GigaOM » Gary Orenstein Archives

Watch out tech world. This CEO knows consumers

Gary Orenstein — Fri, 23 Sep 2011 22:00:04 +0000

Say what you will about Meg Whitman as the new CEO of HP. She knows consumers.

Just a couple of days ago, HP was led by a lifelong enterprise software veteran madly dashing to refashion a global technology behemoth into something he was familiar with. During his tenure the well-regarded, though not as well marketed WebOS has been shunted aside. The PC business is now available for takers. It has spent one quarter of its market cap on a business with a billion dollars in revenue – a mere fraction of HP’s annual sales of $125 billion.

It’s amazing what can happen to a company in a year.

Thankfully the board did get the message quickly, and now Whitman, a former board member since January 2011, is in the drivers’ seat replacing Leo Apotheker

Whitman’s background and leadership style are likely to be on the opposite side of the spectrum from Apotheker’s. She knows consumers, inside and out. Her career path reads like a who’s who of consumer brands, from Procter & Gamble to Disney, Stride Rite to FTD, Hasbro and then eBay, a consumer sensation in the early days of web commerce.

Though only an occasionally eBay user myself, it is clear she helped galvanize a community, pursued a cause, and enabled economic wealth for customers and the company itself. In the heyday of the early web, eBay was near, if not at, the top.

Now at HP she’ll have a chance to apply her consumer touch to one of Silicon Valley’s most storied brands. A brand that many thought had a chance to give Apple a run for its money. A brand that many have relied on for personal technology at home and even more at work. And most importantly, a brand that for years has held one of the most innocuous and underexploited positions in the corporate data center.

Watching the enterprise landscape, I’ve often lamented that HP has one of the most undervalued brands in the data center. It does so much, yet seems to lack the awareness of players like IBM, Cisco, Oracle, VMware, and EMC in their efforts to capture the minds of corporate buyers. It will be interesting if a master marketer will help turn this around.

There is such a great portfolio at HP, so much technology, and so much opportunity. I hope Whitman has the ability to recognize both ends of the spectrum. On the consumer side, we do need someone to give Apple a run for their money. And if it is not HP, I don’t know any other company capable of getting close.

But even more exciting to me is the opportunity to expand HP’s core enterprise business, both hardware, software, services, and now cloud computing. HP servers are well-recognized as tops in the industry, it has a broad networking range, a full line of storage systems, more and more software, and an emerging cloud offering all coupled with the scale to support it. Put this all together, mix in some web and mobile goodness, make sure folks are aware of it and understand it, and who knows what could happen.

Congratulations Meg. And congratulations HP. I hope we are about to see your true capabilities.

Gary Orenstein is the host of The Cloud Computing Show

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

VMware’s slow and steady attack on storage

Gary Orenstein — Sat, 27 Aug 2011 16:00:21 +0000

VMware has transformed the enterprise computing model more than any other company in the past decade. By allowing companies to add multiple virtual servers onto a single physical server, the company kicked off the first building block behind the adoption of private and public cloud architectures — the abstraction of applications and operating systems from specific dedicated hardware. Now VMware is turning its disruption model toward storage.

With the release of vSphere 5 last month and a number of new storage-specific products, VMware is encroaching on storage. Many of these features are outlined in detail on the vSphere blog, currently up to an 11-part series on vSphere 5.0 storage features! While there are a few capabilities here, like the Virtual Storage Appliance (VSA), that make it easy to see the future, it is the breadth of the feature set and the reach of VMware’s tentacles into storage that add up to potentially significant moves. Let’s take a look at a few of the offerings.

The Virtual Storage Appliance is a new software product from VMware that allows customers to create storage area networks right on their existing servers. Before VSA, customers interested in taking full advantage of VMware’s capabilities like vMotion had to pony up for an expensive external storage area network from the likes of EMC or NetApp at the high end or Compellent and EqualLogic in the mid-market. But even the lower prices of Dell’s newly acquired storage products can be too much for a small company to swallow. And the skill set to manage external storage is not always at hand. Enter the Virtual Storage Appliance, which essentially carves out local storage on multiple servers and presents it as a shared resource to all the virtual machines.

It is somewhat ironic that the virtual storage appliance offers customers an option to avoid buying an external storage array, given that VMware is still majority-owned by EMC. And perhaps that is the reason why VMware has focused this offering at small to medium businesses, minimizing overlap with larger customers that tend to need EMC’s standard external storage wares. VMware states on its website that it intends to eliminate the “specialized knowledge in shared storage hardware” that was previously required for virtualization. But if history is to repeat itself in the data center, as it often does, then we can expect the virtual storage appliance to get better and better over time. We can also expect it to offer a wider customer base the option to just buy servers plus VMware licenses and be finished.

VMware also introduced new vSphere Storage APIs for Array Integration (VAAI). Essentially, these APIs allow more storage features to be operated and initiated directly from VMware management tools, somewhat obviating the need to spend more time managing your underlying storage arrays. New Profile Driven Storage allows administrators to assign specific storage requirements to virtual machines that will automatically be matched with the appropriate underlying storage resource. Along the same lines, Storage DRS enables the intelligent placement of virtual machines across data stores within a cluster. Here VMware is assuming the task of assessing which data stores in a group have the best performance and latency characteristics and automatically moving virtual machines between them.

Remember how adeptly VMware wedged its way in between the CPUs and operating systems? IT architects now pick a hypervisor first before almost anything else. And that choice is often VMware. Watch the wedge in the storage market now. Each feature may only be one step toward a greater storage portfolio, but together they represent a clear path. VMware is no longer satisfied to sit back and let storage vendors decide how best to optimize for virtual machines. It is taking its destiny into its own hands and reaching far and wide to grab more of the data center storage stack . . . slowly and steadily.

Be sure to catch the very latest VMware announcements at VMworld next week, August 29 to September 1, in Las Vegas.

Gary Orenstein is the host of The Cloud Computing Show.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

VMware Is The New Microsoft, Just Without an OS

Gary Orenstein — Sat, 14 May 2011 19:00:45 +0000

In the past 10 years VMware has executed a remarkable strategy to topple enterprise software incumbents and emerge as an ecosystem kingpin. More recently, the company has plunged head first into cloud computing from infrastructure to applications. Time and again, it seems as though VMware is beating Microsoft at its own game. But a look deeper reveals that is no surprise.

VMware began with a concept to run multiple operating systems independently on a single physical machine through the use of virtual machines and hypervisors. Simple enough. But the implications are powerful as critical decisions once made at the operating system level quickly became less relevant. Companies such as Microsoft no longer need to be factored in up front for new data center architectures. This is how VMware put the first chink in Microsoft’s armor; operating system choices now take a back seat to hypervisor strategies.

VMware’s promise was visible early enough to EMC that it acquired the company in 2003. But it took until July 2008 for it to inject the company with leadership from Paul Maritz, a 14-year Microsoft veteran who joined EMC in February 2008. That is the same time EMC acquired Pi, a cloud computing company Maritz founded.

Martiz’s resume covers all bases of software infrastructure: platforms, operating systems, development tools, database products, productivity suites, and email. He also brought in two VMware co-presidents from Microsoft giving the executive suite a combined 47 years of Microsoft experience. So the VMware roadmap since then became easy to read. Start with a base infrastructure that can unseat the operating system, woo developers to create a robust ecosystem, and deliver value added applications up the software stack that reach directly to end users.

Pursuing the influential developer community, in 2009 VMware acquired SpringSource in a move to bolster their strength around the Spring Java development framework, and what was noted as a move towards an integrated platform-as-a-service offerings.

Building a collaboration portfolio, in 2010, VMware acquired Zimbra setting off another chapter in the cloud collaboration wars.

Most recently VMware launched Cloud Foundry, a platform-as-a-service that supports development in Java, Ruby on Rails, Node.js and other frameworks, further cementing its role as a market maker for new infrastructure approaches.

Along the way, VMware built a formidable enterprise ecosystem around virtual machine management and the fact that customers want VMware to tell them which other products they should choose to make virtual machine deployment easier. Think Windows Compatibility List 2.0.

With VMware holding the keys to the hypervisor layer and management, then the platform layer, and even the cloud applications layer with email from Zimbra, and presentations from Slide Rocket, why do I need Windows?

This story is far from finished. VMware has less than twenty percent of Microsoft’s market cap today. But if I were tracking the growth, and more importantly, enterprise influence, VMware appears to be making the right moves.

Want to learn more about the big moves in enterprise and cloud infrastructure? Check out the GigaOM Structure Conference June 22 and 23 in San Francisco.

Gary Orenstein is the host of The Cloud Computing Show.

Related research and analysis from GigaOM Pro:
Subscriber content. Sign up for a free trial.

The Dawn of the Super Server

Gary Orenstein — Sat, 05 Mar 2011 17:00:51 +0000

We’re in the midst of a computing implosion: a re-centralization of resources driven by virtualization, many-core CPUs, GPU computing, flash memory, and high-speed networking. Some have predicted, only half-jokingly, that we will be able to buy a mainframe in a pizza box server that fits in a small fraction of a data center rack. That possibility — and in my opinion, inevitability — means we have a lot to watch over the next few years: what I like to call the coming of the Super Server.

The business drivers for the Super Server span power, management, new workloads, and big data needs. Let’s examine each briefly.

Power

Rising data center power bills, combined with a macro push to environmental friendliness, has led to a slew of power-optimized servers. Today, the three-year power bill for data center equipment can often equal or exceed the original capital cost. And with cloud architectures spawning mega data centers costing hundreds of millions of dollars, there’s plenty of room to reshape servers for power savings. We’ve already begun to see the impact with the announcements of Windows running on ARM processors, and emerging server vendors such as Calxeda and Sea Micro focusing on lower power chips that still deliver data center performance.

Management

In addition to power, space, and cooling costs, operating expenses are the other major data center equipment cost post-purchase. Data center administrators usually look to minimize the servers count, or server image count, requiring oversight. Today, through virtualization, architects can minimize the number of physical machines they manage while keeping the same number of server instances available through virtualization. Since virtualization tends to be memory-hungry and storage-hungry, placing more CPU, memory, and storage resources within a single server allow that physical server to manage more virtual machines. Administrators can tackle the same, or greater, number of applications and workloads with less physical equipment: a management and administrative win.

New Workloads and Applications

Our computing habits continue to evolve with Internet development, and new web businesses spur the need for supporting infrastructure. As an example, those running cloud data centers don’t care at all about having CD drives or extra USB ports on their servers, but they do need ways to handle fast and furious updates, millions of video downloads, or voluminous click-tracking. Web application areas like social networking, online video, advertising, and mobile applications require server architectures optimized for transactions, capacity, and web serving, all while minimizing power and management costs.

Big Data Needs

Our Internet-enabled information age has put us in a race to capture, process, and distill more data than ever. When Hadoop emerged as a dominant ,open-source implementation of Map Reduce, it forced a rethinking of storage infrastructure. Previously, many applications requiring large amounts of data made use of centralized storage in large arrays connected together with storage protocols. With the Hadoop Distributed File System, data was intended to be close to the CPU, and on disks within individual servers. So we’ve actually seen a move to pull storage out of the centralized array and back into the server. This triggered a return of larger servers with more internal drive slots to accommodate the storage capacity for Map Reduce operations, a key element of the Super Server.

Want to learn more about big data and the impact on infrastructure? Be sure to check out Structure Big Data March 23, 2011 in New York City.

Gary Orenstein is host of The Cloud Computing Show.

Related content from GigaOM Pro (subscription req’d):

EMC’s Low-End Launch Makes Dell Look Prescient

Gary Orenstein — Sat, 22 Jan 2011 17:00:03 +0000

This week brought the full fanfare of EMC and its “Record Breaking” launch, which some say sounded more like a broken record. Shoving 26 people into a Mini Cooper and motorcycle jumping aside, the main meat of the launch is EMC’s new low-end VNX line, which it claims packs all the features any business storage user might need at an irresistible starting price of under $10,000. In one splashy move, EMC shifted focus away from its high-end Symmetrix products and down to the other end of the spectrum. And all of a sudden Dell’s move to acquire Compellent looks much smarter than it did just over a month ago. (Disclosure: I was the vice president of marketing at Compellent from 2002 to 2005.)

Dell’s acquisition of Compellent in December ended an 18-month flurry of storage M&A that challenged even the most optimistic market projections, and made the sector red hot. In particular, the bidding war for 3PAR that HP won over Dell reached an astronomical $2.4 billion, leaving some wondering how Dell would stack up on the high end of the market. 3PAR essentially represented the only independent high-end storage vendor, and without 3PAR’s scalable disk arrays, Dell would be left standing empty handed against HP, IBM, and EMC. At the time, I wrote that HP’s Bid Makes Sense and Dell could never realize the value that HP potentially could.

This week, however, we find EMC with guns blazing around VNX, a “Simple, Efficient, Powerful, and Affordable” offering countering the stereotypical knocks on EMC products as being complex and expensive. So, back to Dell: Would it have been better off in trying to fight EMC on EMC’s storied high-end enterprise territory with 3PAR’s products, or challenging EMC from below in a channel cage match. Dell knows the channel well, and now has an offering — Compellent — born and raised there for small to medium enterprises.

When Dell bought Compellent without even a bidding war, the deal didn’t have the excitement of the 3PAR acquisition. It also didn’t stand out as much from Dell’s existing acquisition of EqualLogic, the mid-market leader of storage systems based on the IP and Ethernet-based iSCSI storage protocol. With Compellent, though, Dell expanded to solutions including Fibre Channel, a high-end storage protocol, and a sophisticated software suite tailor-made for the storage mid-market.

Now it seems that the Dell-EMC relationship will come to an end. You can still find Dell-EMC storage on the Dell website, but with EMC’s charge straight to the low end, I cannot imagine Dell driving sales of EMC product long into the future. It is hard to imagine that they can co-exist as partners while both aiming company-branded, low cost solutions to the same customers.

The only thought that makes me pause is whether EMC is just throwing a big head fake to the industry. Perhaps the high-end segment is still all that matters, and given that this “Record-Breaking” launch was largely marketing-driven, perhaps it is just a distraction. Maybe EMC just wants to divert attention away from the billions being minted at the top of their line, and this VNX is nothing but an apparition. I doubt that is the case, but it would be interesting.

The likely scenario is a continued battle between giants for the fast growing small-to-medium enterprise segment. This is an area where Dell has excelled, and, in fact, EMC leveraged that capability by selling solutions through Dell. Now it seems like the gloves are off, and Dell looks even smarter for stocking up on ammunition with Compellent in advance of EMC’s charge.

Gary Orenstein is the Host of The Cloud Computing Show.

Image courtesy of Wikipedia Commons contributor che.

Related Content from GigaOM Pro (subscription required)

What Comes After the Storage M&A Gold Rush

Gary Orenstein — Fri, 24 Dec 2010 15:00:55 +0000

Between July 2009 and December 2010, HP, EMC, IBM, and Dell spent almost $10 billion on data storage and warehousing companies including Data Domain, 3Par, Greenplum, Netezza, Isilon, and Compellent. This all-out gold rush was driven by a massive consolidation wave sparked by Cisco’s entry into the server market and Oracle’s acquisition of Sun, making storage the hottest liquidity sector of the last 18 months.

Now that the technology giants have rounded out their storage portfolios, we turn our attention to the next areas ripe for potential acquisitions. These five sectors look particularly enticing for 2011:

Networking

Unfortunately, servers can’t magically glue themselves together in large clustered configurations for public or private cloud computing, but rather, rely on a complex array of networking gear to manage the traffic flow. This equipment may be the last stand against commoditization in the data center, and companies such as Arista, Nicira, and Embrane are pushing both the hardware and software barriers to more efficient solutions. Perhaps Cisco will release the M&A engine it has withheld from the storage market more aggressively on its home networking turf.

Cloud Platforms

Though HP, IBM, Dell, and Oracle have all positioned themselves as purveyors of cloud infrastructure, none of them have delivered a credible answer to Amazon’s EC2 Infrastructure-as-a-Service platform. VMware, OpenStack (sponsored by many, but primarily Rackspace), and Microsoft’s Azure provide similar capabilities, but it seems inevitable that the leading server vendors will want more control over how their products are consumed. So that leaves the smaller players ripe for the picking. Leading contenders include Cloud.com, Eucalyptus, Nimbula, and Joyent among others.

Big Data

We covered the commercialization of big data earlier this year, and since then, Cloudera and 10gen (behind MongoDB) have closed additional financing, and Riptano (behind Cassandra), announced initial financing. IBM and EMC recently bulked up on the last wave of big data and analytics with Netezza and Greenplum, but it’s only a matter of time until the uptake on this new class of big data solutions attracts enough momentum to merit a permanent home in a larger mega corporation.

Virtual Desktop Infrastructure

VDI has taken a crown, if not “the” crown, as an actionable infrastructure project atop virtualization. A combination of new infrastructure capabilities, proliferation of operating systems and devices, and incessant search for reduced costs has led to a VDI resurgence. We looked more closely at why Virtual Desktops Are Hot Again and see this area as a storage-centric deployment that will catch the attention of the tech giants once they digest their most recent acquisitions.

Cloud Gateways

In the grand scheme of storage, the cloud is the smaller but faster growing deployment method compared to on-site. The rapid rise in the number of objects on Amazon S3 to well over 100 billion is just one indicator of this. But conventional enterprises can’t move quickly to the cloud without a much friendlier environment to existing storage access mechanisms. Specifically, enterprises are used to block and file access to storage, not APIs. To fill this gap, a wave of companies promising gateways to cloud storage have emerged. Once again, with significant traction, several are likely to find homes either on the physical infrastructure or service provider side of the equation.

All in all, it promises to be an interesting year in 2011. Did we miss any must-watch sectors? Be sure to leave us your comments. And if you are interested in tracking big data more closely, be sure to check out the Structure: Big Data event coming March 23, 2011, in New York City.

Gary Orenstein is host of The Cloud Computing Show.

Related content from GigaOM Pro (sub req’d):

Why Is Storage So Hot? Cisco Provides a Clue.

Gary Orenstein — Fri, 26 Nov 2010 19:00:17 +0000

In March of 2009, Cisco launched its Unified Computing System, or UCS for short. It signaled to other data center equipment vendors that the separation between servers, networking, and storage was finished. With that single move, the industry set forth on a course to consolidate to integrated solutions, a trend we are in the middle of today.

Major vendors, sensing the integration trend, set off on an acquisition spree that sparked several multi-bullion dollar storage deals. This may be because the server roadmap is well understood, and highly commoditized. We are likely to get more cores, and more virtual machines in servers for a long time to come.

With storage, it’s not as clear. The impact of virtualization on storage is only beginning to rear its head. Deployments of large-scale, big-data solutions are still formative, and cloud storage options abound. Until they lock their blueprints firmly into place, the acquisition hunting might well continue. Remember that storage is the gift that keeps on giving. The data never stops.

Here’s a play-by-play since Cisco’s announcement:

March 2009 | Cisco announces UCS
Cisco enters the server market essentially declaring war with server giants HP, IBM, and Dell.

April 2009 | Oracle announces Sun acquisition
After IBM’s initial dialog with Sun, Oracle enters the hardware market.

April 2009 | Dave Donatelli, long time senior EMC executive joins HP
To counter Cisco’s server entry, HP names a storage heavyweight from EMC to cover server, networking, and storage products, a clear signal they have their own plan for converged infrastructure. However, Donatelli is barred from managing the storage business for one year following a lawsuit from EMC, forcing HP to major storage moves on hold, except for a small acquisition of scale-out file system vendor IBRIX in August 2009.

May 2009 | NetApp tries to buy Data Domain
The two dominant storage companies, NetApp and EMC, sense that the ground is starting to shake. They decide to defensively maintain positions in the storage segment. At the time, I applauded NetApp’s bid because it appeared they were ready to expand software offerings beyond a one-flavor-fits all approach of their core operating system ONTAP.

July 2009 | EMC acquires Data Domain for $2.1 billion
EMC wins control of the de-duplication market leader to bolster and defend its dominant storage position. This takes the wind out of NetApp’s sails and ensures that EMC has a tighter lock on hot storage technologies such as de-duplication which DataDomain first developed and popularized.

November 2009 | HP buys 3Com
While Donatelli is waiting out the storage hiatus, he can keep busy adding to the networking portfolio. HP already had the Procurve line of equipment but more will be required to compete with Cisco.

January 2010 | Oracle/Sun acquisition approved
The game is on. Oracle now has comprehensive server and storage product lines plus all of Sun’s software portfolio.

August 2010 | Dell tries to buy 3Par for $1.15 billion
3Par built a storage system capable of competing with EMC’s Symmetrix business. This is no small feat, which I described as The Simple Reason Dell Bought 3Par. At the time, neither HP nor Dell had a high-end product under their own brand to compete with EMC or NetApp. Dave Donatelli at HP is free of his storage exile. A bidding ware is about to break out.

September 2010 | Mark Hurd is in at Oracle
Oracle mentions it needs expertise from executives who know how to manage a $100 billion dollar business, no doubt a shot across the bow to HP and IBM. There are also rumors that Oracle might buy EMC, which could net the company a significant storage revenue stream and lock up the high end of the storage market.

September 2010 | HP gets 3Par for $2.4 billion
At the time, I was convinced HP needed to win this deal because they have an enterprise sales force capable of selling high-end storage products (rebranded from Hitachi Data Systems), and revenue growth could come sooner than Dell, which lacks the same organizational reach and depth in the high-end storage market.

September 2010 | IBM acquires Netezza for $1.7 billion
Netezza emerged as a strong company in data warehousing, a core element of the data mining renaissance. IBM, which through acquisitions such as Cognos in business intelligence and SPSS in statistical software, knows how to monetize analytics for big business.

November 2010| EMC acquires Isilon for $2.25 billion
Similar to 3Par, Isilon took a long but ultimately successful road in building a product that competes head on with certain portions of EMC rival NetApp’s file storage business. Isilon specializes in scale-out storage, where companies like HP, NetApp, and EMC have made multiple bets.

Did Cisco have any idea what kind of fallout it would trigger with its move? My guess is it went through plenty of scenarios ahead of time. It has the cash and capability to play the M&A game with HP, IBM, and Oracle should it need to. So we may not have seen the end just yet.

Gary Orenstein is the host of The Cloud Computing Show

What’s Behind The Build-Your-Own Data Center Trend

Gary Orenstein — Sat, 13 Nov 2010 17:00:25 +0000

Facebook's planned North Carolina data center.

In the wars of the web, hard goods have become the ammunition of choice. Specifically we see major web and technology companies building their own data centers, a move that appears to be a prerequisite to competing with Internet giants.

Since the early days of Internet hosting, and now with the prevalence of cloud computing, there have always been relatively simple ways to rent data center space, and even to rent servers to power web properties. So why the recent trend to own that infrastructure directly? There are several factors I believe driving the moves of companies like Facebook, Apple, and others in their data center land grab.

Removing “That’s Not Possible”
Without owning the data center and the equipment, web properties must rely on third parties, service agreements, and untold restrictions in the build out of their infrastructure and web site capabilities. For engineers that know what is physically and logically possible with hardware and software, restrictions on server configurations, or power consumption, or available space can be a hard pill to swallow. So instead of complaining to the boss that the job cannot be accomplished within a set of constraints, companies ultimately decide to remove them. Facebook said it will be able to try out more energy efficient options in its own center, while Twitter said it will be better able to optimize its site performance by controlling everything.

Keeping Companies Close to Their Data
As evidenced by the popularity in corporate enterprises for private clouds, it is hard to separate companies from their data. Whether because of perceived, or actual security concerns, companies like to own and control their infrastructure with a hands-on approach. And for companies that live by the operational up time and revenue from web operations, having the hardware close at hand, and directly under company control can make all the difference.

The Wall Street Effect
While web properties should be judged by user numbers, engagement statistics, and revenue, they are also subject to the scrutiny of Wall Street perceptions. One would think that the investment community could sort through the pros and cons of operational versus capital expenses, but data centers are the capital goods of the Internet era that can be touched, seen, and valued perhaps more easily than abstract areas like software intellectual property. It might be just a perception bias, but if Wall Street is involved, you want to do everything you can to look good and if hardware is what they want to see, then hardware is what you show.

Fitting the Conventional Corporate Finance Model
It might make perfect sense for a company to start its web operations in hosted data centers, but eventually the monthly bill gets to be big. Regardless of the industry, businesses have a tendency to want to capitalize large recurring monthly expenses. Particularly when the company thinks it can get a better return on its cash. So while technology restrictions and their removal might be a major driver from the engineering side of the house, the finance team is always looking at ways to optimize the use of company cash. And a physical data center is more easily financed than a large monthly bill from Amazon Web Services.

The Green Angle
The current public perception around energy use and climate impact is also driving a big part of the build-your-own-data-center trend. Referencing the new data center in North Carolina, Barry Schnitt , Director, Corporate Communications and Public Policy at Facebook said, “This project continues Facebook’s strategy of moving from leased data centers to owned facilities that are customized to be more cost-effective and efficient.” As the web and technology giants maneuver to control their image, staying in control of the green angle becomes even more important.

These are just some of the themes driving the build-your-own data center trends. Let us know if you have other ideas about what is behind these moves.

Gary Orenstein is Host of The Cloud Computing Show

Image courtesy of http://www.facebook.com/rutherforddatacenter

Big Data and NoSQL March to the Enterprise

Gary Orenstein — Sat, 30 Oct 2010 16:00:44 +0000

Big data and NoSQL solutions received welcome attention this year from developers and investors alike. The rallying cry of “No SQL” appears to have been tempered to include “Not Only SQL”, adding a degree of civility to an often heavily charged debate. Now big data and NoSQL vendors need to buckle down on their revenue models and make a head-on charge to the enterprise.

Let’s face it; while the web leads the innovation, the enterprise leads the economy. While it was absolutely necessary for a number of high-profile web companies to spearhead the adoption of big data and NoSQL — think Google and Yahoo with Big Table and MapReduce, or Facebook with Cassandra — it is time to explore larger and more repeatable markets with mainstream businesses in the U.S. and around the world.

We looked at who is commercializing big data over the summer, but just recently, we’ve seen both large and small amounts of money raised to direct these efforts. Cloudera, the Hadoop poster child, recently raised $25 million dollars, and according to the company, “The funding will be used to further invest in product development and services to support growth and adoption of Hadoop by enterprise organizations.”

A smaller investment of $2.7 million was announced by Riptano, the company supporting Apache Cassandra, one of the standout software projects to emerge in the NoSQL arena. Though not as direct in its enterprise marketing approach as Cloudera, Riptano offers the professional services, training, and support that I believe are likely to win it more enterprise business over time.

As an API for cloud storage is not enough to win enterprise adoption, a whole class of cloud storage gateway startups has spawned as a result, and I believe we’ll see similar handholding emerge around Big Data and NoSQL. Witness the expanding partner ecosystem around Cloudera, the vendor with the most magical Hadoop dust, or Riptano’s efforts with Quest. Commenting in an article, Matt Pfeil, CEO of Riptano said, “We’ve been surprised to discover much higher use of Cassandra within traditional enterprises than we anticipated.”

So what’s needed to spur more enterprise adoption? Time, for one thing. These new approaches to solving problems need to settle in with architects and developers as they conceptualize new solutions. Earlier this week, leaders from many NoSQL companies gathered for a large panel discussion well summarized by Todd Hoff of High Scalability. His notes include a few “what’s next” steps that fit hand-in-hand with enterprise adoption, including reducing barriers to entry, building features to compete with more established products, and scaling support.

We’ve come a long way in the maturity of new big data and NoSQL solutions, as well as in our understanding of when and where they fit. No doubt we have far to go, but the companies that can match these new approaches with clear and simple enterprise use cases are the ones likely to prosper.

Gary Orenstein is host of The Cloud Computing Show.

Related GigaOM Pro research (sub req’d):

Is EMC the Next Step in Oracle’s Journey to $100B?

Gary Orenstein — Sat, 16 Oct 2010 16:00:17 +0000

Oracle bought Sun Microsystems, hired former HP CEO Mark Hurd and declared that as “Oracle continues to grow, we need people experienced in operating a $100 billion business,” and ever since, the technology world has waited to see what other acquisitions Larry Ellison might have up his sleeve. This past week, we saw strong reactions to the rumor that Oracle might make a bid to buy EMC, due to the acquisition’s outlandish nature and monetary mismatch. Oracle will need to more than triple its revenue to reach that $100 billion target, so anything is possible.

That said, the rumor whips up a bunch of financial questions, because EMC owns 80 percent of VMware. EMC has a market capitalization of around $43 billion, and VMware around $32 billion. Match that up with EMC’s annual revenues of around $15 billion and VMware at $2.4 billion, and it isn’t hard to figure out where most of the value is, as well as where Oracle might be able to get a good deal on the multiple leading storage platforms.

So yes, the idea of Oracle buying EMC and VMware is a little crazy. But the idea of buying EMC and not VMware is within the realm of possibility, at least on paper, with The Register estimating that the non-VMware portion of EMC could be worth as little as $7.9 billion.

This is where things get interesting. The industry appears to be pushing towards server, network and storage consolidation following the moves of HP, IBM, Cisco, and Dell. Even Oracle has pushed a complete hardware and software package with Exalogic and Exadata using technology from Sun Microsystems to deliver an integrated solution. EMC and Network Appliance remain the large pure-play storage companies that could add significant heft to a server vendor that wants to dominate integrated stacks. HP and IBM have too much product overlap, and Dell can’t afford EMC, so that leaves an opening for Oracle and Cisco.

It seems likely that Oracle could be considering an EMC-only bid. I’ve heard some speculate that the reason Oracle became so tied to NetApp for certain solutions was the fear of EMC data center account control. Make no mistake; EMC knows how to close big deals, as their revenue number proves. If the goal for Oracle is to reach $100 billion, NetApp wouldn’t help them as effectively. NetApp currently has an $18 billion dollar market cap and just over $4 billion in revenue.

With Oracle, and potentially Cisco, interested in looking at a the EMC part of the equation, there could be impetus to move this deal forward. Even though Sun had plenty of great storage technology, they never had the commercial product success and storage revenues of EMC. If consolidation between servers and storage is the future, EMC better get cozy with someone soon.

Gary Orenstein is host of The Cloud Computing Show

Photo courtesy of Flickr user ell brown

Related content from GigaOM Pro (subscription req’d):