NetApp - Ask Dr Dedupe

4 Things Frank Slootman Probably Wishes He Didn't Say

Dave Raffo interviewed Frank Slootman this week at SNW.  For those that don't know him, Frank is the former CEO at Data Domain and is now President of EMC's Data Backup Division.  During the interview Frank said things that were, well, un-presidential.  Unfortunately for Frank, we live in a world that records comments like his and preserves them longer than those unidentifiable objects tucked away in the back of your refrigerator.

 

Anyway, Frank made four comments during this interview that stood out as “huh? did he really say that?” types of comments.  I only wish I were in the room so I could have asked some follow up questions.  If I were invited into the room (like that would ever happen) here is the way the interview might have gone:

 

Raffo: But if Disk Library customers want dedupe, they have to buy a Data Domain box?

 

Slootman: Yes. If you have a car and now you want an airplane, are you going to put wings on your car? You have to buy an airplane.

 

DrDedupe:  Frank – I am not sure that is the correct analogy.  A car is designed to travel on the ground, and an airplane is designed to fly through the air.  I think what you meant to say is that if you have a car now and you want a more efficient car, EMC says you should just buy another car, right?

 

Raffo: EMC has sold Quantum's data deduplication software with Disk Library. Will you sell Data Domain software with Disk Library instead?

 

Slootman: No. Disk Library is a straight VTL [virtual tape library] , like it always should have been. It's a brute-force system, no finesse. That's the way it was when it first came out, then they tried to turn it into something it is not by adding deduplication and replication. They bastardized the product, so much so people don't even know what a VTL is anymore. People think VTL is a generic term for backup to disk. People think Data Domain is a VTL, but 90% of the systems we sell are IP-connected, not with a Fibre Channel protocol.

 

DrDedupe:  Thanks for the clarification Frank - EMC was selling a bastardized product and confusing the market.  Good luck in the next all-hands meeting at that division you run now.

 

Raffo: What about the Quantum software EMC has sold with Disk Library?

 

Slootman: We're swapping a lot of those boxes out at zero revenue. We've taken out about a dozen and we'll continue to take out a similar number this quarter. Customers don't want it.

 

DrDedupe:  Wow Frank this is a very interesting disclosure.  Would you care to point out any other products that EMC has sold that your customers don't want?

 

Raffo: Will you continue to work closely with Symantec Corp.'s OpenStorage (OST) API now that you're EMC?

 

Slootman: Yes. I'm not throwing my partners under the bus. We'll compete, but we're all competitors and partners these days. We won't screw them. We'll screw other companies, like CommVault. We {Data Domain] treated them as a good partner and they came after us.

 

DrDedupe:  This is another interesting disclosure Frank.  I am sure our viewers will want to know who the other companies are that you intend to screw.

 

So thats the way the interview might have gone with DrDedupe in the room.  I'm pretty sure that Frank will have some 'splaining to do next time he travels to Hopkinton.  But then again, isn't he starting to sound more like other EMC execs?

 

DrDedupe

The Cost of Efficient Storage can be ZIP

Storage efficiency is a funny thing.  Vendors these days are asking customers to spend money so that they can save money on their data storage.  Huh? This is a bit of a paradox.  Here's the example I use, lets say my wife comes home with a brand new pair of shoes.  "Hey honey, look at these great shoes I just bought!  Regular price was $200 but I got them for 50% off!"  My wife sees this transaction as saving us $100, but I see the transaction as $100 gone from the family checking account, never to be seen again.  Who has the correct perspective?  Well, that all depends on just how necessary those shoes were.  If the closet is overflowing with shoes just like the new ones, this was probably a wasted expense.  However if the shoes were bought to perform a specific function that no other shoe could accommodate (like this), then its probably a worthwhile investment.

Storage efficiency is kind of like that.  You shouldn't invest in it unless it provides some type of new functionality that you don't already own.  Plenty of vendors will offer to sell you cheaper storage, with heavy discounts to entice you, but if its just like all the other storage already in your closet, what are you buying?  More shoes just like the ones you already have?  When you invest in storage efficiency, you should look for something different, something that lets you do things you haven't been able to do before.

Why am I bringing this up?  This week NetApp rolled out a new promo called ZIP, or Zero Investment Promise.  Here's the concept:  instead of buying more shoes, er I mean storage, why not make your existing storage more efficient?  With ZIP, NetApp will loan you a shiny new V-Series Open Storage Controller, but before it hits your dock a promise is made that the V-Series will immediately save you x, (x is defined once we do an onsite evaluation, for free.)  We'll put this promise in writing and give you 90 days to run your tests.  If your tests deliver what we promised, you buy the V-Series (heck its probably already paid for itself by now anyway) - but if you believe we broke our promise, and we agree - the V-Series is yours to keep - gratis.

OK get ready, here comes the fine print.  You and your data center must be in North America (some countries strangely have a problem with promos like this), you cannot already be using V-Series (duh), you have to be a current user of EMC Clariion or HP EVA systems, and you have to agree to use 5 key features on your EMC and HP system, provided via the V-Series (Snapshot, FlexClone, Thin Provisioning, Thin Replication, and of course Deduplication).  That's about it.

So check it out.  And don't assume you have pay alot to bring efficient storage into your world.  In this case, you'll pay ZIP.

DrDedupe

NetApp Replaces OR with AND

I spent last week in New York City, where I was asked to brief many of our largest customers on the merits of deduplication and how it fits into our storage efficiency story.  I also walked the streets of Manhattan in my Dr's garb with a film crew and conducted spontaneous "Man on the Street" interviews, but more on that in another blog...

Anyway back to my briefings.  Most, but not all, of the dozen or so customers I talked to were already implementing deduplication and praised its effectiveness.  A few of the customers I briefed were not actually customers, but rather perspective customers that wanted to learn more about the technology behind NetApp storage efficiency and how it could help them improve their current storage environment.  One thing I've mentioned before but bears repeating is that NetApp deduplication is not limited to NetApp storage systems.  Our V-Series open storage controller brings deduplication and storage efficiency to HP, Sun, EMC and many other arrays.  We didn't make those storage systems, we just made them better.

This brings me to the title of today's blog and the message I always convey during my briefings.  At NetApp, we never ask you to sacrifice performance or resiliency for the sake of reduced storage capacity.  In fact, we'll improve your ability to quickly and safely serve data to your users and applications while at the same time reducing your physical storage footprint.  Thats the message behind our new ad campaign - Replace "Or" with "And" - a clever play on words but one we take to heart.  If you are interested in seeing more - here's a link to our new ad page - including a video I recorded on the topic, yes I do serious videos too.  Also, heres a link to our revamped storage efficiency page featuring yours truly again.  I wonder if the Dr runs the risk of being over-exposed?  Nah you can never see too much of DrDedupe.  Enjoy-

You Lie! Who Coined the Term "Data Deduplication"

Today I came across this quote from Ed Walsh, former CEO of Avamar Technologies:

“Data deduplication was not a term until Avamar used it,” Walsh said. “Data Domain called it capacity optimized storage. At Avamar, we had to teach the market that data deduplication was something that you wanted. Now the market is rife. The technology has proven itself.”

As your purveyor of truth in dedupe, I feel it is my responsibility to get to the bottom of this issue.  I've always given Data Domain credit for coining the term that eventually became my alter-ego. 

So, no I am not going to try and build a case for NetApp.  As stated  out in our award-winning video - "we didn't invent, it we perfected it." - But who really did invent it?  Lets find out together.

First stop, the U.S. Patent Office.  Avamar is listed as the owner of 6 patents.  The oldest, from February 2001, is entitled "Hash File System and Method For Use In Commonality Factoring System."  Whew.  Glad that name didn't last, Dr Commonality Factoring doesn't have the same ring to it.  Well ,anyway, this patent mentions duplicate data but makes no mention of data deduplication.  In fact, after searching all 6 Avamar patents, we find that the word deduplication is never used.

OK, now lets examine Data Domain.  They are listed as the holder of 11 patents.  Their oldest patent is from January 2002 and is entitled "Archival Data Storage System and Method."    Does it mention data deduplication?  Alas, no - it describes an archival storage system with disk drives that spin up and down (an idea, by the way, that seems to have spun down on its own.)

But wait!  A Data Domain patent from July 2006 is entitled "Locality-based Stream Segmentation for Data Deduplication."  The first use of the term belongs to Data Domain!  Do we have a winner?  No, lets just say the score stands at Data Domain 1, Avamar 0, at halftime.  BTW here's some nice halftime entertainment courtesy of the Dr.

OK back to our task. Next - lets try some google searches to see what comes up.  Well, well. Here's an Earnings Release from Avamar dated October 2006, four months after Data Domain's deduplication patent - and an excerpt from that press release: 

"Avamar Technologies, Inc. is a leading provider of enterprise data protection software. Avamar's patented data reduction, single instance store and point-and-click restore technologies have revolutionized backup, recovery and DR strategies for global corporations."

Nary a mention of data deduplication anywhere in this press release.  Now, Ed, if Avamar was using the term data deduplication before Data Domain, wouldn't it have been used in something important like this press release?  Sorry, but I have to declare a shutout here and name Data Domain the winner over Avamar , 2-0.  Ed I believe you've been caught in a little white lie.  Remember, the Dr is watching...

The Dedupe 2.0 Pundits Are Still Swimming in Lake 1.0

I had to chuckle when I saw the concurrent announcements from Permabit and Storage Switzerland that dedupe 2.0 had arrived.  Apparently the acquisition of Data Domain by EMC had signaled the demise of dedupe 1.0, even though I failed to see any technology transformation as a result of that merger.  Data Domain is still doing the same thing they always did and EMC is still doing what they do.  There was no transformative event here guys.

Anyway, the people at Permabit seem to believe that hashing and deduplicating large archival stores is ushering a new era of deduplication and leaving point products behind.  But, er, isn't Permabit another point product?  Last time I looked in a data center there was more than archival data behind the blinking lights on those storage arrays.  In my estimation, deep archival consumes somewhere around 1% of the world's data storage.

Then there's Storage Switzerland.  I guess the Switzerland part is to have one believe that they are a neutral party amidst this vendor-against-vendor world.  But does anyone out there really think they are neutral?  I'll let my readers form their own conclusions.  Anyway, SS says that the core of dedupe 2.0 will be a foundational repository where all previously optimized data will come to rest from primary and secondary storage.  Hmmm, sounds strangely like an archival system to me.  Wait, doesn't Permabit sell archival systems, and they are also talking about dedupe 2.0?  Could there be a connection here?  Say it ain't so Storage Switzerland!

OK lets move back towards reality.  You know when dedupe 2.0 will really arrive?  When people stop talking about dedupe altogether.  Like many technologies that came before it, such as RAID, SCSI, Snapshots, and dozens of others, dedupe will become ubiquitous and acknowledged as a must-have feature from every serious storage vendor.  Inline or Post-process, source or target, local or global - it doesn't matter as long as it dedupes.  Dedupe will run silently in the background, eradicating those pesky duplicate data objects.  NetApp has proven that dedupe can run quite well on a unified platorm that services all tiers of storage (including archival), uses all standard protocols, and serves a wide breadth of applications.  Once a concept is proven, the next steps are constant refinement and optimization, and thats exactly what we are doing.  I can't tell you exactly the day dedupe 2.0 will arrive, but I am pretty sure that NetApp will be among the frontrunners.

DrDedupe

Exchange 2010 Dismisses SIS?

Microsoft Exchange revolutionized the way we communicate.  Its hard to imagine life without email.  Sure there are other email apps out there but aren't they all ladies-in-waiting to the true Queen of email - Microsoft Exchange?

Since version 4.0 was released (about the time the B-52's were singing "Love Shack") Exchange had a cool feature called Single Instance Storage, SIS for short.  SIS was the granddaddy of deduplication, and went something like this:

Email from Joe in Accounting:  "Hey check out this cool bootleg copy of "Love Shack!"

Email from Sally in Marketing, Fred in Shipping, Adam in Sales and about 10 other people "Hey check out what Joe sent me!" 

Well what Joe sent you and you sent to 10 friends and they sent to 10 friends was basically the same email with the same attachment sent over and over again.  The folks in Redmond figured this was going to be a regular thing, so they invented SIS.  With SIS, all those emails were reduced to a single instance, as long as they were held in the same Exchange Store.

Single Instancing saved storage capacity and was free - what's not to like?  Well fast-forward about 15 years and Microsoft had a problem.  More mailboxes, more attachments, and not only bootlegged audio files but attachments of all shapes and sizes - Word Docs, Powerpoints, PDF's, JPEGs, MPEGs, where will it stop?  So Microsoft did what any self-respecting software company would do, they said "you know, storage hardware is cheap these days, and all this SIS stuff is really slowing us down - so please don't use SIS anymore and just buy more storage, OK?

That's right.  Microsoft needed to destroy the monster it had created.  As of Exchange 2007 the message from Redmond was "Given current trends, we expect the value of single instance storage to continue to decline over time. It's too early for us to say whether SIS will be around in future versions of Exchange. However, we want everyone to understand that it is being deemphasized and should not be a primary factor in today's deployment or migration plans."  Worse yet, the word on the streets is that in Exchange 2010, SIS will disappear altogether.

Now before you come down with an acute case of ESISPTD (Exchange SIS Post-Traumatic Disorder) the Dr has some good news for you.  You can use NetApp deduplication to get back what Microsoft taketh away.  On tests run with Exchange 2010, without doing much optimization, we've seen seen space reductions of around 30% via dedupe.  Better yet, since NetApp dedupe is run in the background as a post-process procedure, performance impact on those users sharing their favorite files should be nominal.  I guess you could say that NetApp dedupe lets you go Back to the Future - It's free and it saves storage capacity - whats not to like?

DrDedupe

The Evolution of the Storage Brain - Applications Run Faster With Deduplication

It seems a contradiction - improving system performance with dedupe?  Usually, when someone thinks about implementing dedupe on a storage system, the first thought is "hmmmm...how much is this going to slow down my applications?"  In this blog I am going to tell you how dedupe can actually speed up your applications.

In Data ONTAP 7.3.1, NetApp introduced "Intelligent Caching."  Intelligent Caching refers to FAS and V-Series system memory, as well as secondary memory.  By secondary memory I am referring to the Performance Acceleration Module, or "PAM" for short.  NetApp introduced PAM II in this press release, although it was buried in with a bunch of other important announcements so it was easily missed.  I'll  come back to PAM II later.

So how does system memory, secondary memory, and Intelligent Caching relate to dedupe?  Simple.  When a NetApp storage system holds a dedupe'd data block in its memory, it also holds all the information about the data pointers that reference that particular block.  If this block has been deduped, say, 100 times, and is currently held in memory, there is no need to access the block from disk again, regardless of who is requesting the data block.  The result is faster access to data via reduced disk I/O latency.

The ramifications of this become very interesting.  The higher the dedupe ratio, the more dedupe'd blocks are placed into memory and the faster the system becomes.  We've already run benchmarks of this during VDI boot storms (when everyone turns on their virtual desktops at the same time and the storage system is inundated with read requests) and the results with dedupe and Intelligent Caching show orders of magnitude improvement in read response times.  With PAM II, the total system cache can exceed 2TB, and as Foxy Cleopatra might say, "Now that's a WHOLE lot of memory."  With this larger cache, the laws of probability begin to take affect - a larger cache means more blocks can be cached, and frequently requested blocks will remain in cache longer.  Meanwhile, since dedupe'd blocks have multple pointers they have a higher probability of being cached ...you get the idea.

To me, the combination of intelligent caching and deduplication is a signal that we are entering a new era in the maturity of deduplication, where the intelligence of deduplication is combined with other intelligences (intellligenci?) of the storage system to bring breakthrough results - the evolution of the storage brain.  I'll be exploring other thoughts on this topic in upcoming blogs...

Deceit, Theft, Espionage, and Murder in the Storage Industry - Part 4

It was a sun filled spring day at the Southern Hills Country Club in Tulsa, Oklahoma on Sunday May 27, 1981. Roger Wheeler had just loaded his golf clubs into the trunk of his car after a casual round of golf with friends, just as he had done a hundred times before. As he walked towards the driver’s door of his car, he probably didn’t pay much attention to the dark sedan that silently slid up behind him. Getting into his car, fumbling for his keys and reaching for the ignition, he likely did not notice the two men that left the sedan and were quickly approaching his window. The next morning, I was among the employees of Telex Computer Products to learn that our CEO Roger Wheeler had been murdered, execution style, with a point-blank shot to the head as he sat in his car with the engine running.

For decades, this murder remained unsolved. Over time, however, the events that lead up to this heinous act slowly unfolded and eventually caused the unraveling of one of the most notorious mobs of our time – Whitey Bulger and Boston’s Winter Hill Gang.

Roger Wheeler was aptly named - he was a free-wheeling Oklahoma businessman. Making his fortune as an entrepreneur, he became one of the wealthiest people in Tulsa during the 60’s and 70’s. One of his close friends, Steve Jatras (who succeeded Wheeler as Telex CEO) was a brilliant engineer who saw promise in the burgeoning data storage market and wanted to design a tape drive that could be plugged directly into IBM mainframe computers (see blog #3 of this series for more of Telex’s history.). Wheeler convinced Jatras to join Telex after he made a 17% stake in a company called Midwest Instruments, changed the name to Telex and was voted in as CEO. Jatras agreed to join.

Wheeler himself was not a technologist – when asked by a reporter what Telex did, Wheeler responded “We make widgets, and it’s my job to make sure we make money selling those widgets.” Wheeler continued to wheel and deal as Telex CEO, padding his fortune. One of his personal investments was a majority stake in a number a Jai-Alai Frontons in Connecticut and Florida. A sure money maker, he reckoned. Unfortunately, these establishments were mired in corruption and illegal activities, driven primarily by Bulger’s gang. The mob demand protection money; and in his typical wildcat style, Wheeler refused. Wheeler went so far as to testify in an effort to clean up the Jai-Alai gambling industry. A lofty goal indeed, and one that few men would ever attempt. But this was Roger Wheeler, a man among men.

Meanwhile, Bulger’s gang was becoming more brazen. Johnny Mortorano, aka “The Cook”, was ordered by Bulger to “take care” of the Wheeler problem. This would show loyalty and improve his rank in the gang. The Wheeler "job" was completed on that fateful Sunday in Tulsa. As has so often happened in the past, greed and envy became the eventual undoing of the Bulger gang. When John Callahan, fellow Winter Hill Gang member, exposed the details of the Wheeler slaying in 1982, his body was later found in the trunk of a rental car at the Miami airport.

Federal investigators were leaning on their other informants and beginning to put the pieces of this puzzle together. Once they had enough evidence, they moved in to capture Bulger and charge him as being the ringleader of the Wheeler execution, but Boston FBI agent John Connolly allegedly tipped off Bulger minutes before the FBI arrived and Whitey disappeared.

Four men were eventually indicted in the Roger Wheeler murder case – two served jailed time, one died before sentencing, and Whitey is still in parts unknown and on the FBI’s most wanted list to this day.

As I mentioned in the beginning of my 4-part blog series, if you stick around long enough in life you’ll see things you never dreamed of, things that you never forget that touch you directly or indirectly.

I hope my 4 stories helped make your summer a little more interesting. Now its time to get back to my ‘normal’ blog topics – we have a lot to talk about!

Deceit, Theft, Espionage, and Murder in the Storage Industry - Part 3

Part 3 - Espionage and Plug-In Storage

For this story, we go all the way back to 1972, when mainframes roamed the earth.  In those days, the entire computer industry consisted of 6 companies - IBM and the BUNCH (Burroughs, Univac, NCR, Control Data and Honeywell.)  These companies hovered over their customers like protective parents, and effectively blocked out their competitors with highly proprietary hardware and software.

Enter a little Tulsa-based company called Telex.  Telex emerged from the oil and gas seismic industry (they were previously called Midwestern Instruments) with a clever device used for geophysical logging called the M-3000.  The M-3000 was an analog magnetic recorder, or in today's parlance a tape drive.  Watching the emerging success of IBM's mainframe, the folks at Telex thought "hey that newfangled IBM computer uses tape drives, and we make tape drives!"  The result was the first plug-compatible peripheral in history.  Telex had designed an IBM interface and channel controller that would plug directly into any IBM 360.  And to the chagrin of IBM, the darn thing worked and Telex sold it much cheaper than the IBM model.

Not taking this lying down, IBM decided to try and persuade potential Telex buyers not to associate themselves with this upstart company, to the point where they would refuse to service the entire computer if anyone dared attach a foreign device to their precious mainframe.  Telex was in trouble.

Having little choice, Telex sued IBM on January 21, 1972.  Telex complained that IBM violated antitrust laws and used predatory practices designed to force Telex out of business.  The case trudged through the courts and December of 1973, a district court ruled in favor of Telex and awarded them $352 Million.  IBM quickly appealed the decision and launched a counter-suit, alleging that Telex engaged in industrial espionage.  How else, they reckoned, was Telex able to figure out how to connect their alien device to our mother ship?  IBM claimed that "Telex had engaged in unfair competition by inducing present and former IBM employees to breach their duty of loyalty by revealing trade secrets and other confidential information about IBM's business."  And sure enough, early in 1975, a higher court overturned the earlier decision in favor of Telex and ruled for IBM, fining Telex $18.5 Million.

Having had one of the largest antitrust cases in history overturned, Telex decided in July of 1975 to take its case to the United States Supreme Court.  Tensions mounted as the highest court in the land prepared to make its judgement on Telex's petition.  Both companies decided they had too much to lose if a judgement were found against them.  In the 11th hour, both companies agreed to a "wash settlement" where Telex agreed to remove its Supreme Court petition and IBM agreed to discharge its counterclaim.

So after nearly 4 years of legal wrangling, both parties dropped all complaints and no money exchanged hands.  IBM did, however, soften its policy of allowing 3rd party devices to be attached to its mainframe computers.  The infancy of open systems computing had begun...

Next:  Murder in the Storage Industry 

Deceit, Theft, Espionage, and Murder in the Storage Industry - Part 2

Part 2 - The Case of the Purloined Disk Drives

Back in the early 90's, a couple of pals and I started a company to provide data storage for the exploding Sun Microsystems workstation market.  Our business plan (yeah like we had one of those) was to quickly scoop up the latest and greatest technology, slap it into a desktop shoebox, and beat our competitors to the punch.  The data storage industry was undergoing its own space race, with vendors rolling out a steady supply of bigger, faster disk drives.  The flavor of this day was Maxtor's 525MB drive.  Maxtor had beat everyone else to the magic 500MB level, and these drives were hot commodities.  More hot, in fact, than I realized.

Part of my job at this company was procurement officer.  Translation:  I would buy the weekly edition of Computer Shopper at the local newstand and scour the ads looking for folks offering the latest drives.  Spotted an ad for the treasured Maxtor drive and placed a quick call.  "Sure we have those drives.  How many do you want?"  Placed an order for about 100 of them and as I hung up I thought to myself that it was odd that this seller offered the drives so cheap.  Attributing this to my shrewd negotiating skills, I went about my business.  I few days later, on a Friday as I recall, the drives showed up as promised and I began assembling them without too much further thought.

The following Monday morning was bright and sunny, a beautiful day in the Boston metro area.  I saw another car in the lot as I pulled in to work but didn't think too much of it.  I did notice, however that two men in suits followed me to the front door.  As I reached for my keys, one of them identified himself as an FBI agent and the other as a representative of Maxtor.  Had I recently purchased some 525MB disk drives from Mr X?  I gulped and said yes I had.  They said they wanted to talk to me and the story unfolded.

Turns out Mr X was part of a theft ring.  Employees of Sun had been pocketing these nifty drives and emptying their pockets in his shop.  Sun had noticed some accounting irregularities and began tracking serial numbers.  The FBI wanted to take my drives and see if the serial numbers matched.  Was I willing to cooperate?  Just then my partners arrived, and they looked at me like I was going to single-handedly bring the company down.  Wouldn't you know it, the phone rang at that moment and it was Mr X - asking if I got the drives OK.  "Tell him you want to buy more drives" the FBI agent whispered as I feigned calmness talking to Mr X.  My head was spinning.  Would I cooperate?  Damn right I would!

Things calmed down a bit as the FBI agent loaded the drives into his trunk and thanked me for my cooperation.  The fellow from Maxtor said they would replace all the drives with brand new ones and also thanked us for our help in sending Mr X up the river, apparently they had been trying to nab this guy for quite a while and now finally had the evidence they needed.

I never heard from Mr X, the FBI, or Maxtor again, except of course for the fresh shipment of factory drives that arrived as promised.  After some good-natured ribbing, my partners said that they would take care of procurement from now on.

Coming up...Espionage in the Storage Industry