In my previous post, I discussed why I thought that NetApp’s acquisition of Engenio was a difficult one and why I question the value of the combined entity.  Simply put, there seems to be redundancy in the product line and it makes you wonder how a merged company creates substantial new value.  However, there is another angle that could help explain the move and can be simply explained in three letters – IBM.

NetApp has an existing relationship with IBM where IBM OEMs NetApp filers for their “N-Series” product line.  These products are virtually the same as what you can buy from NetApp directly.  However, the big difference is that they can be sold, quoted and supported by IBM.  IBM does have their own higher end NAS products called SONAS, but the N-series is still an important part of the IBM portfolio.  Clearly this relationship is an important one for both companies.

Engenio is also an important part of IBM’s portfolio since IBM uses Engenio for the DS4000 and DS5000 mid-range disk products.  Like the N-Series, these products are created by Engenio although IBM takes a more active role in customizing these platforms.  Like NetApp, this relationship is an important for IBM since these solutions directly address the mid-market space.

There is an interesting similarity here.  We have two companies both of which have important relationships with IBM joining forces.  If you are IBM, you probably view this as a mixed blessing.  Clearly, post-merger, IBM can feel good about the improved consistency and longevity of two key suppliers, but Netapp/Engenio will also have greater negotiating leverage which could impact product costs and profitability.  At the same time, if IBM were considering an acquisition, the combination of NetApp and Engenio would be more interesting than either company alone.

From a NetApp perspective, the merger with Engenio strengthens their relationship with IBM and brings a new family of block-based storage systems.  However, I believe that it also enhances their value as an acquisition target.  It is obvious why IBM would want to acquire them, but third party Engenio OEMs could be interested as well.  These companies have already invested in Engenio technology and acquiring a combined NetApp and Engenio would instantly provide a credible storage portfolio with a broad range of products.

The combination of Engenio and NetApp may create technical challenges due to the need to support multiple controller technologies, but I believe that the combined entity enhances NetApp’s position as a possible M&A target.  IBM would be a natural player, but the other Engenio OEMs would be interested as well.

Related posts:

1. NetApp and Engenio – A curious acquisition
2. NetApp and Data Domain: ‘Til death or a better offer from EMC do we part
3. NetApp and EMC Duel to the Death for Data Domain

It was with some surprise that I saw the announcement that NetApp was acquiring Engenio this afternoon.  There was a long ensuing discussion on Twitter on this topic and I have serious doubts about the added value of the combined entity.  Here is why.

LSI’s primary go-to-market model has always been through OEMs and so they needed to have solutions that were feature rich enough to create a compelling value.  Obviously, their disk arrays have the basic table stakes such as replication and snapshots, but Engenio was trying to do more.  They wanted to create a modular array that could be extended with advanced software functionality to provide a unified storage platform.  They acquired StoreAge with a goal of bringing storage virtualization into their controller and later OnStor to embed NAS functionality.  Thus the end goal was clear – create an advanced RAID controller that could compete with the industry leaders and leverage their OEMs to go to market.  Interestingly, this model is more similar to NetApp’s strategy then you might think.

NetApp’s core strength is their ONTap OS which powers their storage devices and incorporates the ability to provide simultaneous block and file access and a variety of other advanced software features.  ONTap has been the key enabler of NetApp’s success and it is a technology that other storage vendors try to emulate.  The same core ONTap technology is used throughout NetApp’s product line which allows them to better leverage their development efforts.  From an architectural perspective, ONTap and its related applications (snapshot, replication, WORM, etc…) reside inside the NetApp head along with RAID functionality and the disk shelves in the system are simply JBODs.  NetApp also sells a gateway product called the V-Series which is a head running ONTap which sits in front of third party RAID storage.

Looking at Engenio and NetApp’s product lines, it is clear that there is quite a bit of overlap.  NetApp’s business is enabled by ONTap and the ability to provide efficient and powerful software features in their controllers.  Engenio was trying to do the same thing.  Why would
you want two different controller architectures to solve the same problem?  

If NetApp does keep the Engenio controller architecture they risk diluting engineering efforts.  Prior to Engenio, all engineering could be focused on ONTap and related software applications.  Post Engenio, (assuming they keep Engenio’s controller technology) they will have to maintain two entirely separate controller architectures and code-bases which creates management complexities and dramatically weakens the ONTap everywhere story.

One Twitter user suggested that they could use NetApp V-Series heads with Engenio storage.  That is absolutely true, but why would you?   A V-Series head brings all ONTap functionality (snapshots, replication, WORM, etc…) to any supported storage and by doing so you no longer need those features in the disk array.  Thus back-end storage for V-Series becomes a basic RAID-only solution.  In the case of Engenio, the V-Series would turn an Engenio system into a commodity RAID device with none of the advanced features that Engenio has acquired or developed.

One industry analyst suggests that the acquisition is good because of the profitability of Engenio’s OEM model.  Perhaps, but you have to wonder how the OEMs will react to this.  If I am an actual or potential competitor to NetApp will I still want to OEM Engenio?  It makes you question the longevity (and profitability) of the OEM business.

All of the issues above makes you wonder why NetApp bought Engenio.  Why do you think they did it?  Stay tuned for my hypothesis in part 2 of this post.

Related posts:

1. NetApp and Engenio – Part 2 – A Hypothesis
2. NetApp and Quantum: Why an acquisition would be difficult
3. NetApp Dedupe: The Worst of Inline and Post-process Deduplication

About four months ago, I published a blog post discussing the future of dedicated VMware backup solutions.  The post ignited a bigger discussion and included additional blog entries from W. Curtis Preston, me, Jon Toigo, Virtual Tacit, Veeam and Quest software.  I strongly encourage readers to review the differing perspectives in each post.

I mention all of the above to provide context to Quest’s acquisition of BakBone.  For those who are unfamiliar, BakBone is a backup software provider that plays in the low end of the market.  They position their product, NetVault, as a full service backup application that includes traditional backup, CDP and deduplication.  They have had limited success in the US and Japan appears to be their strongest market.  Their technology was spun-off from AT&T labs.  (As an aside, CommVault was also a spin-off from AT&T Labs.  Those lab guys must have been doing some amazing backup stuff!)

One of the main questions in the blog discussion was how the dedicated VMware backup providers, Veeam and Quest, could compete over the long term with the entrenched backup ISVs who have massive installed bases and larger sales and engineering resources.  Veeam’s position (as described in their post) was that they would stay focused on VM data protection and innovate rapidly thus maintaining a consistent leadership position.  Quest’s response suggested that they wanted to move beyond VMware-only data protection to backup physical servers as well.  The BakBone acquisition is a natural embodiment of this strategy.  However, the acquisition creates new challenges.

The two biggest challenges for Quest are competition and engineering.  The addition of BakBone elevates Quest from a niche player that protects only VMs to a full backup application provider.  Companies that once ignored them such as Symantec, IBM and others will take note.  If the incumbent ISVs see Quest taking marketshare with NetVault (e.g. becoming successful), you can bet that the incumbents will aggressively counterattack with take out programs and other incentives.   The engineering outlook is equally challenging.  Post acquisition, Quest must aggressively innovate in both VMware data protection and physical server protection. VMware innovation should not be a problem, but differentiating in traditional backup is much more difficult since the market is relatively mature and NetVault has had limited success.  The process of integrating vRanger and NetVault must also begin immediately which further adds to the burden.

In summary, it is clear why Quest acquired BakBone – they see a future of combined physical and virtual server protection.  However, you have to wonder whether Quest is spreading themselves too thin.  It is hard enough to do one thing really well (VMware protection) and now they have increased the challenge dramatically by adding NetVault and all of its related agents and features.  Additionally, the incumbent backup ISVs are likely to attack aggressively if threatened which could hinder success. However, there is a large revenue opportunity if (a big if, in my opinion) Quest can pull this off.  Only time will tell if they are successful.

No related posts.

SEPATON has been a great place to work over the last six plus years.  It has been amazing watching the company grow from a small group of about 30 employees in a tiny office space to today’s environment with over 150 employees and large office and lab space.  (Ironically, our current space that once felt so expansive is beginning to feel tight!)  At the same time, the company has grown from only a few small installations to thousands worldwide and hundreds of petabytes protected.  It is truly an amazing story and I have been fortunate to play a small part in it.

SEPATON’s future continues to look bright with lots of cool new features under development.  I would love to tell you about them, but, well, I can’t and you will have to wait and see! 

The big news is that I will be starting next week at HP here in Massachusetts and helping them market their D2D, VLS and physical tape products.  It is an exciting job with great future potential and the partnership between HP and SEPATON on the VLS means that I will continue to collaborate with the SEPATON team.

The great part is that my role will continue to revolve around data protection, and I anticipate continuing this blogging experiment that began over two years ago.  It is not clear whether I will be posting here or on HP.com, but rest assured, I will be pondering the same issues and providing my continued perspectives on all things data protection.

I wish nothing but the best for SEPATON and all the great people who work here.  It has been a fantastic six years.  I am looking forward to the new challenges at HP and helping further increase the success of HP’s data protection platforms.

Note: Regardless of whether I post here, you can always reach me via the contact form on this blog or on my personal blog (e.g. non-data protection) at www.livens.org.

No related posts.

I have been intrigued by the recent Dilbert comic strips highlighting social media.  The piece below was particularly amusing and got me thinking about social media and corporate strategy.  The comic shows the conflict  between social media adoption and corporate culture and portrays a situation that happens all too frequently.  Many companies desire to engage in social media, but when they recognize the commitment and ensuing open discussions back away.

In the storage industry, social media participation varies.  On one extreme, we have companies like EMC or NetApp who have embraced the medium with a large number of active Twitter participants and bloggers.  The broad range of content and opinions suggests that both companies give employees great leeway. (Although it appears that the amount of freedom varies by division.) However, there is a downside.  At times conversations can devolve into competitive fighting that benefits neither party.  These discussions are frustrating and I know of people who have stopped following argumentative Twitter participants from both companies.  These spats are a byproduct of openness, and would be avoided if the companies implemented stronger controls.  However, by allowing open discord, individual personalities can flourish thus creating a more engaging social media experience. 

While EMC and NetApp are highly active on Twitter, there are plenty of companies in the storage and data protection industry who have limited involvement. It is not clear whether the limited participation is due to a lack of recognition of the importance of the medium or perhaps a Dilbert-like scenario where corporate policy or legal requirements prevents participation.  Social media represents an important outlet to interact with end users, peers and colleagues and I believe that companies who are not participating are missing out on an opportunity.  I understand the fear of trying something new or being involved in a competitive fight, but social media enables new connections that are difficult, if not impossible, to recreate using other mediums.

The problems described above are typically found in large companies that are slow to adopt to new mediums and marketing outlets.  Smaller companies, like SEPATON, typically have an advantage because they are often more agile and willing to try new approaches.  In the case of my employer, SEPATON, I started this blog over two years ago and began tweeting soon thereafter.  It has been a great experience and the numerous interactions with other industry participants has been extremely valuable.  I can say with absolute certainty that social media including Twitter has facilitated many new relationships.

In summary, I believe that social media participation is very important.  It can increase corporate recognition and drive new interactions with current and future customers.  While some companies have an open approach which may not be right for everyone, the strategy of ignoring the outlet is not a recipe for success.  Companies must choose the strategy that works most effectively for them, and I am fortunate that SEPATON has embraced a model of openness.  Companies who are struggling with social media would be well-served to remember that their employees represent them every day in phone calls, events, meetings and webinars.  If you trust your employees to represent you in daily activities why not in social media?  I know that this is a struggle for many companies, but remember if you are not participating, your competitors likely are.

Related posts:

1. My experience with social media and the Super Bowl

Last week, I blogged about the 3Par bidding war and how I thought that HP would prevail.  Yesterday, Dell refused to match HP’s latest offer and so unless something crazy happens, HP is now the proud owner of 3Par for the rock bottom price of $2.4B!  The price is more than double Dell’s initial bid of $1.15B and is more than EMC paid for Data Domain.  In order to justify these high bids, Dell and HP must have thought that 3Par could create strong business value.  Now that HP has prevailed and is on the hook for $2.4B, they must execute the transaction and show how 3Par can drive incremental revenue and profits.  Let’s look at some ways HP could leverage 3Par to meet these goals.

Revenue growth is a key metric that will be used to assess the success (or lack thereof) of the 3Par acquisition.  In order to accelerate top line growth, HP cannot just replace EVA and XP sales with 3Par; they must find new avenues for the technology.  Some options include:

• Managed Service (Cloud) providers:  3Par developed unique technology to address this market and 7 of the top 10 infrastructure-as-a-service providers relied on their storage.  HP must continue to pursue managed service providers including HP’s own ES division (formerly EDS).  The credibility and relationships that HP brings as an computing infrastructure provider could enable them to accelerate market penetration and establish an integrated leadership position in this hot market.
• Net new customers: HP must leverage their existing brand and market presence to aggressively sell 3Par technology.  3Par was an early supporter of VMware’s storage APIs and HP could position the new technology as a VM optimized storage platform which would align nicely with their blade server agenda.   To be successful, they must find new opportunities where these message would resonate. 
• Improved competitive positioning: The new technology could allow HP to battle more effectively with competing storage providers.  In this scenario, HP does not necessarily need to create new opportunities, but be more competitive in existing ones.  For example, if today HP wins 30% of competitive situations then you would expect that to increase with 3par.  If their win rate increased to 60% then they would double their revenue.  However, this improvement will take time and will likely be less impactful than the first two.

Revenue growth is critical, and must be considered in the context of profitability.  Profits impact the bottom line and represent the difference between the revenues generated and associated costs.  Every sale (assuming that it is profitable) generates margin, and so net new revenue will increase margin dollars.  However, HP has an opportunity to improve margin percent or, alternatively, reduce costs.

As previously discussed, replacing EVA and XP sales with 3Par will have a limited revenue impact.  However, there is potential for margin improvement by replacing XP sales with 3Par.  XP technology is manufactured by Hitachi Ltd in Japan and HP purchases complete arrays which they re-brand and deliver to end users.  Thus, HP’s cost for the XP is higher than it would be for a wholly owned technology like EVA and now 3Par.  The result is reduced profitability on XP sales.  If HP could replace XP sales with 3Par technology with no revenue change, then you would expect margins to increase which would translate into bottom-line profit improvement.  Thus, HP has a near term incentive to convert XP business into 3Par business and aggressively promote 3Par in lieu of XP.

In summary HP has made a substantial statement about the potential for 3Par by bidding over $2B for the company.  They clearly believe that they can drive increased revenues and profits by integrating the technology into their storage portfolio.  This post highlighted some areas where HP could recognize immediate benefits; however, it is yet to be seen whether HP can realize these gains.  Their success depends on the ability to effectively integrate 3Par’s technology and people into its storage group.

Related posts:

1. Why HP will prevail over Dell in the 3Par bidding war

The Twittersphere and storage industry are abuzz with the ongoing bidding for 3Par.  HP and Dell are aggressively pursuing the company and have a vested interest in 3Par technology. I believe that HP is more motivated to acquire 3Par and will prevail.

Both Dell and HP believe that a 3Par acquisition will generate additional business value, and they both must realize that the losing party will be placed in a difficult situation.  I believe that HP has more to gain by acquiring 3Par and more to lose by failing to do so.  Here is my assessment of the gains and losses by each bidder:

HP gain if they win:

HP has a broad storage portfolio and 3Par’s products would replace/augment the EVA and XP product lines.  EVA was developed by HP, and many storage people suggest that it is desperately in need of a major upgrade.  The XP is an array from Hitachi Data Systems in Japan and is resold by HP.  Thus HP is an interesting position of selling a relatively old, high margin product, the EVA and a newer low margin product, the XP.

The 3Par situation presents an opportunity for HP to alter the situation.  The acquisition would provide cutting-edge technology which could help address EVA’s challenges and provide much improved margins versus the XP.  Of course, HP may continue to sell EVA’s and XP’s, but a 3Par solution could help jump start enterprise array sales and take market share from not only the competition but also from EVA on the low end and XP on the high end.  Long-term, it would not surprise me to see 3Par technology replace the XP for all but the most complex environments and replace high end EVA.

In summary, HP has the opportunity to redefine their enterprise disk array solutions with an acquisition.  This is critical given the needs of re-invigorating the EVA and increasing margins on the high end.

HP loss if they lose:

I can only imagine the competitive FUD (fear, uncertainty and doubt) coming from other disk vendors if they lose.  By bidding so aggressively, HP is stating that they really want 3Par and implicitly indicating that they need to replace their current solutions.  I am sure that HP will argue that the bidding is not a referendum on their currently technology, but competitors will attack this point mercilessly which could impact storage sales.

If HP does want to replace their storage technology then 3Par is critical because there are few alternatives.  The remaining storage players often have great technology but lack the scalability, reliability and performance of the 3Par offering.  In this scenario, HP would need to aggressively pursue a new acquisition candidate to address the intense FUD that would be flung their way.

Dell gain if they win:

Dell has a partnership with EMC on the high end to sell CLARiiON (CX) disk arrays.  It is not clear how many of these arrays Dell sells, but I believe that it is a relatively small number.  Selling a CX system is much more complex and detailed than Dell’s traditional fulfillment model.  The 3Par acquisition represents an opportunity for them to replace the CX product line with a wholly-owned product and to add a sales team experienced in selling storage into large accounts.  Thus, they would better a address a market segment where they currently have a limited footprint which would result in incremental revenue.

Dell loss if they lose:

A failed acquisition will result in status quo for Dell.  They will still be able to attack the enterprise space with EMC’s CX product line, but will lack the technology and enterprise sales talent that 3Par would bring.  EMC has been highly successful selling CX technology and so I believe that Dell could effectively sell against HP/3Par on a technology basis.  The real challenge is the loss of the enterprise sales talent and customer base that 3Par would have brought.  However, I believe Dell could build a larger and more aggressive enterprise storage sales team focused on CX technology which would enable them to compete aggressively with HP/3Par and generate incremental revenue.

From a FUD perspective, Dell is at less risk since they do not own a technology that competes with 3Par.  If competitors tried to attack the failed acquisition, Dell could simply (and reasonably, in my opinion) respond that they saw an opportunity to own enterprise storage technology at a fair price and so pursued 3Par.  They could position the loss of 3Par as a situation where the bidding got out of hand so they decided to stick with their current provider.

Conclusion:

In my opinion, HP has more to gain and lose in this bidding war than Dell.  The addition of 3Par will enable them to instantly improve their storage offerings while minimizing potential FUD from the competition.  Dell, in contrast, could gain greater access to a market segment that they already address through their EMC partnership.  Dell certainly would miss out on sales talent and a customer base, but I believe that this is something that they could address using other methods.

Related posts:

1. HP acquires 3Par: Now what?

One of the biggest challenges with data protection is managing growth.  Some of the common factors that drive increasing capacity requirements include:

• Intrinsic growth – Growth inherent in the environment as users create new data.
• New applications – Companies implement new applications to meet changing business requirements.  These solutions could replace existing technologies or could be net new additions.  Either way, they often generate more data to protect and retain.
• New data types – In today’s multimedia-centric world, there has been a dramatic increase in the number of audio, video and image files being created and protected. These files are much larger and more difficult to compress than traditional content.
• Merger & Acquisition – As M&A activities occur, the acquiring entity must expand their IT infrastructure to absorb the acquired systems and processes.

The combination of these elements drives data growth and creates data backup, recovery and retention challenges.  Growth is constant and the image below (click for a larger view)  is a reminder of how data storage and protection has changed. It also brings the question: Do you want one system that can grow with your environment or many small ones that won’t?  If the latter, then how about I sell you 20 3380′s to replace your 50TB disk array?

Image source: TheFutureBuzz.com, via @sfoskett

No related posts.

Click for larger view – Comic courtesy of xkcd.com

Via Beth Pariseau from TechTarget.

Related posts:

1. W. Curtis Preston Now with TechTarget

Currently the twittersphere and blogosphere is actively discussing Fibre Channel over Ethernet (FCoE).  The conversation was triggered by a post by Hu Yoshida from HDS, and I wanted to share my thoughts.

One of the most interesting responses was this one by Nigel Poulton where he explains the infrastructure required for FCoE.  He goes into great detail highlighting the lossless nature of FCoE and the required hardware and cabling.  The key takeaway is that FCoE is not iSCSI; it does not use generic 10 Gigabit Ethernet (10 GigE) hardware.  You will need different cabling and advanced switches to meet the more stringent demands of FCoE.

The requirement for specialized FCoE hardware will increase costs versus iSCSI.  iSCSI can leverage existing LAN  investments and benefits from the economies of scale of Ethernet hardware.  The specialized nature of FCoE means that it will ship in lower volumes than vanilla 10 GigE and thus it will take longer for FCoE hardware prices to decline.  This will impact the ROI of FCoE solutions since the higher cost will extend the payback period.  Thus, FCoE will be at a price disadvantage to iSCSI; however, I believe that the two technologies address different markets.  FCoE mimics the reliability and consistency of Fibre Channel, but will also mimic the high price.  iSCSI in contrast, runs on mainstream Ethernet infrastructures and will provide a more cost effective although less reliable and slower performing option.

A real world experience would be useful in clarifying my thoughts.  I sold iSCSI solutions (EqualLogic, if you must ask) to small businesses in a previous job.  The prospects were companies who wanted the benefit of a shared storage environment and had limited budgets and sophistication.  They were excited about iSCSI because it provided the benefit of a SAN without the cost and complexity of Fibre Channel.  In my opinion, these customers will continue to favor iSCSI over FCoE.  However, I believe that the target FCoE customer is a larger environment that already has an FC SAN in place.

It will be interesting to see where the market moves from a backup perspective.  iSCSI is less common in today’s Gigabit Ethernet data protection environments.  As 10 GigE prices decrease, high performance iSCSI will become widely available.  Will this drive penetration for iSCSI backup targets?  It definitely could for smaller companies like the one described above.  However, I believe that enterprises have a strong need for the performance and reliability of FC/FCoE and will favor these technologies over iSCSI in the datacenter.

Another question is whether companies will favor FCoE over FC.  Given the newness of FCoE, I believe that it will take a few years before companies start seriously implementing FCoE in the core especially since most enterprises already have large investments in FC.  FC speeds will also continue to increase and keep the technology competitive.  Long-term, FCoE will co-exist in corporate datacenter infrastructures along with FC, and it will be interesting to see how companies divide their I/O workload between the two.

Related posts:

1. TS7650G and Fibre Channel Drives