the tag archive for the ‘datadomain’ tag.

Do you have a clearly defined Recovery Point Objective (RPO) for your data?  What about a clearly defined Recovery Time Objective (RTO)?

One challenge I run in to quite often is that, while most customers assume they need to protect their data in some way, they don’t have clear cut RPO and RTO requirements, nor do they have a realistic budget for deploying backup and/or other data protection solutions.  This makes it difficult to choose the appropriate solution for their specific environment.  Answering the above questions will help you choose a solution that is the most cost effective and technically appropriate for your business.

But how do you answer these questions?

First, let’s discuss WHY you back up… The purpose of a backup is to guarantee your ability to restore data at some point in the future, in response to some event.  The event could be inadvertent deletion, virus infection, corruption, physical device failure, fire, or natural disaster.  So the key to any data protection solution is the ability to restore data if/when you decide it is necessary.  This ability to restore is dependent on a variety of factors, ranging from the reliability of the backup process, to the method used to store the backups, to the media and location of the backup data itself.  What I find interesting is that many customers do not focus on the ability to restore data; they merely focus on the daily pains of just getting it backed up.  Restore is key! If you never intend to restore data, why would you back it up in the first place?

What is the Risk?

USA Today published an article in 2006 titled “Lost Digital Data Cost Businesses Billions” referencing a whole host of surveys and reports showing the frequency and cost to businesses who experience data loss.

Two key statistics in the article stand out.

  • 69% of business people lost data due to accidental deletion, disk or system failure, viruses, fire or another disaster
  • 40% Lost data two or more times in the last year

Flipped around, you have at least a 40% chance of having to restore some or all of your data each year.  Unfortunately, you won’t know ahead of time what portion of data will be lost.  What if you can’t successfully restore that data?

This is why one of my coworkers refuses to talk to customers about “Backup Solutions”, instead calling them “Restore Solutions”, a term I have adopted as well.  The key to evaluating Restore Solutions is to match your RPO and RTO requirements against the solution’s backup speed/frequency and restore speed respectively.

Recovery Point Objective (RPO)

Since RPO represents the amount of data that will be lost in the event a restore is required, the RPO can be improved by running a backup job more often.  The primary limiting factor is the amount of time a backup job takes to complete.  If the job takes 4 hours then you could, at best, achieve a 4-hour RPO if you ran backup jobs all day.  If you can double the throughput of a backup, then you could get the RPO down to 2 hours.  In reality, CPU, Network, and Disk performance of the production system can (and usually is) affected by backup jobs so it may not be desirable to run backups 24 hours a day.  Some solutions can protect data continuously without running a scheduled job at all.

Recovery Time Objective (RTO)

Since RTO represents the amount of time it takes to restore the application once a recovery operation begins, reducing the RTO can be achieved by shortening the time to begin the restore process, and speeding up the restore process itself.  Starting the restore process earlier requires the backup data to be located closer to the production location.  A tape located in the tape library, versus in a vault, versus at a remote location, for example affects this time.  Disk is technically closer than tape since there is no requirement to mount the tape and fast forward it to find the data.  The speed of the process itself is dependent on the backup/restore technology, network bandwidth, type of media the backup was stored on, and other factors.  Improving the performance of a restore job can be done one of two ways – increase network bandwidth or decrease the amount of data that must be moved across the network for the restore.

This simple graph shows the relationship of RTO and RPO to the cost of the solution as well as the potential loss.The values here are all relative since every environment has a unique profit situation and the myriad backup/restore options on the market cover every possible budget.

Improving RTO and/or RPO generally increases the cost of a solution.  This is why you need to define the minimum RPO and RTO requirements for your data up front, and why you need to know the value of your data before you can do that.  So how do you determine the value?

Start by answering two questions…

How much is the data itself worth?  

If your business buys or creates copyrighted content and sells that content, then the content itself has value.  Understanding the value of that data to your business will help you define how much you are willing to spent to ensure that data is protected in the event of corruption, deletion, fire, etc.  This can also help determine what Recovery Point Objective you need for this data, ie: how much of the data can you lose in the event of a failure.

If the total value of your content is $1000 and you generate $1 of new content per day, it might be worth spending 10% of the total value ($100) to protect the data and achieve an RPO of 24 hours.  Remember, this 10% investment is essentially an insurance policy against the 40% chance of data loss mentioned above which could involve some or all of your $1000 worth of content.  Also keep in mind that you will lose up to 24 hours of the most recent data ($1 value) since your RPO is 24 hours.  You could implement a more advanced solution that shortens the RPO to 1 hour or even zero, but if the additional cost of that solution is more than the value of the data it protects, it might not be worth doing.  Legal, Financial, and/or Government regulations can add a cost to data loss through fines which should also be considered.  If the loss of 24 hours of data opens you up to $100 in fines, then it makes sense to spend money to prevent that situation.

How much value does the data create per minute/hour/day?

Whether or not your data itself has value on it’s own, the ability to access it may have value.  For example, If your business sells products or services through a website and a database must be online for sales transactions to occur, then an outage of that database causes loss of revenue.  Understanding this will help you define a Recovery Time Objective, ie: for how long is it acceptable for this database to be down in the event of a failure, and how much should you spend trying to shorten the RTO before you get diminishing returns.

If you have a website that supports company net profits of $1000 a day,  it’s pretty easy to put together an ROI for a backup solution that can restore the website back into operation quickly.  In this example, every hour you save in the restore process prevents $42 of net loss.  Compare the cost of improving restore times against the net loss per hour of outage.  There is a crossover point which will provide a good return on your investment.

Your vendor will be happy when you give them specific RPO and RTO requirements.

Nothing derails a backup/recovery solution discussion quicker than a lack of requirements.  Your vendor of choice will most likely be happy to help you define them but it will help immensely if you have some idea of your own before discussions start.  There are many different data protection solutions on the market and each has it’s own unique characteristics that can provide a range of RPO and RTO’s as well as fit different budgets. Several vendors, including EMC, have multiple solutions of their own — one size definitely does not fit all.  Once you understand the value of your data, you can work with your vendor(s) to come up with a solution that meets your desired RPO and RTO while also keeping a close eye on the financial value of the solution.

My recent post about Compression vs Dedupe, which was sparked by Vaughn’s blog post about NetApp’s new compression feature, got me thinking more about the use of de-duplication and compression at the same time.  Can they work together?  What is the resulting effect on storage space savings?  What if we throw encryption of data into the mix as well?

What is Data De-Duplication?

De-duplication in the data storage context is a technology that finds duplicate patterns of data in chunks of blocks (sized from 4-128KB or so depending on implementation), stores each unique pattern only once, and uses reference pointers in order to reconstruct the original data when needed.  The net effect is a reduction in the amount of physical disk space consumed.

What is Data Compression?

Compression finds very small patterns in data (down to just a couple bytes or even bits at a time in some cases) and replaces those patterns with representative patterns that consume fewer bytes than the original pattern.  An extremely simple example would be replacing 1000 x “0”s with “0-1000”, reducing 1000 bytes to only 6.

Compression works on a more micro level, where de-duplication takes a slighty more macro view of the data.

What is Data Encryption?

In a very basic sense, encryption is a more advanced version of compression.  Rather than compare the original data to itself, encryption uses an input (a key) to compute new patterns from the original patterns, making the data impossible to understand if it is read without the matching key.

Encryption and Compression break De-Duplication

One of the interesting things about most compression and encryption algorithms is that if you run the same source data through an algorithm multiple times, the resulting encrypted/compressed data will be different each time.  This means that even if the source data has repeating patterns, the compressed and/or encrypted version of that data most likely does not.  So if you are using a technology that looks for repeating patterns of bytes in fairly large chunks 4-128KB, such as data de-duplication, compression and encryption both reduce the space savings significantly if not completely.

I see this problem a lot in backup environments with DataDomain customers.  When a customer encrypts or compresses the backup data before it gets through the backup application and into the DataDomain appliance, the space savings drops and many times the customer becomes frustrated by what they perceive as a failing technology.  A really common example is using Oracle RMAN or using SQL LightSpeed to compress database dumps prior to backing up with a traditional backup product (such as NetWorker or NetBackup).

Sure LightSpeed will compress the dump 95%, but every subsequent dump of the same database is unique data to a de-duplication engine and you will get little if any benefit from de-duplication.   If you leave the dump uncompressed, the de-duplication engine will find common patterns across multiple dumps and will usually achieve higher overall savings.  This gets even more important when you are trying to replicate backups over the WAN, since de-duplication also reduces replication traffic.

It all depends on the order

The truth is you CAN use de-duplication with compression, and even encryption.  They key is the order in which the data is processed by each algorithm.  Essentially, de-duplication must come first.  After data is processed by de-duplication, there is enough data in the resulting 4-128KB blocks to be compressed, and the resulting compressed data can be encrypted.  Similar to de-duplication, compression will have lackluster results with encrypted data, so encrypt last.

Original Data -> De-Dupe -> Compress -> Encrypt -> Store

There are good examples of this already;

EMC DataDomain – After incoming data has been de-duplicated, the DataDomain appliance compresses the blocks using a standard algorithm.  If you look at statistics on an average DDR appliance you’ll see 1.5-2X compression on top of the de-duplication savings.  DataDomain also offers an encryption option that encrypts the filesystem and does not affect the de-duplication or compression ratios achieved.

EMC Celerra NAS – Celerra De-Duplication combines single instance store with file level compression.  First, the Celerra hashes the files to find any duplicates, then removes the duplicates, replacing them with a pointer.  Then the remaining files are compressed.  If Celerra compressed the files first, the hash process would not be able to find duplicate files.

So what’s up with NetApp’s numbers?

Back to my earlier post on Dedupe vs. Compression; what is the deal with NetApp’s dedupe+compression numbers being mostly the same as with compression alone?  Well, I don’t know all of the details about the implementation of compression in ONTAP 8.0.1, but based on what I’ve been able to find, compression could be happening before de-duplication.  This would easily explain the storage savings graph that Vaughn provided in his blog.  Also, NetApp claims that ONTAP compression is inline, and we already know that ONTAP de-duplication is a post-process technology.  This suggests that compression is occurring during the initial writes, while de-duplication is coming along after the fact looking for duplicate 4KB blocks.  Maybe the de-duplication engine in ONTAP uncompresses the 4KB block before checking for duplicates but that would seem to increase CPU overhead on the filer unnecessarily.

Encryption before or after de-duplication/compression – What about compliance?

I make a recommendation here to encrypt data last, ie: after all data-reduction technologies have been applied.  However, the caveat is that for some customers, with some data, this is simply not possible.  If you must encrypt data end-to-end for compliance or business/national security reasons, then by all means, do it.  The unfortunate byproduct of that requirement is that you may get very little space savings on that data from de-duplication both in primary storage and in a backup environment.  This also affects WAN bandwidth when replicating since encrypted data is difficult to compress and accelerate as well.

Well, not exactly.  What you really need is a restore solution!

I was discussing this with a colleague recently as we compared difficulties multiple customers are having with backups in general.  My colleague was relating a discussion he had with his customer where he told them, “stop thinking about how to design a backup solution, and start thinking about how to design a restore solution!”

Most of our customers are in the same boat, they work really hard to make sure that their data is backed up within some window of time, and offsite as soon as possible in order to ensure protection in the event of a catastrophic failure.  What I’ve noticed in my previous positions in IT and more so now as a technical consultant with EMC is that (in my experience) most people don’t really think about how that data is going to get restored when it is needed.  There are a few reasons for this:

  • Backing up data is the prerequisite for a restore; IT professionals need to get backups done, regardless of whether they need to restore the data.  It’s difficult to plan for theoretical needs and restore is still viewed, incorrectly, as theoretical.
  • Backup throughput and duration is easily measured on a daily basis, restores occur much more rarely and are not normally reported on.
  • Traditional backup has been done largely the same way for a long time and most customers follow the same model of nightly backups (weekly full, daily incremental) to disk and/or tape, shipping tape offsite to Iron Mountain or similar.

I think storage vendors, EMC and NetApp particularly, are very good at pointing out the distinction between a backup solution and a restore solution, where backup vendors are not quite as good at this.  So what is the difference?

When designing a backup solution the following factors are commonly considered:

  • Size of Protected Data – How much data do I have to protect with backup (usually GB or TB)
  • Backup Window – How much time do I have each night to complete the backups (in hours)
  • Backup Throughput – How fast can I move the data from it’s normally location to the backup target
  • Applications – What special applications do I have to integrate with (Exchange, Oracle, VMWare)
  • Retention Policy – How long do I have to hang on to the backups for policy or legal purposes
  • Offsite storage – How do I get the data stored at some other location in case of fire or other disaster

If you look at it from a restore prospective, you might think about the following:

  • How long can I afford to be down after a failure?  Recovery Time Objective (RTO): This will determine the required restore speed.  If all backups are stored offsite, the time to recall a tape or copy data across the WAN affects this as well.
  • How much data can I afford to lose if I have to restore? Recovery Point Objective (RPO):  This will determine how often the backup must occur, and in many cases this is less than 24 hours.
  • Where do I need to restore the application? This will help in determining where to send the data offsite.

Answer these questions first and you may find that a traditional backup solution is not going to fulfill your requirements.  You may need to look at other technologies, like Snapshots, Clones, replication, CDP, etc.  If a backup takes 8 hours, the restore of that data will most likely take at least 8 hours, if not closer to 16 hours.  If you are talking about a highly transactional database, hosting customer facing web sites, and processing millions of dollars per hour, 8 hours of downtime for a restore is going to cost you tens or hundreds of millions of dollars in lost revenue.

Two of my customers have database instances hosted on EMC storage, for example, which are in the 20TB size range.  They’ve each architected a backup solution that can get that 20TB database backed up within their backup window.  The problem is, once that backup completes, they still have to offsite the backup, and replicate it to their DR site across a relatively small WAN link.  They both use compressed database dumps for backup because, from the DBA’s perspective, dumps are the easiest type of backup to restore from, and the compression helps get 20TB of data pushed across 1gbe Ethernet connections to the backup server.  One of the customers is actually backing up all of their data to DataDomain deduplication appliances already; the other is planning to deploy DataDomain.  The problem in both cases is that, if you pre-compress the backup data, you break deduplication, and you get no benefit from the DataDomain appliance vs. traditional disk.  Turning off compression in the dump can’t be done because the backup would take longer than the backup window allows.  The answer here is to step back, think about the problem you are trying to solve–restoring data as quickly as possible in the event of failure–and design for that problem.

How might these customers leverage what they already have, while designing a restore solution to meet their needs?

Since they are already using EMC storage, the first step would be to start taking snapshots and/or clones of the database.  These snapshots can be used for multiple purposes…

  • In the event of database corruption, or other host/filesystem/application level problem, the production volume can be reverted to a snapshot in a matter of minutes regardless of the size of the database (better RTO).  Snapshots can be taken many times a day to reduce the amount of data loss incurred in the event of a restore (better RPO).
  • A snapshot copy of the database can be mounted to a backup server directly and backed up directly to tape or backup disk.  This eliminates the requirement to perform database dumps at all as well as any network bottleneck between the database server and backup server.  Since there is no dump process, and no requirement to pre-compress the data, de-duplication (via DataDomain) can be employed most efficiently.  Using a small 10gbps private network between the backup media servers and DataDomain appliances, in conjunction with DD-BOOST, throughput can be 2.5X faster than with CIFS, NFS, or VTL to the same DataDomain appliance.  And with de-duplication being leveraged, retention can be very long since each day’s backup only adds a small amount of new data to the DataDomain.
  • Now that we’ve improved local restore RTO/RPO, eliminated the backup window entirely for the database server, and decreased the amount of disk required for backup retention, we can replicate the backup to another DataDomain appliance at the DR site.  Since we are taking full advantage of de-duplication now, the replication bandwidth required is greatly reduced and we can offsite the backup data in a much shorter period of time.
  • Next, we give the DBAs back the ability to restore databases easily, and at will, by leveraging EMC Replication Manager.  RM manages the snapshot schedules, mounting of snaps to the backup server, and initiation of backup jobs from the snapshot, all in a single GUI that storage admins and DBAs can access simultaneously.

So we leveraged the backup application they already own, the DataDomain appliances they already own, storage arrays they already own, built a small high-bandwidth backup network, and layered some additional functionality, to drastically improve their ability to restore critical data.  The very next time they have a data integrity problem that requires a restore, these customer’s will save literally millions of dollars due to their ability to restore in minutes vs. hours.

If RPO’s of a few hours are not acceptable, then a Continuous Data Protection (CDP) solution could be added to this environment.  EMC RecoverPoint CDP can journal all database activity to be used to restore to any point in time, bringing data loss (RPO) to zero or near-zero, something no amount of snapshots can provide, and keeping restore time (RTO) within minutes (like snapshots).  Further, the journaled copy of the database can be stored on a different storage array providing complete protection for the entire hardware/software stack.  RecoverPoint CDP can be combined with Continuous Remote Replication (CRR) to replicate the journaled data to the DR site and provide near-zero RPO and extremely low RTO in a DR/BC scenario.  Backups could be transitioned to the DR site leveraging the RecoverPoint CRR copies to reduce or eliminate the need to replicate backup data.  EMC Replication Manager manages RecoverPoint jobs in the same easy to use GUI as snapshot and clone jobs.

There are a whole host of options available from EMC (and other storage vendors) to protect AND restore data in ways that traditional backup applications cannot match.  This does not mean that backup software is not also needed, as it usually ends up being a combined solution.

The key to architecting a restore solution is to start thinking about what would happen if you had to restore data, how that impacts the business and the bottom line, and then architect a solution that addresses the business’ need to run uninterrupted, rather than a solution that is focused on getting backups done in some arbitrary daily/nightly window.

I attended an event tonight put on by Optistor Technologies, a local vendor, which was centered on data de-duplication.  Reps from EMC, DataDomain, and SilverPeak were there, chatting with customers about how their respective products leverage de-duplication technology to save you money.  The keynote speaker was Keith Colburn, captain of the Wizard, from Discovery Channel’s Deadliest Catch and they had paired the evening with a spread of crab and prawns to snack on.  Anyway, Keith admitted during his speech that he “doesn’t know a thing about WAN, re-dupe (he meant de-dupe), or any of that stuff” but I think he did an admirable job discussing the importance of picking the right vendors to work with, ones who have adequate support resources and top notch technology.  He related some of the problems he’s had in the Bering Sea and how he attempts to mitigate risk as much as possible but sometimes discovers problems that he never anticipated.  That’s where the tie-ins with vendor choice came together.  All in all it was a fun event and I got a chance to catch up with some of the engineers and account managers I’ve dealt with in the past couple years.

Okay, now that I’ve talked about backing up the datacenter with NetBackup and DataDomain, and backing up remote sites with NetBackup and PureDisk, it’s time to discuss how to get all that data offsite to protect against a catastrophic event at the datacenter.

As mentioned before we have a primary datacenter with the majority of our systems including the backup environment, and a secondary “disaster recovery” datacenter to which we replicate tier 1 applications for business continuity purposes.  Since we really wanted to get away from using tapes and instead store the backups on disk in our datacenter we have a second backup environment in the DR datacenter and we replicate the backup data there.

There are several ways to replicate backup data between two sites but most of them have drawbacks..

1.) Duplicate the backup data from disk to tape and ship the tapes to the remote site to be ready for restore.  This is the easiest and probably cheapest way.  But there’s that pesky tape yet again with it’s media handling and shipping.  And restore could take a while since you have to deal with restoring the catalog from tape, then importing the media, etc.

2.) Duplicate the backup data directly from the local disk to the disk in the second location across the WAN.  This is not very feasible with any significant amount of data because every byte of data that is backed up in the datacenter has to be copied across the much slower WAN.  It could take many days to duplicate a single nights’ backup.  You’d also need a special Catalog backup job that wrote to a storage device across the WAN.  The good here is that the backup application knows there is a second copy of the data and knows how to find it.

3.) Replicate the data with the backup storage devices’s native replication.  Whether it’s PureDisk, Avamar, or DataDomain, pretty much every source-based or target-based deduplication solution has replication built in that leverages the deduplication to reduce the amount of data that traverses the WAN.  The advantage here is that you can have a copy of all of your backup data in a second location in a much shorter time than a traditional copy process.  If your deduplication device stores the data with 10:1 compression, then your WAN usage is reduced by 90%.  The savings in practice is actually better than that.  The drawback is that the backup application (hence the catalog) has no knowledge that there is a second copy of the backup data and after recovering the catalog, you would need to import all of the disk-based media which could take a long time.

4.) Leverage NetBackup Lifecycle Policies with Symantec OpenSTorage (OST) and an OST-capable backup storage system like DataDomain or PureDisk(with PDDO).  Basically this has all the advantages of option #2, where it is a catalog-aware duplication, combined with the advantages of WAN bandwidth savings from option #3.  Time to copy the data offsite is much shorter due to deduplication, and time to restore is very fast since the data is already in the catalog and available on the disk.

OpenSTorage (OST) is a network protocol that Symantec developed to interface with disk-based backup storage systems and DataDomain was an early adopter of OST.  OST allows Netbackup to control replication between OST-capable storage systems and keep track of the replicated copies of backups in the Catalog just as if Netbackup had made both copies itself.  OST is also used as the protocol to send the backup data to the storage device as opposed to CIFS/NFS or VTL.  DataDomain appliances support OST as does PureDisk when used in conjunction with the PDDO option discussed earlier.   In NetBackup, replication controlled by OST is called “optimized duplication” and is controlled primarily through Lifecycle Policies.

Traditionally, when creating NetBackup job policies, the administrator will specify a Storage Unit (either a disk storage unit or a tape library or drive) that the job policy will send backups to.  Lifecycle Policies are treated like Storage Units as far as the Job Policy is concerned but the Lifecycle Policy includes a list of storage units, each with it’s own data retention, that the backup data must be stored onto in order for NetBackup to consider the data fully protected.  Typically there is a “Backup” target which is where the actual data coming from the client is stored, followed by one or more “Duplication” targets.  After the backup job completes, NetBackup will copy the backup data from the “backup” location to all of the “duplication” locations.  This works with pretty much any type of storage and you can mix and match tape and disk in the same policy.  Since these are duplication operations, NetBackup will read ALL of the data from the backup location, and write ALL of the data to each duplication location.  This can take a long time even on the local network and trying to offsite a lot of data over the WAN is not very feasible.

With OST, the lifecycle policy operates exactly the same except that it uses “optimized duplication”, instructing the storage device to copy the file rather than performing the copy through a media server.  So in the case of DataDomain, OST issues the command to the DDR, the DDR then copies the file to the second DDR in the remote site and gets all the benefits of deduplication and compression between the two.  The media server doesn’t actually do any work.  Once the duplication is complete, the DDR notifies NetBackup and the catalog is updated with a record of the second copy of the backup.  Lifecycle Policies are fully automated, you can’t even restart a failed duplication, so in the event of a transient failure like a WAN hiccup NetBackup will retry a duplication job forever until it succeeds in order to satisfy the lifecycle policy.

As you can probably surmise, this is REALLY nice for a tape-less backup environment.  Our DD690 offsites over 9TB of data every night DURING the backup window.  When the last backup job completes, the offsite copies are complete within 30 minutes.  And there is absolutely no management of the offsite process or duplication jobs besides configuring the lifecycle policies up front.  The drawback to regular Netbackup lifecycle policies is that all duplications are taken from the initial backup copy which limits what you can do with the copies.

Enter NetBackup 6.5.4…  Despite the small 6.5.3 -> 6.5.4 version number change, the 6.5.4 release had quite a few new features added.  The biggest one was a revamping of the Lifecycle Policy engine to allow for nested duplications.  Now you can create a copy of a backup, then create multiple copies from the copy, then create copies from the other copies.  Why is this useful?

Remember when I discussed using NetBackup with PDDO to backup remote sites?  Well the data backed up from the remote site is all stored in the primary datacenter and we need to get the second copy to the DR datacenter.  Plus, we wanted to have a small cache of recently backed up data sitting on the remote media server for fast restore.  Well, nested lifecycle’s are the key.  The lifecycle writes the initial backup copy onto the media server’s local disk which is configured as a capacity-managed staging area (ie: it stores as much as it can and expires data when it needs more space for new backups).  The lifecycle then creates a duplicate of the backup onto the PureDisk storage unit in the primary datacenter.  Since bandwidth to the remote site is very limited we don’t want to copy it from the remote site twice so the lifecycle has a second duplication nested under the first to copy it to the DR datacenter.  The source of the second copy is the primary datacenter copy, NOT the remote media server copy.

Where else can we use this?  Let’s consider our tape-less datacenter backups..  We backup the clients to the DataDomain in our primary datacenter, then using a lifecycle policy and OST, create a copy on the DataDomain in the DR datacenter.  If we also wanted to have a tape copy for long term archive or vaulting we could create a nested duplication to make a copy to a tape library in the DR datacenter from the disk copy that is also in the DR datacenter.  Without nested lifecycle’s the only workable solution would be to create the tape in the primary datacenter.  Every copy of the backup made via the lifecycle policy whether it is using OST or not is maintained by the catalog and easily used for restore.  Furthermore, using OST as the protocol between Netbackup and DataDomain actually increases throughput to the DataDomain DDR systems by approximately 2X vs VTL/CIFS/NFS.

Now to the caveats.. Optimized duplication via OST is only available when you are using OST as the protocol between the media server and the storage unit.  This means it doesn’t work with VTL even when the DataDomain IS the VTL.  OST only works over an ethernet network which is why we skipped VTL completely and used 10gbps networks for the DDR connections.  We even skipped VTL/Tape for the NAS systems, connected them directly to the 10gbps network and use 3-way NDMP to backup them up over the network, through the media servers, to the DataDomain.  We get the benefit of lifecycle policies, optimized duplication, and I may have mentioned before–no pesky tape even with NDMP/NAS backups.  And the interesting thing is that with the 10gbps connection, the NDMP dumps are faster than direct fiber to tape.

There were other enhancements to NetBackup 6.5.4 centered around OST functionality but the lifecycle policy improvements were huge in my opinion.

To cover the catalog replication, we run Netbackup hot catalog backups to a CIFS share that is hosted by the DataDomain.  The DDR replicates that share using DataDomain native replication to the DDR in the DR datacenter where the same data is available via a similar CIFS share.  Our standby Netbackup master server is already connected to the CIFS share for catalog restore and connected to the DDR via OST.  A single operation restores the catalog from the replicated copy.  In a real disaster we can begin restoring user data within 30 minutes from the DR datacenter.

I support a very diverse environment with a mix of Windows, Netware, Linux, Solaris, and Mac clients running on standard servers as well as VMWare ESX, plus two different brands of NAS, a few iSeries systems, and an Apple XSAN thrown in for good measure.  We have hundreds of applications running on these systems including SQL, Oracle, MySQL, Sharepoint, Documentum, and Agile.  These applications are mostly contained in our primary datacenter but we also have a few remote datacenters for specific applications and for disaster recovery as well as a couple remote business offices.

Recently I’ve been working on a project to replace our existing backup application with a new one.  We were experiencing extremely long backup windows, low throughput per client, and high backup failure rates with our existing solution and it was time to make a change of some kind.  The goal was to protect all of our systems regardless of their location with both an onsite backup in our primary datacenter and an offsite copy for disaster recovery purposes.  Additionally we wanted to use little or no tape.  After research, lots of vendor meetings, a consulting engagement, and lengthy debate we chose Symantec NetBackup with Symantec NetBackup PureDisk and DataDomain.  This combination was chosen for several reasons which will become clearer below.

For those of you who are not familiar with these products here’s a brief description..

Symantec Netbackup is a traditional backup solution that is designed to move data from many clients, as fast as possible, to disk or tape.  It is similar to EMC Networker, Symantec BackupExec, and any number of other backup products.  NetBackup supports a wide variety of clients, NAS devices, applications (SQL, Exchange, etc), as well as tape libraries and disk storage for the backed up data.  Since it simply copies all of the data that resides on the client directly to the backup server it is not particularly tuned for backing up remote offices across the WAN but it can easily flood a local LAN during a backup.

Symantec NetBackup PureDisk is currently a separate solution from the base NetBackup product; it is designed specifically for backing up data over the WAN.  Puredisk is a “source-dedupe” solution and is very similar in function to EMC’s Avamar product with which I have a long standing love affair. PureDisk performs an incremental-forever style of backup where only the data that changed since the last backup is copied to the backup server.  It then uses deduplication technology to reduce the resulting backup dataset down to an even smaller size before it gets copied across the network.  The data is collected and stored (in it’s deduplicated form) on the backup server.  With this design PureDisk saves network bandwidth as well as disk space on the backup server making it ideal for backups across the WAN, VPN, etc.  Symantec’s goal is to merge PureDisk into NetBackup as a single solution at some point probably next year.  PureDisk backup servers can replicate backed up data to other PureDisk backup servers in de-duplicated form for redundancy across sites. The downside to PureDisk is that raw throughput on a PureDisk backup server is not high enough for datacenter use and client support is more limited than the standard NetBackup product.

DataDomain (now part of EMC) has been making it’s DDR products for a while now and has been very successful (prompting the recent bidding war between NetApp and EMC to purchase the company).  DataDomain appliances are “target-dedupe” devices that are designed to replace tape libraries in traditional backup environments, like Netbackup.  The DDR appliance presents itself as a VTL (virtual tape library) via SAN, a CIFS(Windows) file server, and/or a NFS(UNIX) file server making it compatible with pretty much any type of backup system.  DataDomain also supports Symantec’s OpenSTorage (OST) API which is available in Netbackup 6.5.  The DDR system receives all of the data that Netbackup copies from backup clients, deduplicates the data in real-time, then stores it on it’s own internal disk.  Because the DDR is purpose built and has fast processors it can process data at relatively high throughput rates.  For example, a single DD690 model is rated at 2.7TB/hour (about 6gbps) when using OST.  The deduplication in a DDR provides disk-space savings but does not reduce the amount of data copied from backup clients.  DDRs can also replicate data (in deduplicated form) to other DDRs across the LAN or WAN, great for offsite backups.

For an explanation of de-duplication, check out my prior post on the topic..

Two of the challenges we faced when designing the final solution had to do with the cost per TB of DataDomain disk and the slightly limited client OS support of PureDisk.  But we had a clean slate to work from–there was no interest in utilizing any of the existing backup infrastructure aside from the two IBM tape libraries we had.  We were not required to use the libraries but we wouldn’t be buying new ones if we planned on using tape as part of the new solution.

For the primary datacenter we deployed NetBackup Master and Media servers, a DataDomain DD690, and connected them to each other with Cisco 4900M 10gbps switches.  We deployed a warm-standby master server plus a media server and another DD690 in our DR datacenter but did not use 10gbs there due to the additonal cost.

With this set up we covered all of the clients in our primary datacenter.  Systems that have large amounts of data (like Microsoft Exchange, SAS Financials, etc) were connected directly to the 4900M switches (via 1gbps connections).  Aggregate throughput of the backups during a typical night averages 400-500MB/sec with all of the data going to the DataDomain.  The Exchange server’s flood their network links pushing over 100MB/sec per server when backing up the email databases.  We currently back up 9TB of data per night with 3 media servers and a single DDR in about 5 hours.  Our primary bottlenecks are with the VCB Proxy server (we need more of them) and the aging datacenter core network having an aggregate throughput of a barely more than 1gbps.

But what about those remote sites?  What does OST really add?  How do you tackle the NAS backups without resorting to tape?  All that and more is coming up soon…

This is the 3rd part of a multi-part discussion on capacity vs performance in SAN environments.  My previous post discussed the use of thin provisioning to increase storage utilization.  Today we are going to focus on a newer technology called Data De-Duplication.

Data De-Duplication can be likened to an advanced form of compression.  It is a way to store large amounts of data with the least amount of physical disk possible.

De-duplication technology was originally targeted at lowering the cost of disk-based backup.  DataDomain (recently acquired by EMC Corp) was a pioneer in this space.  Each vendor has their own implementation of de-duplication technology but they are generally similar in that they take raw data, look for similarities in relatively small chunks of that data and remove the duplicates.  The diagram below is the simplest one I could find on the web.  You can see that where there were multiple C, D, B, etc blocks in the original data, the final “de-duplicated” data has only one of each.  The system then stores metadata (essentially pointers) to track what the original data looked like for later reconstruction.

Image from eWeek Article

The first and most widely used implementations of de-dupe technology were in the backup space where much of the same data is being backed up during each backup cycle and many days of history (retention) must be maintained.  Compression ratios using de-duplication alone can easily exceed 10:1 in backup systems.  The neat thing here is that when the de-duplication technology works at the block-level (rather than file-level) duplicate data is found across completely disparate data-sets.  There is commonality between Exchange email, Microsoft Word, and SQL data for example.  In a 24 hour period, backing up 5.7TB of data to disk, the de-dupe ratio in my own backup environment is 19.2X plus an additional 1.7X of standard compression on the post de-dupe’d data, consuming only 173.9GB of physical disk space.  The entire set of backed up data, totaling 106TB currently stored on disk, consumes only 7.5TB of physical disk.  The benefits are pretty obvious as you can see how we can store large amounts of data in much less physical disk space.

There are numerous de-duplication systems available for backup applications — DataDomain, Quantum DXi, EMC DL3D, NetApp VTL, IBM Diligent, and several others.  Most of these are “target-based” de-duplication systems because they do all the work at the storage layer with the primary benefit being better use of disk space.  They also integrate easily into most traditional backup environments.  There are also “source-based” de-duplication systems — EMC Avamar and Symantec PureDisk are two primary examples.  These systems actually replace your existing backup application entirely and perform their work on the client machine that is being backed up.  They save disk space just like the other systems but also reduce bandwidth usage during the backup which is extremely useful when trying to get backups of data across a slow network connection like a WAN.

So now you know why de-duplication is good, and how it helps in a backup environment.. But what about using it for primary storage like NAS/SAN environments?  Well it turns out several vendors are playing in that space as well.  NetApp was the first major vendor to add de-duplication to primary storage with their A-SIS (Advanced Single Instance Storage) product.  EMC followed with their own implementation of de-duplication on Celerra NAS.  They are entirely different in their implementation but attempt to address the same problem of ballooning storage requirements.

EMC Celerra de-dupe performs file-level single-instancing to eliminate duplicate files in a filesystem, and then uses a proprietary compression engine to reduce the size of the files themselves.  Celerra does not deal with portions of files.  In practice, this feature can significantly reduce the storage requirements for a NAS volume.  In a test I performed recently for storing large amounts of Syslog data, Celerra de-dupe easily saved 90% of the disk space consumed by the logs and it hadn’t actually touched all of the files yet.

NetApp’s A-SIS works at a 4KB block size and compares all data within a filesystem regardless of the data type.  Besides NAS shares, A-SIS also works on block volumes (ie: FiberChannel and iSCSI LUNs) where EMC’s implementation does not.  Celerra has an advantage when working with files which contain high amounts of duplication in very small block sizes (like 50 bytes) since NetApp looks at 4KB chunks.  Celerra’s use of a more traditional compression engine saves more space in the syslog scenario but NetApp’s block level approach could save more space than Celerra when dealing with lots of large files.

The ability to work on traditional LUNs presents some interesting opportunities, especially in a VMWare/Hyper-V environment.  As I mentioned in my previous post, virtual environments have lots of redundant data since there are many systems potentially running the same operating system sharing the disk subsystem.  If you put 10 Windows virtual machines on the same LUN, de-duplication will likely save you tons of disk space on that LUN.  There are limitations to this that prevent the full benefits from being realized.  VMWare best practices require you to limit the number of virtual machine disks sharing the same SAN lun for performance reasons (VMFS locking issues) and A-SIS can only de-dupe data within a LUN but not across multiple LUNs.  So in a large environment your savings are limited.  NetApp’s recommendation is to use NFS NAS volumes for VMWare instead of FC or iSCSI LUNs because you can eliminate the VMFS locking issue and place many VMs on a single very large NFS volume which can then be de-duplicated.  Unfortunately there are limits on certain VMWare features when using NFS so this may not be an option for some applications or environments.  Specifically, VMWare Site Recovery Manager which coordinates site-to-site replication and failover of entire VMWare environments does not currently support NFS as of this writing.

When it comes to de-duplication’s impact on performance it’s kind of all over the map.  In backup applications, most target based systems either perform the work in memory while the data is coming in or as a post-process job that runs when the backups for that day have completed.  In either case, the hardware is designed for high throughput and performance is not really a problem.  For primary data, both EMC and NetApp’s implementations are post-process and do not generally impact write performance.  However, EMC has limitations on the size of files that can be de-duplicated before a modification to a compressed file causes a significant delay.  Since they also limit de-duplication to files that have not been accessed or modified for some period of time, the problem is minimal in most environments.  NetApp claims to have little performance impact to either read or writes when using A-SIS.  This has much to do with the architecture of the NetApp WAFL filesystem and how A-SIS interacts with it but it would take an entirely new post to describe how that all works.  Suffice it to say that NetApp A-SIS is useful in more situations than EMC’s Celerra De-duplication.

Where I do see potential problems with performance regardless of the vendor is in the same situation as thin provisioning.  If your application requires 1000 IOPS but you’ve only got 2 disks in the array because of the disk space savings from thin-provisioning and/or de-duplication, the application performance will suffer.  You still need to service the IOPS and each disk has a finite number of IOPS (100-200 generally for FC/SCSI).  Flash/SSD changes the situation dramatically however.

Right now I believe that de-duplication is extremely useful for backups, but not quite ready for prime-time when it comes to primary storage.  There are just too many caveats to make any general recommendations.  If you happen to purchase an EMC Celerra or NetApp FAS/IBM nSeries that supports de-duplication, make sure to read all the best-practices documentation from the vendor and make a decision on whether your environment can use de-duplication effectively, then experiment with it in a lab or dev/test environment.  It could save you tons of disk space and money or it could be more trouble than it’s worth.  The good thing is that it’s pretty much a free option from EMC and NetApp depending on the hardware you own/purchase and your maintenance agreements.

“The opinions expressed here are my personal opinions. Content published here is not read or approved in advance by my Employer (EMC) and does not necessarily reflect the views and opinions of EMC nor does it constitute any official communication of EMC.”

StorageSavvy on Twitter

Follow

Get every new post delivered to your Inbox.

Join 511 other followers