16-09-2016, 21:27 #1
[EN] Oracle Cloud: guerra de preços com AWS e Azure
Generation2 IaaS delivers twice the compute, twice the memory, four times the storage and ten times more I/O at a 20% lower price than Amazon Web Services.
Q1 FY17 Non-GAAP SaaS and PaaS Revenues grew 82% in Constant Currency
Redwood Shores, California—September 15, 2016
Oracle Corporation today announced fiscal 2017 Q1 results. Total Revenues were $8.6 billion, up 2% in U.S. dollars and up 3% in constant currency.
Cloud plus On-Premises Software Revenues were$6.8 billion, up 5% in U.S. dollars and up 6% in constant currency.
Cloud software as a service (SaaS) and platform as a service (PaaS) revenues were $798 million, up 77% in U.S. dollars and up 79% in constant currency.
Total Cloud Revenues, including infrastructure as a service (IaaS), were $969 million, up 59% in U.S. dollars and up 61% in constant currency.
Operating Income was $2.6 billion and Operating Margin was 31%. Non-GAAP Operating Income was $3.4 billion and non-GAAP Operating Margin was 39%. Net Income was $1.8 billionwhile non-GAAP Net Income was $2.3 billion. Earnings Per Share was up 10% to $0.43, while non-GAAP Earnings Per Share was up 4% to$0.55. GAAP and non-GAAP Earnings Per Share was negatively impacted by three factors: 1 cent because of a higher tax rate due to more cloud sales being in the U.S, half of one cent because of borrowing, and 1 cent due to strengthening of the U.S. dollar.
Short-term deferred revenues were $9.5 billion, up 4% in U.S. dollars and up 5% in constant currency compared with a year ago. Operating cash flow on a trailing twelve-month basis was $13.7 billion.
"Our Cloud business plus our On-Premises Software business grew 7% in constant currency in the first quarter, on a non-GAAP basis," saidOracle CEO, Safra Catz. "The overall top-line growth of our two strategic businesses was driven by non-GAAP SaaS and PaaS revenue growing 82% in constant currency, substantially outperforming our guidance. As our SaaS and PaaS business continues its rapid growth, we expect its gross margins to climb from 62% this quarter toward our 80% target."
"This year we are on track to sell more than $2 billion of SaaS and PaaS annually recurring revenue," said Oracle CEO, Mark Hurd. "We believe this will be the second year in a row that Oracle has sold more SaaS and PaaS than any cloud services provider. In the first quarter alone, we added more than 750 new SaaS customers including 344 new SaaS Fusion ERP customers – that's more ERP customers than Workday has sold in the history of their company."
"Next week at Oracle OpenWorld, we will introduce the second generation of our Infrastructure as a Service," said Larry Ellison, OracleChairman and CTO. "Our Generation2 IaaS delivers twice the compute, twice the memory, four times the storage and ten times more I/O at a 20% lower price than Amazon Web Services. IaaS represents a huge new cloud opportunity for Oracle to layer on top of our rapidly growing SaaS and PaaS businesses."
The Board of Directors also declared a quarterly cash dividend of $0.15 per share of outstanding common stock. This dividend will be paid to stockholders of record as of the close of business on October 12, 2016, with a payment date of October 26, 2016.
Última edição por 5ms; 16-09-2016 às 21:35.
16-09-2016, 21:47 #2
Oracle Profit Rises on Cloud Computing Growth
Oracle’s cloud-computing business posted another strong sales gain, but the quarterly growth was offset by steep declines in its conventional software-licensing business.
Sept. 15, 2016
Oracle is playing catch-up to Amazon.com Inc. and Microsoft Corp. as companies shift more of their computing operations off site and into the cloud. The company is rapidly building out its cloud business, particularly in the markets of selling access to web applications, known as software as a service, and selling access to tools to program and manage apps as well as analyze data, called platform as a service.
Those are among the hottest markets in enterprise technology. A recent survey from management-consulting group McKinsey & Co. found that the percentage of companies using public cloud-computing infrastructure as the primary environment for at least one workload will climb to 37% in 2018 from 25% in 2015.
In the latest quarter, Oracle’s total cloud revenue rose 59% to $969 million. Because Oracle books overseas sales in foreign currencies, the strong U.S. dollar meant that cloud sales would have grown 61% had it been measured a constant-currency basis.
But Oracle’s larger and lucrative business of selling software licenses to companies that run programs on site is shrinking rapidly. Sales of new software licenses fell 11% to $1.03 billion.
“It’s very difficult to create an upside scenario when you transition to the cloud,” said Stifel Nicolaus & Co. analyst Brad Reback. That is because margins are initially lower on subscription sales than on software licenses.
Mr. Reback said Oracle has missed per-share-earnings expectations in three of the past nine quarters.
Larry Ellison, Oracle’s co-founder and former CEO who is now executive chairman and technology chief, said the company had higher taxes than expected because it “overexecuted” in the cloud. That hurt margins, and reduced profits.
“This is actually the fact that we’ve done a very good job in the U.S. growing our cloud business,” he said in a call with analysts. “That had the strange effect of raising our tax rate. This is the best bad news you could get.”
The solid showing of Oracle’s cloud business comes days before the software giant welcomes an estimated 60,000 developers, partners and customers to its annual Oracle OpenWorld conference in San Francisco. There, the company is expected to debut new cloud services and outline its strategy to compete against rivals such as Amazon and Microsoft, which developed their web-services offerings earlier.
To compete with those giants, Oracle needs to, among other things, ramp up spending on the massive data centers required to handle customers’ computing processes and storage in the cloud. However, in the latest quarter, Oracle spent $299 million on capital expenditures, down from $446 million in the year-earlier period. The decrease helped to keep operating margins at 31%, flat with the year-earlier figure.
For the period ended Aug. 31, Oracle reported a profit of $1.83 billion, or 43 cents a share, compared with $1.75 billion, or 40 cents a share, a year earlier. Excluding stock-based compensation and other items, per-share earnings were 55 cents, compared with 58 cents a share expected by a consensus of analysts, according to Thomson Reuters.
Oracle is in the midst of acquiring NetSuite Inc. for $9.3 billion, which could help the company make inroads with smaller corporate customers. Co-Chief Executive Safra Catz said on the conference call that the deal, announced in July, has cleared antitrust reviews everywhere except the U.S. “where our waiting period expires at the end of September.”
Oracle, though, may need to sweeten its offer to acquire the maker of subscription-based, on-demand business applications after NetSuite’s largest institutional shareholder, T. Rowe Price Group Inc., said it wouldn’t tender shares in support of the current $109-a-share deal.
The firm cited “inherent conflicts of interest” between NetSuite and Mr. Ellison, who along with his family held nearly 40% of NetSuite’s common shares as of April, according to regulatory filings. Last Friday afternoon, Oracle extended the deadline from Sept. 15 to Oct. 6 “to facilitate the completion of outstanding antitrust reviews.”
JMP Securities Patrick Walravens said in a recent research note that the company will likely have to boost its offer to close the deal.
16-09-2016, 22:14 #3
Oracle Seeking Data Center Space for Cloud Push
An aerial view of Oracle's Utah Compute Facility in South Jordan, Utah.
By Rich Miller - July 8, 2016
There’s a major new player in the cloud data center arms race, as database giant Oracle Corp. is seeking to add capacity to support its push into cloud services, according to a new report.
As Oracle shifts its business model toward the cloud, it is seeking to lease large amounts of wholesale data center space, according to North American Data Centers, which tracks data center inventory across the country. Oracle began adding leased server space last year, when it leased 5 megawatts of capacity with Digital Realty in suburban Chicago, according to the report.
Now Oracle is seeking to expand its data center footprint in northern Virginia, currently the nation’s hottest market for cloud real estate. Real estate sources in northern Virginia say Oracle may seek to deploy as much as 25 megawatts of data center capacity in the region, according to Jim Kerrigan of North American Data Centers.
Oracle isn’t the only cloud player seeking large amounts of data center real estate. Microsoft is continuing its aggressive cloud expansion to support its Azure cloud services, and hopes to line up 30 megawatts of additional space in northern Virginia as well as additional capacity in the Chicago market, Kerrigan says. That’s on top of the 47 megawatts of wholesale space that Microsoft leased in the first quarter of 2016.
Neither Oracle nor Microsoft have finalized plans in northern Virginia or made public announcements of expansions in the region. But their data center shopping spree illustrates how the growth of cloud computing is turbo-charging the market for wholesale data center space.
Oracle Beefs Up Its Cloud
Oracle CEO Larry Ellison was once known for his disdain for cloudy jargon. But in recent years Oracle has begun shifting to a cloud delivery model.
Oracle now has offerings in all three sectors of the cloud market: software-as-a-service (SaaS), platform-as-a-service (PaaS) and infrastructure-as-a-service (IaaS). Oracle’s SaaS services include enterprise resource planning (ERP) and human resources applications, while its PaaS offerings are led by its Oracle Database Cloud Service, which allows customers to run databases in Oracle’s server farms rather than their on-premise facilities.
In 2015 Oracle entered the IaaS market, which has been dominated by Amazon Web Services. Oracle Cloud offers elastic compute, storage and networking services on a subscription basis.
These IaaS services may be attractive to companies that are running applications on Oracle’s SaaS and PaaS platforms and want to reduce latency in moving data between their public and private clouds (eliminating the round trip to a private cloud at Amazon, for example).
“There’s a huge amount of demand or our infrastructure-as-a-service from our existing SaaS customers and our new SaaS customers, and even bigger amount of demand for infrastructure-as-a-service from our database customers,” Ellison said in the company’s earnings call last month.
“We’re growing fast in SaaS. We’re growing fast in PaaS,” Ellison added. “Now we need to grow fast in infrastructure-as-a-service. We’ve made the investments, we have the right technology with our second-generation data centers, and we’re very excited about the potential.”
Competing with Amazon requires some infrastructure, as demonstrated in the major data center expansion announcements this year from Google and Microsoft. So it’s not surprising that Oracle would also need to add capacity to compete in this market.
Oracle runs its online services out of data centers in Austin, Utah and Scotland. It also houses servers with some third-party data center providers.
The evolution of the Oracle Utah Compute Facility (UCF) in South Jordan, Utah illustrates the ongoing changes in technology and best practices, and Oracle’s emphasis on continuous refinement of data center design to strike a balance between sustainability, cost and operational efficiency. The Oracle data center team described the company’s data center evolution last fall at the 7×24 Exchange conference in San Antonio.
But the buildout of the Utah facility occurred over two years. By shifting to wholesale space, Oracle would be able to fill data halls with servers on a shorter timeline. In recent projects, wholesale providers have demonstrated the ability to deploy new capacity from the ground up in as little as six months.
The newest data halls in the Oracle Utah facility operate with a PUE of 1.18, a significant efficiency improvement over earlier designs. The facility uses indirect evaporative cooling and flywheel UPS systems.
The design features a ductless cooling system using multiple “decks” housing plenums to separate and transport hot and cold air. Supply air is dropped into the data hall, and from there is distributed to provide cooling for electrical rooms and telecom space. Racks are housed in a containment system, with a chimney system venting exhaust air from the hot aisle into the upper return plenum.
16-09-2016, 22:20 #4
Huge Cloud Expansion Underway in Amazon US East
By Rich Miller - September 8, 2016
Amazon Web Services is aggressively adding data centers in Northern Virginia to boost capacity for its Amazon US East region, the company’s largest cluster of cloud computing server farms.
The expansion supports the rapid growth of Amazon Web Services’ lucrative cloud operation, which faces growing competition from Microsoft, Google and Oracle. The three tech titans are also building and leasing data centers as they race to gain ground on Amazon.
Northern Virginia is one of the primary battlegrounds in the cloud war. It’s of major strategic importance to Amazon Web Services (AWS), whose Amazon US East region spans more than 25 data centers across Loudoun and Prince William counties.
A key player in Amazon’s expansion is Northern Virginia is Corporate Office Properties Trust (COPT), a real estate investment trust that is a leading landlord for the U.S. government and defense contractors. COPT is known for leasing specialized space for national security tenants, but has quietly become one of the largest providers of cloud computing real estate.
1.3 Million SF of Data Centers
COPT has leased nearly 1.3 million square feet of data center space to Amazon Web Services over the past three years, according to SEC filings, all of it in Northern Virginia. That footprint spans eight different leases with Vadata Inc., the business unit that operates Amazon’s data centers, with annual rent of $15.3 million.
COPT recently delivered two new data centers to a tenant in Ashburn, and will complete a third facility later this year. Industry sources say the tenant is Amazon, and that the cloud builder is also the likely customer in two data centers COPT is building in nearby Manassas. Between them, the five new data centers span more than 815,000 square feet of data center space to be deployed over 15 months.
The developer doesn’t specify which of its buildings are leased by Amazon, consistent with its emphasis on confidentiality in working with government agencies. That aligns with the approach at AWS, which is careful in its disclosures about the infrastructure supporting its massive cloud computing operation, which now generates nearly $10 billion in annual revenue.
US East: Where Amazon’s Cloud Began
While the Amazon Web Services cloud spans the globe, a huge chunk of its Internet infrastructure is concentrated across Loudoun and Prince William counties in northern Virginia. It’s America’s busiest Internet intersection, the place where all the networks meet, anchored by the Equinix interconnection hub in Ashburn. That’s why Amazon’s initial company-built data centers sprang up in Ashburn and Sterling (the town next door).
These initial data centers form the nucleus of Amazon US East, its oldest and largest of its cloud regions. Amazon doesn’t disclose the full scope of its infrastructure, but third-party estimates peg its U.S. data center network at about 700 megawatts of IT capacity, with as much as 500 MW of that total focused in US East.
The recent construction builds on Amazon’s existing footprint. The company operates nine data centers in Sterling and six in Ashburn (with three more under construction) as well six in Manassas (with two more in the works). The company also has two data facilities in Chantilly, and is planning to expand its infrastructure into Haymarket in Prince William County.
Why Northern Virginia Matters
Amazon’s huge presence is one reason why Northern Virginia is the home of the cloud. Loudoun County is home to more than 60 data centers, with 6 million square feet of data center space, while Prince William County says it hosts an additional 2 million square feet of data centers. Microsoft and Oracle have both been beefing up their data center capacity in “Data Center Alley” this year.
Northern Virginia is a market COPT knows well from its work with government agencies and systems integrators. COPT Data Center Solutions owns more than 60 data center facilities, representing about 3.5 million square feet of space. A big chunk of that – more than 1.5 million square feet – is built for U.S. Government and defense contractors, the company says. Service provider customers include Northrop Grumman (which operates two data centers for the Virginia Information Technologies Agency) and Computer Sciences Corp and AT&T, which each lease about 300,000 square feet of space from COPT.
Amazon’s single-tenant buildings represent about a third of the data center footprint for Corporate Office Properties Trust. COPT also operates the DC6 wholesale data center facility in Manassas (formerly known as PowerLoft).
Amazon’s Data Center Template
In its work with Vadata/Amazon, COPT builds out powered shells: undeveloped space with the power and fiber connectivity already in place. Amazon then fills the building with its custom-built data center infrastructure. Amazon fine-tunes its custom servers, storage and networking gear to get the best bang for its buck, offering greater control over both performance and cost.
Amazon standardizes its data centers to house between 50,000 and 80,000 servers, according to company presentations. That consistent approach can seen in COPT’s recent projects. All three of the new Ashburn data centers are 149,000 square feet, as are three of the data centers COPT has built and leased in Manassas.
Amazon Web Services has improved its profit margin in recent months, which company executives credit to the data center operation. “We’ve been seeing some great efficiencies in our infrastructure, both internally as Amazon and also as part of AWS,” said Amazon CFO Brian Olsavsky in the company’s recent earnings call. “We have great people working on not only better efficiency, but also driving cost out of our acquisition prices.”
The relationship with Amazon also appears to be helping COPT’s bottom line. Over the past year, the REIT’s revenue from leasing data center shells has grown from $5 million per quarter to $7.3 million per quarter. COPT has also benefited from improved leasing at its DC6 wholesale facility, filling 15.7MW of the available 19MW of power, boosting revenue from $3.8 million per quarter to $6.8 million per quarter.
COPT’s success in the cloud sector led to a $104 million gain from a deal in which it formed a joint venture with GI Partners, one of the most experienced data center investors. The venture, known as GI-COPT DC Partnership, acquired six of COPT’s single-tenant data centers, including several leased by Amazon Web Services. “This venture demonstrates the strength of demand for strategically located data center properties leased to high credit tenants,” said Stephen Budorick, President & CEO of COPT.
The New Frontier: Haymarket
Amazon isn’t done expanding its cloud infrastructure in Northern Virginia. The new frontier for Amazon US East appears to be Haymarket, a small town just west of Manassas. Vadata has applied for generator permits totaling 75 megawatts at a site in Haymarket. The development timeline on this expansion is not clear, as a proposed Dominion Virginia power line to support Amazon’s growth has been delayed by lengthy wrangling.
Dominion’s plan for an overhead 230kV line has been opposed by a group of local residents, who have termed the project a “very expensive extension cord” for Amazon. The Coalition to Protect Prince William County wants Dominion to divert the power line or bury it in sensitive areas close to homes and wildlife. The state of Virginia will have the final say, but has been considering arguments and motions in the case for more than a year.
An interesting wrinkle in Amazon’s infrastructure is the geography of Amazon’s specialized cloud for government IT workloads, which is currently hosted in the Pacfic Northwest rather than Northern Virginia. GovCloud is an isolated AWS region designed to host sensitive data and regulated workloads in the cloud, while meeting strict U.S. government compliance requirements. AWS was one of the first cloud providers to meet the Federal Risk and Authorization Management Program (FedRAMP) High baseline, a set of security requirements for cloud services that includes over 400 security controls.
The Road Ahead
Amazon may yet extend GovCloud to its US East region. In the meantime, Amazon’s relationship with COPT may help reassure federal and local government agencies that the Amazon US East facilities meet their requirements.
“Security is in COPT’s DNA,” the company says, noting its historic focus on security. That includes Anti-Terrorism/Force Protection (ATFP) compliance, with a controlled perimeter greater than 150 feet from the building, biometric access controls, man traps, video surveillance zones, and extensive experience building SCIF (Sensitive Compartmented Information Facility) space for government clients.
What seems clear is that the cloud-driven building boom in Northern Virginia is likely to continue. There’s at least 4 million square feet of cloud data centers being planned in Loudoun and Prince William counties, including the new projects in the pipeline for Digital Realty, DuPont Fabros Technology, Equinix, RagingWire, INFOMART, Sabey and Iron Mountain.
That’s why Amazon is busy building as well, ensuring it doesn’t run out of capacity for the clouds to come.
“There’s plenty of room for multiple vendors in this business,” said Amazon’s Olsavsky. “What we focus on is innovating on behalf of customers and expanding the geographic footprint to make our services more widely available.”
16-09-2016, 22:26 #5
Microsoft Accelerates its Cloud Growth With Data Center Deals
By Rich Miller - April 13, 2016
The cloud computing arms race is accelerating, and the battle will be waged with data centers. The leading players are moving quickly to amass the capacity they will need for the clouds to come, resulting in huge deals for data center space under development.
Microsoft has signed massive leases for data center space in three key markets, according to a new report from a leading data center real estate specialist. The company has reserved 47 megawatts of space with deals in northern Virginia, Silicon Valley and San Antonio, which will enable Microsoft to quickly ramp up its cloud operations to meet demand.
Microsoft’s leasing, which includes several of the largest deals in the history of the data center industry, was highlighted in a new report from Jim Kerrigan of North American Data Centers, who tracks wholesale inventory across the U.S.
Kerrigan reports Microsoft has signed two leases with CyrusOne – including a massive 22 megawatt deal for space in Sterling, Va. and a 9-megawatt lease in a San Antonio – and is also the tenant in a 16 megawatt lease with DuPont Fabros Technology in Santa Clara, Ca.
The deals highlight how Microsoft is both building and buying infrastructure to support the rapid growth of its cloud business, sometimes in the same market. That’s the case in San Antonio, where Microsoft operates a huge company-built data center but is also reportedly leasing space from CyrusOne, which recently completed its 196,000 square foot San Antonio II facility.
Neither Microsoft nor its service providers have confirmed the recent leases. But the expansion is consistent with Microsoft’s public discussion of its cloud ambitions, and will help the company keep pace with its chief rivals in the fast-growing public cloud, which provides instant capacity for companies to run applications in other companies’ data centers.
“We operate over a million servers in data centers all over the world,” said Mark Russinovich, the Chief Technical Officer for the Microsoft Azure cloud platform, in his keynote at the recent Open Compute Summit. “Azure’s footprint operates in 22 regions. We’ve got six more coming online this year, and many more will be coming in the future.”
Cloud Building Boom
The rapid growth of cloud computing has spurred a building boom for the major Internet platforms, as Google, Microsoft, Amazon and Facebook are all super-sizing their cloud campuses to add data center capacity. Meanwhile, developers of third-party data center space are also scaling up their operations so they can lease space to cloud builders.
The rapid procurement of data center space by Microsoft follows last month’s announcement that Google is ramping up its data center capacity, with plans to add 12 new compute regions for Google Cloud Platform over the next year.
Both Microsoft and Google need cloud capacity as they seek to keep pace with market leader Amazon Web Services (AWS), which has built a massive cloud infrastructure to support its growth. That includes a large cluster of data centers in northern Virginia, where it is estimated to have a capacity of about 500 megawatts of power to support servers and storage.
Amazon is building new capacity in several sites in northern Virginia, including “Data Center Alley” in Ashburn as well as further west in the Manassas area.
Last week Amazon CEO Jeff Bezos said that AWS will become a $10 billion business this year. In a letter to shareholders, Bezos said AWS has 33 Availability Zones across 12 geographic regions worldwide, and will add another five regions and 11 Availability Zones in the coming year.
Regions are becoming more important for global cloud growth, as surveillance controversies have sharpened the focus on data sovereignty and customer privacy. As more countries seek to keep citizens’ data within their borders, cloud providers are adding data centers in these markets to accommodate these demands.
“The number of region for Azure is more than the number of region for AWS and Google,” said Kushagra Vaid, the General Manager for Azure Cloud Hardware Engingeering. “We view this global presence as a differentiator.”
How Open Compute Powers the Azure Cloud
At the Open Compute Sumit, Vaid noted that the servers now being deployed in the Azure cloud are based on Open Compute Server (OCS) designs, which Microsoft contributed when it joined the Open Compute Project in 2014.
“This is what’s powering Microsoft’s cloud,” said Vaid. “Over 90 percent of what we deploy today is based on this OCS design. We see this as a very key part of a broad ecosystem that takes the experience of a major cloud provider and applies it to the enterprise.”
Indeed, when Microsoft looks at Open Compute, it sees the enterprise. The key benefit, Vaid said, is that opening its designs makes it easier for customers using Azure for their public cloud presence to use identical infrastructure in their corporate data centers, making it easier to create hybrid infrastructures that can seamlessly move data and apps between private clouds and the public Azure platform.
“We get to tap into a larger supply ecosystem,” said Vaid. “Our enterprise customers tell us ‘we want to use what you’re using,'” he said. “We take our customers to Quincy (Washington) and tour them through our data center. They say ‘this is fantastic. I just want to buy what you’re buying.’ ”
They can now introduce those customers to vendors who support Microsoft’s Open Compute Server. “Organizations of all sizes can take advantage of our innovations in order to improve the performance, efficiency, power consumption and costs of their data centers,” said Russinovich.
Good News for Data Center Builders
The robust leasing activity by Microsoft demonstrates how the growth of cloud computing is providing a huge boost for data center developers. It’s long been expected that cloud business models would eventually impact the market for on-premises data center space, especially smaller server closets. But cloud growth is now driving phenomenal demand for additional “plug-n-play” data center capacity, a trend that could continue for some time to come.
“Large blocks of built-out space will be key as cloud providers attempt to gain market share by taking down existing MTDC (multi-tenant data center space),” Kerrigan wrote in his latest market report.
The focal point for that demand is northern Virginia, which has seen vigorous demand for space. “The relatively low costs for space and power, coupled with the unique density of carriers, content providers, and cloud firms, have driven demand for data centers in the area and will continue to do so,” said Kerrigan, who noted that the state of Virginia recently extended data center incentives that will provide significant tax breaks on servers and software through 2035.
16-09-2016, 22:35 #6
Size matters in the cloud data center world, says Microsoft
By Stuart Lauchlan
September 15, 2016
Earlier this month, Microsoft made the bold claim of being “the first global provider to deliver complete cloud from UK data centers” as it eyed up the public sector market. At the time, I raised a skeptical eyebrow as to how this claim was justified, given the presence of firms like Salesforce and IBM operating in the UK government market as well with their own in-country data centers.
While I’m no closer to a specific answer to that, with happy timing Scott Guthrie, Microsoft’s EVP Cloud and Enterprise, used an appearance at this week’s Deutsche Bank Technology Conference to drill down into the firm’s global data center thinking and stake some further bold claims. Guthrie began by emphasizing the global nature of cloud computing:
The cloud needs to be something that works all over the world, and in particular, can meet the unique compliance, data sovereignty requirements that organizations have around the world in order to run their business. We’ve been hard at work building out our cloud infrastructure over the last several years and we now provide our services in more than 34 Azure regions around the world, where customers can basically in a matter of seconds deploy and run their own code and deliver their own set of applications and solutions to their employees and their customers.
Guthrie wasn’t shy of setting out some big numbers to back up his pitch:
We’ve got more than 120,000 new customer subscriptions being created every month, about 1.6 million production databases are now being hosted in Azure alone, more than 2 trillion Internet of Things messages each week, more than 5 million organizations that have been through identity server and integrated their user employee security as part of our cloud, 4 million developers.
All of this leads to the inevitable conclusion that size matters. Guthrie boasted:
We have more regions than AWS (Amazon Web Services) and Google combined.
In particular, we don’t just have broad coverage of regions but also meet unique data residency promises that no other cloud vendor delivers. Whether those are in China, whether that’s in Germany, whether that’s for US Government and Department of Defense, we can make guarantees around who has access to the data. For example, in Germany and China, we make promises [that] no American, and no Microsoft employee has any operational control over those data centers or that data to be able to meet those specific markets.
Size is also on the agenda when it comes the data centers themselves, Guthrie claimed, citing an example of one in the East US Region which is two miles long and will ultimately host a million servers. You need to be a particular type of company to pull that off, he argued:
This kind of scale at that global level is something we fundamentally believe, really very few companies in the world are going to be able to provide, and for the most part we see ourselves [in a] two-horse race today, with AWS as that provider.
The only other credible vendor could be Google, admitted Guthrie, but insisted that it lacks the enterprise capability at present:
Even the way they have built their infrastructure historically, it’s really been optimized search. If you look at, for example, the number of regions around the world they operate in, they currently, I think, have four versus our 34.
That undermines Google’s prospects in the enterprise cloud space, suggested Guthrie:
If you want to compete in the UK and you want to go to a bank in the UK and say, ‘Hey, you should consider using us for cloud. [But] we can’t guarantee your data is going to stay in the country, we can’t actually meet your regulator needs, and I’m not sure when we do support whether we’re going to be able to have a human on the end of the line who would pick up the phone’. That ends up being a non-starter for that conversation. So I do think they are going to struggle a little bit, certainly over the next couple of years in terms of building up that enterprise credibility and being able to get to that point.
There’s also the security aspect to factor in, with Guthrie pitching the line that this is far from being an issue that’s going away as a factor in cloud adoption:
In general the threat environment that we all live in now is significantly scarier than it was a decade ago versus two decades ago. The adversaries out there are getting more sophisticated, the types of attacks that are happening are happening more frequently, and that’s just going to be the new normal for us going forward. The important thing when you think about security is, you’ve got to be paranoid. You can’t take anything for granted, and any vendor who says, ‘Use my stuff and you will be perfectly secure’, run away from them because they don’t get security or they are lying.
That prospect – being lied to by your vendor – raises the question of trust, a theme which Microsoft’s ‘friend-with-benefits’ in the cloud, Salesforce, has been majoring on for some time. Guthrie posits trust this as a major differentiator between Microsoft and the competition, including, presumably, Salesforce:
We have now more trust certifications with our cloud than any other vendor. We meet – whether it’s individual industries, whether it’s individual country boundaries – the credentials necessary for companies to bet their business on top of us.
How that claim goes down with Salesforce in a week when Microsoft’s other big news was that it has won HP Inc’s CRM business away from Salesforce, remains to be seen. But Guthrie did attempt some diplomacy about that deal, even if it wasn’t entirely successful:
I think the HP Inc take out yesterday wasn’t a Salesforce take out, it was…they are very large Salesforce shop – or they were until yesterday. They are planning a massive migration and big bet on Dynamics.
And HP Inc isn’t the only one, he hinted:
It’s one of the more public ones that we’ve had, but we’re starting to see some really good success in the market at a broad level, and particularly, right enterprise. Now let’s [just] say, for CRM in particular, enterprise is our sweet spot…in particular for customer engagement solutions, I think we’ve got a very strong enterprise product and we’re growing very, very fast.
16-09-2016, 23:12 #7
CenturyLink Verizon Business Data Services Feud Erupts
9/16/16 at 1:35 PM by Joan Engebretson
A CenturyLink Verizon business data services feud arose today, with CenturyLink accusing Verizon of “doubletalk” on the issue of business data services (BDS) competition. It’s a topic that is being intensely debated as the FCC considers whether to impose price controls in this market, which includes a wide range of wholesale data services.
CenturyLink notes in a press release that in filings made jointly with competitive carrier association Incompas, Verizon has argued that competition is lacking in the BDS market. But, according to CenturyLink, that point of view is contradicted in a white paper that Verizon submitted in support of its planned acquisition of XO Communications, a competitive provider of BDS services.
“[T]he white paper highlights the intense competition in the BDS marketplace,” comments CenturyLink Vice President of Federal-Regulatory Affairs Melissa Newman in a letter sent to the FCC.
CenturyLink Verizon Business Data Services Positions
Traditionally, large incumbent carriers such as Verizon and CenturyLink have been on the same side with regard to BDS and have fought price controls. But as Verizon’s wireless business becomes increasingly important and the company faces the prospect of purchasing a large amount of data links from other carriers, the company worked out what an Incompas executive has called a “compromise” BDS proposal that supports FCC oversight of pricing for non-competitive markets and services
Verizon’s plan to acquire XO was not driven primarily by a desire to gain XO’s BDS business. What Verizon really wants is XO’s wireless spectrum licenses in frequency bands suitable for 5G deployments. But along with those come the XO BDS business and it’s not surprising that Verizon is trying to assuage FCC concerns that the acquisition would reduce competition in that market.
One other large incumbent carrier besides Verizon that has broken with the pack on BDS is Windstream, which increasingly has emphasized enterprise business services in recent years. Windstream has to buy a lot of circuits from other carriers to gain nationwide scope and according to the company’s chief financial officer, the ratio of Windstream BDS connectivity purchased to BDS connectivity sold is four-to-one.
Verizon provided a statement to Telecompetitor regarding CenturyLink’s position.
“CenturyLink’s characterization of the Verizon/INCOMPAS joint proposal is wrong,” said Kathy Grillo, senior vice-president and deputy general counsel, public policy and government affairs for Verizon. “Our joint proposal is not a commentary on the current state of competition. It is a compromise intended to resolve an issue that has been overhanging the industry for years.”
Grillo went on to say, “We have always maintained that regulation is only necessary where competition does not discipline the market. Of course, many markets are already competitive. In markets that are not sufficiently competitive, Verizon and INCOMPAS have offered a framework to address them. Instead of trying to poke holes in the only meaningful compromise on the table, we encourage all stakeholders to work together to find a reasonable path forward.”
17-09-2016, 11:53 #8
IBM cloud chief: The next phase of cloud is a race to add value
Robert LeBlanc explains how IBM’s portfolio of SoftLayer IaaS, BlueMix PaaS and cognitive Watson services add value atop basic cloud services
By Brandon Butler and John Gallant
Network World | Sep 12, 2016
At 105 years old, IBM has been through more than a few major technology transformations. Arguably, none is bigger than the current retooling of the company around cloud computing – an effort overseen by Robert LeBlanc, senior vice president, IBM Cloud. LeBlanc says IT leaders and business executives aren’t caught up in Wall Street’s worries over IBM’s revenue declines or the pace of the cloud transformation.
Instead, they’re focusing on the higher-value capabilities – like Watson cognitive computing or internet of things functions – IBM is layering atop what LeBlanc calls the ‘phase one’ cloud that will help them transform their own businesses.
In an interview with Network World Senior Editor Brandon Butler and IDG Chief Content Officer John Gallant, LeBlanc talked about how IBM is tailoring its cloud services to specific vertical industries and what Big Blue is doing to enhance its Platform- and Infrastructure-as-a-Service capabilities. He also discussed why partnerships with companies ranging from VMware, Box, SAP and Workday are strengthening IBM’s cloud play.
John Gallant: We want to make sure our readers understand the overall cloud strategy IBM has embraced and the major transition IBM is going through. What is the overall cloud strategy?
It’s a three-pronged strategy. One is that we’re very much focused on cognitive solutions and that’s really bringing analytics and AI to all of the solutions, which are very much industry oriented. We started with our focus on healthcare and we’re extending that to other industries. The second is that everything we do is on the cloud. We are focused on building all of these solutions on the idea of cloud. We’re also focused on bringing some of our traditional offerings onto the idea of cloud. As an example, all our relevant middleware offerings are now available on the cloud so the client can run it on-prem and they can run it inside the cloud. The third is we’re doing everything in an industry context and that really feeds into what we’re seeing from clients here and the next generation of the cloud.
Let me expand on that a little bit. In the first wave, everybody looked at the cloud as a way to save money. I could get cheaper compute, cheaper storage, cheaper networking and the cloud was a good way for people to get better economics. In some cases, they looked at that as: ‘Hey, I’m going to move my application to the cloud or I’m going to burst to the cloud. Instead of buying that extra server I’ll just run it on the cloud when I need to’. I think we’re moving to the next phase of cloud and that’s where the cloud is becoming a platform for innovation.
What I mean by that is the cloud is enabling clients to do things today that they otherwise wouldn’t have been able to do – certainly not at the speed and scale that they have in the past. This is starting to spawn new business models. It’s spawning new business processes. It’s allowing clients to move into mobile at a rate and speed that they’ve never seen before. The reason is they now have access to capability and technology that before would have required a level of investment. They would have to procure servers, configure them, get them all ready, buy software and that literally can take months when in most cases now, with little or no investment they can get access to newer technologies. That’s what I call the value and where the cloud is now becoming a platform for innovation.
A couple of examples on the IBM side and what we’ve done with Watson: We’ve got over 30 cognitive services now that are available on the cloud. In fact, the only way we’re delivering our cognitive services is on the cloud. Whether I’m a small startup company or I’m a large enterprise, I have access to that set of technology capabilities without having to make a large, upfront investment. That enables me to do a lot more sandboxing, a lot more testing, trying things and seeing where I can come up with new innovation.
We’re the first to put blockchain services up on the IBM cloud and we offer different levels of security with that blockchain capability. Look at Internet of Things and the connection of all these physical devices, as well as all the services you need to ingest data at rates that most companies or enterprises couldn’t ingest at. You’re talking billions of pieces of information being ingested at a pretty dramatic rate.
Hence, why we bought the Weather Company. When we bought the Weather Company, everyone said: ‘Wow, IBM is getting into the weather data business’. Yes, weather data is important and we think that weather data actually impacts a lot of business processes, whether it be supply chain or airline scheduling. We have data that we can now help clients with but more importantly, what the Weather Channel gave us is the ability to ingest data and deal with data at scale. We got a lot of technology that we’re applying into the internet of things space. Again, it’s value added on top of the cloud.
JG: In her presentations, CEO Ginny Rometty talks about this transition in terms of IBM focusing on higher value opportunities. What does that mean from a cloud perspective?
Some of the cloud vendors will tell you to bring your applications, bring all your data and you can run it on my cloud. That’s what I call phase one of the cloud. Phase two of the cloud is the additional value. I can bring IoT services, I can bring cognitive services, I can bring a PaaS platform, our Bluemix. We’ve got over 150 services on our Bluemix platform which runs on the cloud, really allowing the higher value. Those are the things that clients can do more with. Instead of just giving them compute and storage and elasticity and allowing them to scale up and down and pay for what they use, it’s also the functions that will run on the cloud that I now have access to that I can [use to] build my next new application or my next solutions. Instead of having to build everything from scratch, I now have a real set of building blocks on which to build those next-generation applications.
Brandon Butler: We saw that cloud revenue was up 30 percent in the last quarter to about $3.4 billion. Can you outline what falls into that cloud revenue bucket?
We split our cloud revenue externally into two pieces. We talk about the total cloud and we always talk about the trailing 12 months because the world of cloud is moving very quickly. Our last 12-month trailing is $11.6 billion, of which $6.7 billion of that is as-a-service. That’s where we provide the capability to the client as a service and that’s what a lot of people think about when they think about the cloud.
As an example, our SoftLayer platform where we provide the compute and storage as a service, that’s all included in that number. The difference between as-a-service and the total cloud is what clients are doing in the private cloud. That’s where clients are building out their own clouds and in a lot of cases they’re actually building hybrid clouds in which they have a private cloud connecting to a public cloud. In fact, most enterprises are building some kind of hybrid solution. There are some applications and solutions that will run just on the public cloud but in most cases the clients are trying to connect back to traditional transactional systems, to their databases, to their data and so they’re building up these private clouds. In fact, if you look at data from IDC and others, there still is more spend in private cloud than there is in the public cloud. You were at VMworld recently, correct?
You saw some of the data that [VMware CEO] Pat Gelsinger showed about private clouds versus public clouds. We’re seeing it as a combination, so when we say $11.6 [billion] that includes what we’re doing for clients on private cloud. Remember that also includes some of the things that we do in terms of our strategic outsourcing and our consulting business where we’re helping clients build their cloud strategy. We’re helping them build their private clouds as well as what they do in the public cloud.
17-09-2016, 11:56 #9
BB: Is your sales force fully onboard with selling those cloud services versus the traditional on-premise offerings? What have you done to incent the sales force to make that shift?
There are always two things you do with a sales force. One is you have to train the sales force. They have to be aware of the offerings and how to put the offerings in context with what the client is trying to do. We are training our sales force in the cloud and honestly, when you talk about the cloud you have to talk about it in the context of industry. We are educating our sales force in how you put the cloud in the context of the industry I’m in.
I talked about business models and business processes being affected by the cloud but you have to understand the industry you’re in to understand the business process and models that are being affected. That’s where most of the disruption, in fact, is happening. One is certainly getting our sales force to understand how we can bring value and what we have in our portfolio to help them make the transition. We’re undergoing a transformation. Every client, every industry I’ve talked to in the last year is going through some form of transformation. Some people call it digital transformation. There are different names for it but everyone is going through that transformation. That’s number one. Number two is how we compensate. We have changed the compensation model so a big part of a salesperson’s compensation is their ability to sell our cloud offerings.
JG: Investors and financial analysts have lots of questions about what’s happening with IBM’s revenue and the length of this transition and whether it is taking too long. Do you get those same kinds of questions from IT buyers? Are they concerned about that transition or do they see this as an absolutely necessary move for IBM? What kinds of questions are they asking you about it?
I don’t get a lot of questions about the financials from clients. They’re more [concerned] about where we see the cloud going. What are the things I need to do to prepare for that transformation? Almost every client I talk to is committed to the cloud. There’s no maybe anymore. Everyone is going to the cloud. Now they’re trying to decide what to move to the cloud. There are certain business processes, certain applications that are natural to move to the cloud. Some of them will stay on-premise. Sometimes that on-premise is driven by regulatory, security, the amount of data they have. There are a lot of factors that go into it but in most cases they end up with hybrids.
A lot of the dialogue I have with the client is: ‘What do I move, how do I move it, when do I move it?’ They’re looking for help and guidance. They’re looking for capability. I don’t have a lot of discussions about IBM financial results with clients. What they want is to make sure we’re in it for the long haul. They know as they make the transition this is a decade-long commitment that they’re making. The other piece that I hear a lot from clients is [about] culture. How do they make the shift in culture? They’re always asking for my experience and insights into how you make that transformation because inside of IBM we’ve had to make that transformation.
JG: From the CIO perspective, when you talk to them about the IBM transition, what would you say is the single biggest worry they have? What’s the one thing they don’t want IBM to screw up?
I guess they’re always worried about two things. Are you bringing the A-team skills to help me make the transformation that I’m undergoing and are you going to stay in the public cloud longer term? I point them to our strategy. We’re as dependent on the cloud, if not more, than our clients are. If I’m not in the cloud, then I don’t have a Watson business because everything I do is on Watson. If I’m not on the cloud I’m not in the internet of things business, I’m not in Blockchain. All of those value-added services and solutions that we’re building for ourselves and for our clients are all on top of the cloud. The cloud is a foundational element for the IBM business and also for our clients. I talk to them about the level of investment we’re making, which is in the billions of dollars and about the progress we’ve made in the last 18 to 24 months, really starting when we acquired SoftLayer in 2014.
BB: Robert, I want to get back to hybrid cloud. In her annual letter to IBM shareholders, Ginny said that hybrid cloud is the fastest-growing segment of the cloud market and that IBM is the global leader in hybrid cloud for the enterprise. How is IBM better than, say, Dell/EMC or Hewlett Packard Enterprise for hybrid cloud?
Think about hybrid cloud: It is the combination of private and public. For some players, it’s hard to be a hybrid cloud provider if you don’t have part of the equation. Being a world-class public cloud provider is a perquisite to being a world-class hybrid cloud provider. You’ve got to enable the client to build these solutions and put the solution or the application and the data where it makes sense for their business. As an example, if you’re building a new mobile application, most mobile applications today are being built out on the cloud. You’ve got to have a world-class cloud platform to enable mobile. You’ve got to be global. Some of these players are more local. Most enterprises are global. Even smaller companies have aspirations to be global so you’ve got to have data centers that are around the world. We’ve got almost 50 data centers now as part of our cloud around the world. In fact, just last week I was in Korea opening up our Korea cloud data center.
Clients are looking for the ability to deploy their solution anywhere around the world and to have data local. There are a lot of companies that have data regulation, data security, privacy regulations that force some of the data to be in-country. We give that flexibility to the client so they can decide where the application runs. They can decide where the data resides. If I’m running an application, I may have an instance of that application running in Country A and the same instance running in Country B. That really is the client’s decision.
When you think about some of the other players, they may have one piece or the other. We’ve got the on-prem; that’s our history, that’s where we come from. We understand how to securely connect applications to data, securely connect back into transactional systems. That is where we have a unique set of capabilities, a unique understanding, since we were a major player in that generation of on-prem IT. Our ability to map and to connect them to the public cloud and a world-class public cloud solution is unique.
I’ll give you an example.If I’ve got a large application and it’s handling a lot of data, for me to move the application to cloud is one thing. For me to move the data is another thing. If I have transactional systems that are producing new data every day, if I’m having to replicate the data back and forth, that becomes a very expensive proposition. What I hear from clients is: ‘Yes, compute and storage is a big part of the public cloud but I’ll tell you, I’m spending more money on networking and moving data than I am on compute and storage.’ Having to move data and replicate data, if you have to do that in a hybrid world, is not a cost-effective way to manage my environment.
17-09-2016, 11:57 #10
BB: Partnerships seem to be another important part of your strategy. You have an expanded partnership with VMware that you spoke about at VMworld. You have a partnership with Workday. Can you explain the overall partner strategy for advancing the cloud position and what should customers expect in the way of future partnerships?
One type of partnership enables clients to do things they couldn’t do before. VMware would be a great example of that. I would put Apple in the cloud in that bucket where we enable clients to move workloads from on-premise onto the cloud. They can consistently manage and have a consistent platform and a stack where they can build the application once and run it where it makes sense, whether it be on-prem or in the cloud. Then you want to automate that such that it’s easy for them to move from one to the other. Sometimes they’re going to start in the cloud and build the application and move it on-prem. Sometimes they’ll take an on-prem current application and move it to the cloud. We want to make it easy for them to go back and forth to map to whatever business model or whatever business objectives they have.
We had Marriott onstage with us (at VMWorld). That was an important thing for Marriott because they want to have that flexibility. It’s technology that clients use on-prem to enable them to use that in the cloud and that drove a lot of the VMware partnership. It’s the same thing with Apple because one of the pieces of the Apple relationship is Swift. As you know, Swift is heavily used in building the mobile application but clients need to be able to connect the mobile application to backend systems. We enable Swift to be run in the enterprise and in the cloud. That’s enabling new capability in the connection of this hybrid notion. That’s one class of partnership.
The other class of partnership is like what we’ve done with SAP, Box and Workday, in which they’re looking for a world-class public cloud so that they don’t have to spend all their time and energy investing in their own private cloud. They want to go public cloud and for them, the other thing that’s attractive is my global presence. Because they’re all trying to expand their businesses into global markets and the fact that I’m sitting already in data centers all through Europe, I’ve got nine data centers in Asia, enables them to take their offerings and get into new markets quickly without having to invest all of their money on building out the infrastructure. It opens up the opportunity for them to expand their business into new markets.
JG: You had talked earlier about the industry-specific focus of your efforts. Some analysts we had talked to in advance of this spoke about a recent internal structural change, the creation of an industry platforms group. Can you tell us about that?
We made an announcement internally here at IBM. Look at what we did with Watson Healthcare. That really became a model that we believe will permeate through multiple industries. We will go after the other industries and replicate what we’ve been able to do in healthcare in those industries. Financial services would be a natural for IBM – given our position in financial services and what we’re doing in technologies like blockchain – and helping financial services companies move into the cloud.
JG: Just so I understand, Robert, how is that group structured? That’s one group that oversees all of the industry-specific capabilities you’re bringing to market?
Yes. There are certainly a lot of things we do that are common to industries. The cloud is an example. The cloud is the cloud. The cloud doesn’t know an industry. There are separate capabilities that we will build on top of the cloud that are unique to a certain class of applications or certain industries. Blockchain is a great example for financial services. You saw new acquisitions in healthcare of Truven, Merge, and Phytel. There is a set of applications and data that is applicable to healthcare. You would anticipate that we would continue to do that now in other industries and being able to take our Watson technology and apply our Watson cognitive technology into multiple industries. Most industries are data driven and process driven and any time you can take data and apply things like cognitive and analytics, you get insight. When you get insight you improve the process, you come up with new business models and you can drive that digital transformation. Every industry that we look at is going through some form of digital transformation. What we want to do is we put that unit together to focus in on specific industries as we go forward.
JG: I would assume this also has a big impact on who you sell to. Is it bringing in more line of business executives that you have to talk to about these industry-specific capabilities?
I think that’s a good observation. The people who are making decisions are not just the technical team or the CIOs anymore. The lines of business are heavily involved. When cloud was all about compute and storage that was a CIO decision. Now that the cloud is becoming a discussion around business process, business models, data specific to an industry, the line of business is now being involved so it is a different sell. It is much more of a consultative sell, hence why Bridget van Kralingen, who ran our Global Business Services unit, was a natural to run this. We believe it’s much more of a business transformation than it is just technology. Technology is a means to an end. I’m partnering closely with Bridget to make sure that I provide the best platform, the cloud that enables that level of transformation that happens in an industry context.
BB: Robert, I wanted to ask you about the public cloud. In Gartner’s latest Magic Quadrant report, Amazon Web Services continues to be the market leader. Microsoft is seen as a second place vendor and it seems like there’s a whole host of other vendors fighting for market share, including Google, IBM and others. What do you think of that characterization of the public IaaS cloud market and where do you see IBM fitting in?
Gartner has a very narrow definition of what the cloud is. They actually talk about Infrastructure-as-a-Service and that’s what their report is based on. We think it is too narrow. We don’t think it’s a client view. Forrester just put out a new report and they positioned IBM as a leader, especially when clients are trying to do hybrid. Gartner tends to have a very narrow view, probably the most narrow view of the market that I’ve ever seen, but even in Gartner, if you look at some of their other Magic Quadrants, they position us extremely well. They excluded a lot of capability we have. They excluded our Platform-as-a-Service capability. They excluded our bare metal capability, which is a big differentiation for our cloud over other clouds. It’s something that enables us to do things that others cannot. If you look at our VMware partnership, a lot of that was enabled because we had the bare metal capability. We can do things for clients that some of the other public cloud providers cannot do. We think that Gartner took a very naïve view, a very small, narrow view of the market.