Dedupe this: NetApp buying Data Domain
Surprise deal obscures good results
Comment Announcing its intent to be king of the dedupe hill, NetApp is buying Data Domain for $1.5bn. The announcement overshadowed NetApp's fiscal 2009 earnings which beat Wall Street estimates.
The IT world has been agog with rumours of a big takeover, Cisco and EMC being often mentioned, but Data Domain and NetApp was an unexpected pairing. Neither company is weak or in trouble and neither is known to be boldly acquisitive. Yet NetApp has struck a strong and decisive deal with Data Domain: a somewhat out-of-character deal given its track record of buying technology to be integrated into its existing offerings, with the exception of Decru.
The most recent Data Domain earnings  showed revenues up 50 per cent year-on-year and 7 percent sequentially. It made a profit too, showing it was capable of growth in the midst of the recession. It is certainly no financial basket case.
Data Domain is the energetic leader of the deduplication market, powering its inline deduping technology with Intel processors and intent on becoming a storage platform company in its own right, or so we thought. It has a strong line of products which it has sold vigorously, but with two perceived weaknesses. So far it has no answer to Ocarina's content-aware compression, the ability to dedupe graphic images and videos such as JPEG and MPEG files that traditional dedupe can't touch. Secondly Data Domain has yet to cluster its boxes together and so scale out.
Competitors such as Sepaton and Exagrid can cluster their deduping boxes and so offer greater scale than Data Domain. This is a capability that many expected Data Domain to deliver, after it has introduced Nehalem-based controllers.
NetApp is paying about $25/share, a 40 percent premium when the deal was announced yesterday. There is an $11.45 cash plus 0.75 NetApp shares for each Data Domain share. The deal is expected to close in 60 - 120 days, with NetApp gaining increased earnings in its fiscal 2010.
NetApp says it is buying because the two businesses are complementary, Data Domain's boxes being used to deduplicate files backed up to disk, and not competing with NetApp's FAS lines of online file and block storage. Data Domain will be able to grow its international sales faster by going through NetApp's European and Asia/Pacific distributors and resellers than by growing its own international channels.
There is overlap with NetApp's existing VTL (Virtual Tape Library) line but that will be sorted out. NetApp isn't saying anything specific about this or Data Domain's roadmap or staffing, other than to affirm broad commitment to the roadmap and to the principle of keeping Data Domain as a growing and profitable business run as a relatively independent operation. We might imagine that NetApp's own VTL product line developments are now on hold and that the line is facing practical end of life, with due arrangements for replacement products, support and other considerations yet to be made.
NetApp must have decided that it is cheaper to buy its way to dedupe product success than to continue to develop its own deduping VTL  product line.
Alacritus was bought by NetApp in April, 2005, for its virtual tape library technology, and resulted in the separate VTL product line, it not being integrated into ONTAP. We might take the view that NetApp is effectively replacing that technology four years later with Data Domain's.
Data Domain also gives it many new accounts into which it can sell its FAS hardware and software products, and the deduplication market has a long way to go before it reaches saturation.
Data Domain motivations
Why is Data Domain selling? Why have Frank Slootman and his team decided enough is enough? Surely it can't be the lure of faster international sales development? Surely it can't be a need for product development funding? They are sitting on a cash generation machine with Nehalem and clustering set to help continue a blistering growth path. The motivation here is not clear. Data Domain could buy Ocarina and deal with that technology gap problem at a stroke. The clustering technology must be in Data Domain's development labs already. What problem does Data Domain face that requires a NetApp acquisition as its solution?
Is it the case that Slootman and his team recognise that a stand-alone deduplication supplier cannot survive and prosper? Data Domain cannot transform itself into a general storage platform now that every storage array supplier has its own deduplication strategy in place, witness EMC with its Avamar acquisition and Quantum relationship.
This view would say NetApp is the haven Data Domain absolutely requires in the longer term because, as EMC has said over and over, deduplication is a feature, not a platform.
Looked at another way, this acquisition says as much about the relative weakness of NetApp's existing VTL product line as it does about the future for stand-alone deduplication storage array suppliers. Both are now probably finished, with the exception of niche, high-end, scale-out dedupe products such as those from Exagrid and Sepaton.
NetApp's previous experience of acquisitions could be described as patchy. Technology integration has proved problematic in the case of Spinnaker and clustering with it taking several years before clustering comes to maturity. NetApp's merged and cluster-capable ONTAP v8.0 release is expected later this year.
The Topio any-to-any replication technology acquisition  resulted in NetApp learning that there wasn't as much of a market for that functionality as people thought. Nevertheless, with Data Domain being acquired and run as a separate business unit within NetApp, these problems shouldn't occur. Unless another bidder emerges and with both boards in favour it looks as if the acquisition is a done deal and a great deal.
Simple put, NetApp is buying the hottest deduplication technology and product line in the market.
In other news, NetApp announced fourth quarter fiscal 2009 revenues of $880m, down from the year ago quarter's $938m, with net income of $75m, $15m less than the year before. Full year 2009 revenues were $3.4bn, 3 percent higher than a year ago, with net income of $87m, well down on 2008's $310m. The earnings were better than Wall Street's expectations, NetApp having controlled its costs well. No guidance was given for the next quarter. ®