Archive for April, 2010

Public vs. Private Options in the Cloud

The demand for cloud computing is perpetually increasing, which means that business and technology managers need to clear up any questions they have about the differences between public and private clouds—and quickly at that.

The St. Louis-based United Seating and Mobility is one company that faced the common dilemma of choosing between a public or private cloud. The company—which sells specialized wheelchairs at 30 locations in 12 states—initially used phones and email to stay up to date on vendor contracts and other matters before monitoring these developments with off-the-shelf applications on its own servers. Finally, United Seating and Mobility decided to move to the public cloud.

United Seating and Mobility’s director of operations Michael DeHart tells Baseline Magazine of the move, “The off-the-shelf applications didn’t collaborate. You’d log on to all of the apps and try to remember which one needed which password.” Staffers across the nation now share the information seamlessly via the enhanced tools available in the public cloud.

Another example illustrating the difference between the public and private cloud is the Cleveland Cavaliers. The NBA team uses a private cloud to run its arena’s website. Going private allowed for increased one-on-one interaction with the cloud provider partner while simultaneously giving the franchise more resources to handle increased traffic to the site. Traffic on the area site has been known to spike when, for example, the team makes the playoffs or a major artist is coming to the venue. “When you’ve booked Miley Cyrus you’d better be ready,” says the Cleveland Cavaliers director of web services Jeff Lillibridge.

Despite choosing different versions of the cloud, both United Seating and Mobility and the Cleveland Cavaliers have noticed that few enterprise managers will be able to avoid the topic of private verses public clouds. According to research firm IDC, worldwide cloud services revenue will reach $44.2 billion in 2013, compared to $17.4 billion last year.

Business and technology professionals remain stumped about what private and public clouds are despite the increased demand for worldwide cloud services. Examples of public clouds include Google AppEngine, IBM’s Blue Cloud, LotusLive Engage and Amazon’s Elastic Compute Cloud (EC2). A public cloud is a shared technology resource used on an as-needed basis and available via the Internet while a private cloud is created specifically for the use of one organization.

Enhanced by virtualization technologies, both concepts are making way for an “evergreen” approach to IT in which enterprises can obtain technologies when they need them without purchasing and maintaining a host of in-house services.

Bob Zukis, national leader of IT strategy for PricewaterhouseCoopers (PwC) says, “It all stems from the legacy model of ‘build it and forget about it.’ Changes taking place in the industry are making it much more efficient and effective to provision what IT needs. So ‘build it and forget about it’ no longer meets the needs of the business. Whether you’re going with a public or private cloud, you’re pursuing a way to increase your technological resources in a more efficient flexible way.”

In addition to being evergreen, this movement is also green-friendly. Says Frost and Sullivan’s Vanessa Alvarez, “Cloud computing allows for resources and paying only for what they use. When an application is not utilizing resources, those resources can be moved to another application that needs them, enabling maximum resource efficiencies. If additional capacity or resources are no longer needed, virtual servers can be powered down or shut off.”

Organizations continue to struggle to choose between private versus public clouds. On one hand, private clouds offer security and increased flexibility compared to traditional legacy systems, but they have a higher barrier of entry compared to public clouds. In contrast, private cloud services require that an enterprise IT manager handle technology standardization, virtualization and operations automation in addition to operations support and business support systems.

“With public clouds, you provision your organization very quickly, by increasing service, storage and other computing needs, “says Zukis. “A private cloud takes a lot more time because you’re essentially rearchitecting your legacy environment.” Although public clouds don’t require this organizational shift and are thus faster and more convenient, they fail to provide the same amount of transparency as private clouds. Says Zukis, “It’s not always clear what you’re buying off the shelf with public clouds.”

Assessing the Value of Security

Another major issue in the cloud debate is security. All organizations value security but each has to decide between balance between cost and convenience, on one hand, and data security, on the other. Some organizations might have a higher threshold for potential violations than others and thus require a need-for-speed strategy.

Head of strategic sales and marketing at NIIT Technologies Aninda Bose, who has analyzed both cloud structures through her job and also in her position with nonprofit research organization Project Management Institute, states that the public cloud is the better option for an enterprise dealing with high-transaction/low-security or low data value. An example illustrating this is a local government office, which needs to tell a citizen that their car registration is up for renewal and simply needs to give the citizen a renewal date—a perfect situation for public cloud hosting.

Examples better suited for the private cloud model due to the sensitivity of their data include a federal agency, financial institution or health care provider. Mark White, principal with Deloitte Consulting, explains, “Accounting treatments and taxation applications are not yet fully tested for public cloud services. So enterprises with significant risk from information exposure may want to focus on the private cloud approach. This caution is most relevant for systems that process, manage and report key customer, financial or intelligence information. It’s less important for ‘edge’ systems, such as salesforce automation and Web order-entry applications.”

Sioux Falls, South Dakota-based medical-practice company The Orthopedic Institute is very data-dependent and concluded that the private cloud structure best fit its needs—specifically because the company must comply with strict rules for protecting patient information laid out by HIPAA (Health Insurance Portability and Accountability Act).

IT Director David Vrooman explains that The Orthopedic Institute was seeking to change it domain name from Orth-I.com but after exploring possibilities with the exclusive provider of .md domains MaxMD it determines that MaxMd could also provide private cloud services for highly secured, encrypted email transmissions. Moreover, the cost of entry was less than doing it in-house. “We didn’t want to use one of our servers for this because it would have amounted to a $20,000 startup cost. By going with a private cloud option, we launched this at one-fifth of that expense—and it only took an afternoon to get it started, ” says Vrooman. “It would have taken at least a week for my staff and me to get this done. And because MaxMD has taken over the email encryption, I’m not getting up at 3am to find out what’s wrong with the server.”

Some industry experts warn that traditional views about security and cloud computing may be changing, however, and that includes organizations which are dependent on highly secured data. CPA2Biz, the New York-based American Institute of Certified Public Accountants, wanted to provide its 350,000 members with access to the latest software tools for its business resources-providing subsidiary. CPA2Biz worked with Intacct to create a public cloud model for its CPA members. The program was launched in April and since then concerns have about security have been addressed and hundreds of firms are supporting approximately 2,000 clients through the public cloud services offered through CPA2Biz.

“Only those in the largest of member organizations would be able to consider a private cloud system. Plus, we don’t believe there are security advantages to a private cloud system,” says vice president of corporate alliances at CPA2Biz Michael Cerami. “We’ve selected partners who operate highly secure public cloud environments. This allows us to provide our members with great collaborative tools that enable them to work proactively with their clients in real time.”

The Choice

Going back to United Seating and Mobility, the organization was interested in the public cloud structure because it isn’t dependent on high-volume, automated sales. The company uses IMB’s LotusLive Engage for online meetings, file-sharing and project-management tasks.

DeHart estimates that it would have taken up a server and a half had it done this in house saying, “Being on the public cloud allows us to avoid this entirely. It’s a leasing-versus-owning concept—an operational expense versus a capital one. And the Software-as-a-Service offerings are better than what we could get off the shelf. We certainly can’t use this cloud to work with any sensitive health data. But we can run much of our business operations on it, freeing up our IT people to focus on email, uptime and cell phone services.”

Now, take the Cleveland Cavaliers. They opted for private cloud services to support the website for their venue, Quicken Loans Arena, aka “the Q.” Fans can search for information about upcoming events on TheQArena.com and are directed to a business called Veritix is they want to buy tickets. The arena site acts as a traffic conduit for Veritiix, thus a private cloud was the best option and the team partnered with Hosted Solutions. Since the current NBA season began last fall, the site’s page views and visits have seen an increase of over 60 percent and the number of unique visitors has increased by 55 percent. The team avoids uncertainly about who is minding the data by employing Hosted Solutions.

The private cloud also enables the team to manage site traffic that can jump significantly in the case of a last-second, playoff-determining shot, for example. “The need to scale was significant but we didn’t want to oversee our own dedicated hosting,” says Lillibridge. “It would have been more expensive, and we would have had the headache of managing our own servers. We needed dedicated services that would avoid this, while allowing our capacity to increase during peak times and decrease when we don’t have a lot of traffic.”

There is no clear cut answer for whether the private or public cloud is better, rather companies needs to assess their own individual requirements for sped, security, resources and scalability. To learn more about which Cloud option is right for your enterprise, contact a Nubifer representative today.

Advertisements

A Guide to Windows® Azure Platform Billing

Understanding billing for Windows® Azure Platform can be a bit daunting, so here is a brief guide, including useful definitions and explanations.

The Microsoft ® Online Customer Service Portal (MCOP) limits one Account Owner Windows Live ID (WLID) per MOCP account, and the Account Owner has the ability to create and manage subscriptions, view billing and usage data and specify the Service Administrator for each subscription. While this is convenient for smaller companies, large corporations may need to create multiple subscriptions in order to design an effective account structure that will able to support and also reflect their market strategy. Although the Service Administrator (Service Admin WLID) manages deployments, they cannot create subscriptions.

The Account Administrator can create one or more subscriptions for each individual MOCP account and for each subscription, the Account Administrator can specify a different WLID as the Service Administrator. It is also important to note that the Service Administrator WLID can be the same or different as the Account Owner and is the person actually using the Windows ® Azure Platform. Once a subscription is created in the Microsoft ® Online Customer Service Portal (MOPC), a Project appears in the Windows ® Azure portal.

The relationship between components is clearly displayed in the diagram below:

Projects:

Up to twenty services can be allocated by one project. Resources in the Project are shared between all of the Services created and the resources are divided into Compute Instances/Cores and Storage accounts.

The Project will have 20 Small Compute Instances that you can utilize, by default. These Small Compute Instances could be a variety of combinations of VM sizes as long as the total number of Cores across all deployed services within the Project doesn’t exceed 20.

To increase the number of Cores, simply contact Microsoft ® Online Services customer support to verify the billing account and provide the requested Small Compute Instances/Cores (subject to a possible credit check). You also have the ability to design how you want to have the Cores allocated, although be default the available resources are counted as number of Small Compute Instances. See the conversion on Compute Instances below:

Compute Instance Size CPU Memory Instance Storage
Small 1.6 GHz 1.75 GB 225 GB
Medium 2 x 1.6 GHz 3.5 GB 490 GB
Large 4 x 1.6 GHz 7 GB 1,000 GB
Extra large 8 x 1.6 GHz 14 GB 2,040 GB

Table 1: Compute Instances Comparison

The compute Instances are shared between all the running services in the project—including Production and Stage Environments. This allows you to have multiple Services with different number of Compute Instances (up to the number of maximum available for that Project).

5 Storage accounts are available per Project, although you can request to increase this up to 20 Storage accounts per Project by contacting Microsoft ® Online Services customer support. You will need to purchase a new subscription if you need more than 20 Storage accounts.

Services:

A total of 20 Services per project are permitted. Services are where applications are deployed; each Service provides two environments: Production and Staging. This is visible when you create a service in the Windows ® Azure portal.

A maximum number of five roles per application are permitted within a Service; this includes any combinations of different web and worker roles on the same configuration file up to a maximum of 5. Each role can have any number of VMS, see below:

The Service has two roles in this example, and each role has a specific worker role. Web Role, web tier, handles the Web interface, while the Worker Role, business tier, handles the business logic. Each role can have any number of VMs/Cores up to the maximum available on the project.

If this service is deployed from the Azure ® resources perspective, the following resources will be used:

1 x Service

–       Web Role = 3 Small Compute Nodes (3 x Small VMs)

–       Worker Role = 4 Small Compute Nodes (2 x Medium VMs)

–       2 Roles used

Total resources left on the Project:

–       Services (20 -1) = 19

–       Small Compute Nodes (20 – 7) = 13 small compute instances

–       Storage accounts = 5

For more information regarding the Windows Azure pricing model, please contact a Nubifer representative.

Amazon’s Elastic Compute Cloud Platform EC2 Gets Windows Server Customers from Microsoft

Amazon has launched an initiative for Microsoft customers to bring their Windows Server licenses to Amazons EC2, Elastic Compute Cloud Platform. This initiative is in tandem with a brand new Microsoft pilot program which allows Windows Server customers with an EA (Enterprise Agreement) with Microsoft to bring their licenses to Amazon EC2. Peter DeSantis, general manager of EC2 at Amazon, said in a recent interview with eWEEK that these customers will pay Amazon’s Linux On-Demand or Reserved Instance rates and thus save between 35 to 50 percent, depending on the type of customer and instance.

Also in his interview with eWEEK, DeSantis said that Amazon customers have sought support for Windows Server and Amazon has delivered support for Windows Server 2003 and Windows Server 2008. Customers with EA agreements with Microsoft began to ask if those agreements could be applied to EC2 instances, thus the new pilot program. Amazon announced the new initiative on March 24 and began enrolling customers instantaneously. According to DeSantis, enrollment will continue through September 12, 2010.

Amazon sent out a notice announcing the program and stated the following criteria as requirements laid out by Microsoft to participate in the pilot: your company must be based or have legal entity in the United States; your company must have an existing Microsoft Enterprise Agreement that doesn’t expire within 12 months of your entry into the Pilot; you must already have purchased Software Assurance from Microsoft for your EA Windows Server licenses; you must be an Enterprise customer (this does not include Academic Government institutions).

eWEEK revealed some of the fine print for the project released by Amazon:

“Once enrolled, you can move your Enterprise Agreement Windows Server Standard, Windows Server Enterprise, or Windows Server Datacenter edition licenses to Amazon EC2 for 1 year. Each of your Windows Server Standard licenses will let you launch one EC2 instance. Each of your Windows Server Enterprise or Windows Server Datacenter licenses will let you launch up to four EC2 instances. In either case, you can use any of the EC2 instance types. The licenses you bring to EC2 can only be moved between EC2 and your on-premises machines every 90 days. You can use your licenses in the US East (Northern Virginia) or US West (Northern California) Regions. You will still be responsible for maintaining your Client Access Licenses and External Connector licenses appropriately.” To learn more about Microsoft’s and Amazon’s Cloud offerings visit Nubifer.com.

Legal Risks for Companies to Consider Before Embracing the Cloud

Along with its never-ending stream of possibilities in revolutionizing the invention, development, deployment, scale, updating, maintenance and payment for data and applications, cloud computing brings a variety of legal risks to the table, and companies must consider these before entering a highly optimized public cloud.

Risk from uncertainty over where sensitive data and applications physically dwell arises from what Baselinemag.com calls the “nationless state” of the public cloud. Among these ricks are jurisdictions where laws governing the protection and availability of data are very different than what companies are used to. Information in the cloud can also be widely distributed across various legal and international jurisdictions (which each have different laws concerning security, privacy, data theft, data loss and intellectual property) due to the virtual and dynamic nature of cloud computing architecture.

Furthermore, when operating in the cloud, issues concerning privacy, data ownership and access to data cause many questions to arise. National or international legal precedents for cloud computing may be few and far between, but companies nonetheless must ensure that they can immediately access their information and that their service provider has appropriate backup and data-retrieval procedures in place.

A new paradigm of licensing—in which traditional software license agreements will be replaced with cloud service agreements—will be replaced with cloud service agreements as a result of the legal framework of cloud computing. Lawyers representing cloud service providers will subsequently try to reduce the liability of their clients by proposing contracts with the service provided “as is” without a warranty. Under this new paradigm, the service is provided without any assurance or promise of a specific level of performance. This added rick must be evaluated within the context of the benefits derived from the cloud as well as the proposed data which will be stored in the cloud.

Cloud computing also causes issues for companies that have to meet increasingly stringent compliance and reporting requirements for the management of their data. These issues pose major risks in protecting companies’ sensitive data and the information assets their customers have entrusted them to watch over.

In summary, enterprises must make sure that their cloud service providers specify where their data dwells, the legal framework within those specific jurisdictions and the security, backup, anti-hacking and anti-viral processes the service provider has set up. Despite these risks, cloud computing has enormous benefits which should make companies eager to take advantage of its optimization, scalability and cost savings that cloud computing provides. While embracing the cloud, companies must simply conduct a more detailed legal analysis and assessment of risks, much like they would with traditional IT services. For more information on security relating to Cloud Computing, please visit Nubifer.com.

Microsoft Not Willing to Get Left in the Dust Left by Cloud Services Business

Microsoft may be the largest software company on the globe, but that didn’t stop it from being left in the dust by other companies more than once and eWEEK reports that when it comes to cloud services Microsoft is not willing to make the same mistake.

Although Microsoft was initially weary of the cloud, the company is now singing a different tune and trying to move further into the data center. Microsoft had its first booth dedicated solely to business cloud services at the SaaSCon 2010, held at the Santa Clara Convention Center April 6 and 7. Microsoft is positioning Exchange Online (email), SharePoint Online (collaboration), Dynamics CRM Online (business apps), SQL Azure (structured storage) and AD/Live ID (Active Directory assess) as its lead services for business. All of these services are designed to run on Windows Server 2008 in the data center and sync up with the corresponding on-premises applications.

The services are designed to work hand-in-hand with standard Microsoft client software (including Windows 7, Windows Phone, Office and Office Mobile), thus ensuring that the overarching strategy is set and users will have to report on its cohesiveness over time. Microsoft is also offering its own data centers and its own version of Infrastructure-as-a-Service for hosting client enterprises apps and services. Microsoft is using Azure—a full online stack comprised of Windows 7, the SQL database and additional Web services—as a Platform-as-a-Service for developers.

Featuring Business Productivity Online Suite, Exchange Hosted Services, Microsoft Dynamics CRM Online and MS Office Web Apps, Microsoft Online Services are up and running. In mid-March Microsoft launched a cloud backup service on the consumer side called SkyDrive, which is an online storage repository for files which users can access from anywhere via the Web. SkyDrive may be a very popular service, as it offers a neat (in both senses of the word) 25GB of online space for free (which is more than the 2GB offered as a motivator by other services).

SkyDrive simply requires a Windows Live account (also free) and shows that Microsoft really is taking the plunge. For more information on Microsoft’s Cloud offerings, please visit Nubifer.com.

ERP and CRM Integration Via Business Intelligence for the Cloud

The masterminds behind Crystal Reports are unveiling a new business intelligence cloud offering being sold through channel partners. Not only do solution providers get an ongoing annuity on the sale, but they can perform the integration work to link the cloud-based BI to the data source (whichever ERP/CRM solution it is, such as Oracle, Salesforce.com, SAP or something else).

Traditional VARs gaging the potential of the cloud business model may have a difficult time seeing how much money per user per month will be enough for a business to reap the benefit of the cloud. Indicee executives Mark Cunningham, CEP, and Craig Todd, director of partnerships, understand the businesses are accustomed to the big sale upfront and ongoing services after that sale. Cunningham and Todd were both part of the team that created the Crystal Reports business intelligence software–which sold to Seagate before becoming part of SAP–and decided to bring their technology expertise into the cloud.

Although Cunningham and Todd knew that business was moving into the cloud and that their expertise had revealed that channel partners are the ideal way to connect with end customers, they just didn’t know how to merge those two ideas. Said Todd to Channel Insider, “The biggest single difference in what SaaS is removes those boxes. It has initially been seen as a threat by some of our partners.”

“A lot of VARs are worried about being disintermediated. Their expertise in installing software is no longer required.But the ones we’ve been working with the last few months see it as an opportunity,” continued Todd.

Arxis Technology in Simi Valley, California, an ERP, CRM and BI specialist, is one such partner. The 25-person company has two offices in California as well as offices in Chicago and Phoenix. Director of sales and marketing Mark Severance told Channel Insider that whether the customer is deploying on-premises solutions or in the cloud solutions the revenue comes out even. “The biggest thing people are having a hard time with is that you are used to the big upfront sale. But, honestly, from our perspective, if you have great products and do a great job taking care of the customer, then there’s a business model for that you do,” explains Severance.

Severance said that the annuity part of the business (in which Arxis receives a commission per user per month on an ongoing basis) will eventually make up for the lack of large upfront sale. Additionally, Arxis can offer the integration and implementation services which customers need, which means setting up the BI solution’s data sources, whether they may be Salesforce.com or an internal CRM or ERP solution.

Arxis continues to offer traditional on-premises CRM and ERP software sales and implementation; the biggest vendor Arxis works with currently is Sage. Arxis offers a BI solution from Business Objects in on-premises and cloud form and recently added Indicee’s cloud-based BI solution for a variety of reasons. One major reason is that some customers are unable to afford an on-premises-based BI solution and thus a cloud-based solution is more economically accessible.

Severance further pointed out that most of computing is making the transition into the cloud. While companies used to feel safe having their server in-house, they now want to be able to access there data whenever, wherever they are, from whichever device they are using.

Indicee’s Cunningham and Todd also pointed out that VARs can provide their end customers with training services as well as services like change management. Said Todd, “There’s an exciting opportunity here for traditional VARs. This creates a platform that allows partners to focus on the V and A in the VAR–the value add.”

Pricing at Indicee starts at $69 per user per month, with a five-user pack priced at $150 per month. The VAR cut generally is a 20 percent commission on sales of five packs or more, calculated monthly and paid out quarterly, but Todd noted that it is dependent on how much work the VAR is completing to get the customer.

Gartner predicts sales of $150 million by 2013. Cunningham notes that SaaS is poised for growth and that if solution providers are seeking to enter the cloud, business intelligence is a lucrative starting point, even with its required integrative work. To learn more about CRM Applications in the Cloud, please visit Nubifer.com.

The Role of Multitenancy in the Cloud

The debate over whether or not multitenancy is a prerequisite for cloud computing wages on. While those pondering the use of cloud apps might think they are removed from this debate, they might want to think again, because multitenancy is the clearest path to getting more from a cloud app while spending less.

Those in the multitenancy camp, so to say, point out that there is only a slight only difference between two subscription-based cloud apps is that one is multitenant and the other is single-tenant. The multitenant option will offer more value over time while lowering a customer’s costs and the higher degree of multitenancy—i.e. the more a cloud provider’s infrastructure and resources are shared—the lower the customer cost.

At the root of the debate is revenue and cost economics of cloud services. Revenues for most cloud app providers come from selling monthly or annual per-seat subscriptions. These bring in just a portion of the annual revenue that would be generated by an on-premise software license with comparable functionality. The challenge for selling software subscriptions comes from reducing operating costs to be able to manage with less. If this is not achieved, the provider may have to do more than an on-premise vendor does—like run multiple infrastructures, maintain multiple versions, perform upgrades and maintain customer-specific code—with less money. The answer to this conundrum is multitenancy. Multitenancy extends the cost of infrastructure and labor across the customer base. Customers sharing resources down to the database schema is perfect for scaling.

As the provider adds customers, and those customers benefit from this scaling up, the economies of scale improve. The cloud app provider is able to grow and innovate more as costs decrease and in turn value increases. Over time customers can expect to see more value (like in the form of increased functionality), even if costs don’t lower. For more information of Multitenancy, visit Nubifer.com.