OakLeaf Systems: Windows Azure and Cloud Computing Posts for 7/20/2009+

Sunday, July 26, 2009

Windows Azure and Cloud Computing Posts for 7/20/2009+

Windows Azure, Azure Data Services, SQL Azure Database and related cloud computing topics now appear in this weekly series.

••• Updated 7/25/2009: Minor updates and additions
•• Updated 7/23 and 7/24/2009: Correction to Azure instance cost, Azure ROI Calculator, Google Apps Sync, other additions
• Updated 7/21 and 7/22/2009 9:30 AM PDT: Additions
Updated 7/20/2009 2:30 PM PDT: New Azure Toolkit and SDK July 2009 CTP (see Infra)

Note: This post is updated daily or more frequently, depending on the availability of new articles in the following sections:

Azure Blob, Table and Queue Services
SQL Azure Database (SADB)
.NET Services: Access Control, Service Bus and Workflow
Live Windows Azure Apps, Tools and Test Harnesses
Azure Infrastructure
Cloud Security and Governance
Cloud Computing Events
Other Cloud Computing Platforms and Services

To use these links, click the post title to display the single article you want to navigate.

Azure Blob, Table and Queue Services

• Eric Florenzano reviews the following (mostly key-value) alternatives to SQL RDBMSs in his My Thoughts on NoSQL post of 7/21/2009:

Tokyo Cabinet / Tokyo Tyrant
CouchDB
Redis
Cassandra

Steve Lesem’s Cloud Storage and The Innovator's Dilemma 7/19/2009 essay begins:

Too many think of cloud storage as just another or the next type of storage. As usual with this view, it is associated with a view that the "next" storage type is bigger, faster and cheaper. Because each generation of storage is always bigger, faster and cheaper. As such, proponents of this view generally believe that access via traditional approaches, like WebDAV, NFS, cifs and others, is a critical capability. Some may even argue that Web Services APIs are not the critical differentiation of Cloud Storage. We disagree.

Cloud storage is a radical change. It enables new application types. The critical capability for cloud storage is a Web services API access, revealing the full promise of SOA (Service Oriented Architecture). Second, the services that are revealed by the API access go far beyond "put" and "get". Anytime and anywhere access, tagging, sharing and collaboration, geo storage via a single namespace, and policy management of storage are some of the services that the new applications will expect to find in the storage clouds they chose. Also, storing massive amounts of data in the cloud and having these services available to act on all the data is required.

SQL Azure Database (SADB, formerly SDS and SSDS)

••• James Hamilton’s HadoopDB: MapReduce over Relational Data post of 7/25/2009 contends that Hadoop “could run over a full relational database”:

MapReduce has created some excitement in the relational database community. Dave Dewitt and Michael Stonebraker’s MapReduce: A Major Step Backwards is perhaps the best example. In that posting they argued that map reduce is a poor structured storage technology, the execution engine doesn’t include many of the advances found in modern, parallel RDBMS execution engines, it’s not novel, and its missing features.

In Mapreduce: A Minor Step Forward I argued that MapReduce is an execution model rather than storage engine. It is true that it is typically run over a file system like GFS or HDFS or simple structured storage system like BigTable or Hbase. But, it could be run over a full relational database. …

• John Willis passes along RightScale first to support IBM DB2 [Express] in the cloud (Press Release) on 7/22/2009 with the full text of the release, starting:

RightScale, Inc., the leader in cloud computing management, today announced that the RightScale Cloud Management Platform is the first solution that lets users create, manage and automate IBM DB2 Express-C 9.7 database software on the cloud. Now, RightScale users can more easily build, test and deploy applications on leading clouds such as Amazon EC2 using the most advanced version of IBM DB2 directly from the RightScale platform. …

• Markus Klems tackles Relaxed Consistency in this 7/22/2009 essay with links to additional references:

Since Werner Vogels’ famous blog post & paper about eventual consistency (aka relaxed consistency) some time has passed. Relaxed consistency is a concept related to the complexity of distributed systems. Since it can take some time for a read or write operation to be properly validated and passed through the various layers of a distributed system, the question arises, “how does this affect user experience?” Clearly, not in a good way.

Werner Vogels explains that for certain applications it might be necessary to tolerate a certain amount of system inconsistency in order to speed up the user-perceived time it takes to complete an operation (such as adding an item to the famous shopping cart). The concept is somewhat similar to Ajax, however, Ajax happens on the presentation layer whereas relaxed consistency happens somewhere in a back-end layer of the software stack. …

http://www.allthingsdistributed.com/2007/12/eventually_consistent.html

http://www.allthingsdistributed.com/2007/10/amazons_dynamo.html

http://research.yahoo.com/files/pnuts.pdf

Fortunately, SADB doesn’t suffer from the problems associated with relaxed or eventual consistency.

.NET Services: Access Control, Service Bus and Workflow

• Peter G. Jones’ Azure .Net Service Bus & TCP post of 7/22/2009 provides a real-world application for the .NET Service Bus:

We recently implemented the .Net Service bus to expose some in-house WCF services to the world wide world. It may be useful for me and others if I describe how to do this :) This setup allows you to switch between TCP and HTTP relay binding with configuration.

For the in-house systems you need to create a host – you can’t use IIS as the service bus requires a connection to be initiated by both ends of the communication. We created a Windows Service application that also runs as a console app. This is much simpler when developing and debugging.

Richard Seroter explains Sending Messages From Azure Service Bus to BizTalk Server 2009 in this 7/19/2009 post:

In my last post, I looked at how BizTalk Server 2009 could send messages to the Azure .NET Services Service Bus. It’s only logical that I would also try and demonstrate integration in the other direction: can I send a message to a BizTalk receive location through the cloud service bus?

Richard’s preceding post is Securely Calling Azure Service Bus From BizTalk Server 2009 of 7/12/2009.

Live Windows Azure Apps, Tools and Test Harnesses

•• Davide Zordan shows you how to deploy David J. Kelly’s Simon Silverlight 3 game to Azure in his Simon in the cloud: deploying your existing Silverlight application to the Windows Azure platform. You can run a demo of the ported project here.

•• David Pallman’s Azure ROI Calculator post of 7/22/2009 announces:

… [T]he availability of Neudesic's Azure ROI Calculator, available online at http://azureroi.cloudapp.net. This is a beta tool we are soliciting feedback on.

The ROI Calculator lets you estimate what your monthly cloud computing charges will be on the Azure Services Platform. It also helps you determine what your return on investment will be if you migrate an application over to the cloud. …

•• Om Malik and Michael Cote chose Omar Del Rio’s Personal Radio Station as the international winner of newCloudApp() - The Azure™ Services Platform Developer Challenge on 7/22/2009. According to the help file:

With Personal Radio Station you can share your music with your friends, recommend music and post comments about the music you are listening to.

It is as simple as signing into your Windows Live account and your Mesh account (Your Mesh account will work as an storage drive for your favorite songs).

Jim Nakashima announces the availability of Windows Azure Tools and SDK July 2009 CTP on 7/20/2009. VS 2008 gets a new dialog for creating Cloud Services with multiple WebRoles and WorkerRoles, as well as Associating an ASP.NET Web Application (including MVC) as a Web Role. For more details, see Azure Infrastructure. Here’s Jim’s example:

Azure Infrastructure

••• Paul Krill’s Spring may be coming soon for Microsoft's Azure Computerworld post of 7/24/2009 speculates:

Java developers fond of the popular Spring Framework might gain an interesting deployment option -- the Microsoft Windows Azure cloud platform -- if an alliance can emerge between Spring proponents and Microsoft.

And given Microsoft's newly stated intention to possibly align with Spring advocates, it looks like such an option could be in the works.

•• David Chappell replies to my mischaracterization of his cloud presentation in a Comment to my A Comparison of Azure and Google App Engine Pricing post of 7/19/2009:

Just to be clear, the presentation linked to above (it's not a white paper) wasn't sponsored by Microsoft. Rather, it was a keynote I wrote for a Silicon Valley cloud computing conference, and it expresses my personal perspective. (In fact, this is true of everything I write, regardless of the sponsor.)

My goal in that talk wasn't to show which platforms were richer than others. Instead, I was trying to find a more useful framework for comparison than IaaS/PaaS, which I really do think is too simplistic. For example, AWS has some aspects of PaaS, such as SimpleDB and SQS. And depending on which dimension you look at, Windows Azure can look like either IaaS or PaaS. It's priced like IaaS--by instance hour--but you give it an application to run rather than explicitly creating a VM, like PaaS. The simple IaaS/PaaS breakdown once made sense, but it's no longer enough.

This is a big part of why I think scenario-based comparisons can sometimes be more useful. Still, things have changed since I wrote that talk: AppEngine now supports more of my scenarios with its addition of tasks and the Task Queue API. If I were to update the comparison, AppEngine would today cover pretty much the same terrain as Windows Azure.

•• David Linthicum’s Why on-premise SOA technology vendors should move to cloud computing pricing post of 7/24/2009 challenges vendors to “Move to service-based pricing, I dare you!”

Let's face it, there are no guarantees around the success of SOA technology within the enterprise. We pay our million-plus dollars for huge license agreements, prior to one server going online, and then find out after implementation that the technology has fallen short and the project [is] a failure. Sound familiar?

I don't put this on the SOA technology vendors, by the way. In many instances, the SOA architects just did not ask the right questions and testing was not done; when you do that, you get what you deserve. However, if you're looking at this from the SOA architect's point of view, they would say that they have to expend the money when the budget dictates and make specific assumptions that the technology will work as advertised.

But make sure the CPU pricing emulates Google’s usage-based formula for App Engine, not Azure’s fixed daily charge, regardless of user activity.

•• David Worthington reports that Burton research director Anne Thomas Manes recommends not using unmanaged code in Azure projects in his Analyst: Beware of unmanaged code in Windows Azure article of 7/24/2009 for Software Development Times:

The March CTP of Azure introduced .NET Full Trust, a method for developers to call into unmanaged COM DLLs; Platform Invocation Services, which enables managed code to call native code; and PHP support.

Anne Thomas Manes, research director with the Burton Group, said that Microsoft has three primary motivations for including unmanaged code: Certain aspects of Windows Communication Foundation (WCF) and Windows Workflow Foundation may require Full Trust, including Microsoft Message Queuing, a channel that is used in WCF; PHP is popular for new development; and COM support is necessary for hosting legacy applications.

… Manes cautioned that developers should have a strong business case for using COM in the cloud.

"COM development has a legacy of viruses and is unsafe in the cloud. Azure is a target of hackers. People should not write COM applications on Azure," she added.

•• C. Burns and B. Guptil co-authored Saugatuck Research’s MIT Cloud Computing Forums: Organizational Politics Cloud[s] Adoption Research Alert of 7/22/2009 (site registration required).

From June 11 through June 18, Saugatuck Research VP Charlie Burns took part in four expert panel and networking reception events hosted by the MIT Technology Review, that examined the realities of Cloud Computing, and their effects on user business and IT strategy, planning and management (See Note 1, and Research Alert MIT Cloud Computing Forums: Executives Don’t Know What They Don’t Know, RA-610, published 24Jun09).

[I]n this Research Alert, we highlight one important question that was asked of the event attendee’s: What is single greatest challenge to the adoption of Cloud Computing within their own firms? Interestingly, 40 percent of the 149 respondents indicated “organization/politics.” The next-ranked challenges were “legacy assets” (31 percent) and “workload migration” (29 percent).

John Gauntt will author a Mobile Cloud Computing Project for GigaOm Pro that’s expected to publish after Labor Day, according to this 7/22/2009 post:

I’ll be canvassing both the mobile and cloud computing sectors for data and insight.We’re still knocking around the outline which should be sorted by next week. Suffice to say that it’s going to pivot between an app-centric and service-centric view of Act II of the Mobile Internet.

Ping me with ideas and contacts if you’ve got a strong view on mobile cloud computing.

•• Nikhil Chinchwade started a new thread on 7/22/2009 in the Azure forum with this question: July CTP feature (Multiple web role and worker role). Will it be deployed on same VM or different VM's?. The official answer from Microsoft’s Li-Lun Luo was:

[E]ach instance will be deployed on a separate VM. So of course different roles will run on different VMs.

Nikhil then asked:

I mean, if I have configured two web roles under same cloud service (per the new feature provided in July CTP) with 1 instance each, would it use 2 VM's or 1?

Aleks Gershaft of the Azure team clarified Li-Lun’s use of the term instance to mean *Role instance:

It will be two VMs. At the current time, each role instance is given a separate VM. [Emphasis added.]

Which means US$0.12 per hour, $2.88 per day, or $86.40 per month per Role instance after RTW. I found this hard to swallow in an environment as competitive as cloud computing platforms. (7/23/2009: Compute cost corrected from $0.15 per hour in original post.)

As developer Pita.O points out:

The whole idea of the feature request for multiple roles per instance was for us to use one VM efficiently. …

That was my understanding also. Hopefully, the Azure team will enable by RTW all role instances to run on one VM when traffic is light.

• Elizabeth Montalbano’s Company Helps ISVs Assess Cost of Microsoft's Windows Azure review of 7/22/2009 for PC World describes PreEmptive’s new DotFuscator release, which:

[I]ncludes a way for ISVs (independent software vendors) to monitor an application not only to find out how many computing resources it requires when running on Windows Azure, but also to find out how people are using it, said Sebastian Holst, the chief marketing officer at PreEmptive, based in Mayfield Village, Ohio. …

PreEmptive charges US$12,000 for a development group to use the full suite, and then if users want to subscribe to a PreEmptive-hosted dashboard to view the data, it's $2,000 per user per year, Holst said.

A bit pricey, n’est-ce pas?

• James Watters takes NY Times editorial contributor and Harvard law professor Jonathan Zittrain to the woodshed in a scathing NYT Kicks Off Cloud Paranoia Series response of 7/21/2009 to Zittrain’s Lost in the Cloud op-ed piece of 7/19/2009. Zittrain claims the cloud is a threat to innovation. As Watters observes, Zittrain confuses Facebook and Apple’s AppStore with cloud computing:

Innovation will be alive and well because the fundamental technologies at the core of cloud computing are designed for massive, vibrant, explosive, awesome, and amazing application innovation. There will always be a big place in the market for companies who achieve design simplicity by limiting what can be done on their platforms—Apple and Facebook may march to massive market share by this principle—but as long as the technologies underpinning the network are open, programmable, extensible, modular, and dynamic as they are and will be, innovation is in good hands.

Ask any new cloud platform what they really want in abundance—it’s developers. As always, as in the desktop era, you win them by giving them the best place to innovate and create.

I agree with Watters that Zittrain’s essay completely missed the mark.

• Gordon Haff analyzes the trade-off between local and cloud-based computing in his Moore's Law vs. the Cloud article of 7/21/2009 for CNet News:

… [A]lthough cloud computing tracks improvements in networks, it doesn't necessarily sync up so cleanly with the parallel improvements going on in computers themselves. As a commenter put it in a recent post of mine: "The thing that I don't understand about the move to "cloud-based services" is that it seems at odds with Moore's Law. Specifically, devices are going to have more & more processing power, disk space & memory - why would you want to offload processing to the cloud?" …

• Simon Davies’ Windows Azure SDK July SDK Powershell Role Sample of 7/21/2009 describes:

[A] new sample project in the [Azure July 2009] SDK called PowerShellRole , this sample highlights the fact that Powershell is available - and usable – from Windows Azure. There is a sample web role that enables Powershell commands to be executed, the following screenshot shows using the Clouddrive Powershell provider to list the contents of a blob store:

Here’s the full What’s New for the July 2009 CTP?

Support for developing and deploying services containing multiple web and worker roles. A service may contain zero or more web roles and zero or more worker roles with a minimum of one role of either type.
New project creation dialog that supports creating Cloud Services with multiple web and worker roles.
Ability to associate any ASP.NET Web Application project in a Cloud Service solution as a Web Role
Support for building Cloud Services from TFS Build
Enhanced robustness and stability

Phil Wainewright questions whether Microsoft is “really aiming to become the leading cloud provider in those spheres?” in his Assessing Microsoft's Cloud Intentions post of 7/20/2009 to ebizQ:

… I'm sure the Azure and Online Services teams are ambitious for their own products, but Microsoft as a whole has to balance their aspirations with those of other business units, many of which bring in much larger revenues. No such considerations hold back Amazon in formulating the business strategy for its cloud computing services, or Google as it promotes its Google Apps offering to enterprises. It's pretty clear what these companies are trying to achieve. Whereas Microsoft's intentions are always liable to ambivalence, compromised by the demands of other, more established product lines. …

Maria Spinola’s White Paper: An Essential Guide to Possibilities and Risks of Cloud Computing of 7/19/2009:

The goal of this White Paper is to provide a realistic perspective of the possibilities, benefits and risks of Cloud Computing; what to look for, what to avoid, and also some tips and best practices on implementation, architecture and vendor management strategies. It is important to consider all those aspects before you decide either to move (but without putting the carriage before the horse) or not to move your systems, applications, and/or data to to the “Cloud”, in a “hype free” approach.

Click here to download the White Paper

Eric Lai claims “One integrator who uses rival cloud services questions Microsoft's math” in his Windows Azure will outcompete Amazon Web Services on features, total cost post of 7/15/2009 to InfoWorld’s CloudComputing blog:

While The Register declared Azure to be cheaper than Amazon.com's price for hosted Windows and more expensive than a Linux instance, Silicon Valley Insider called Azure's price "not significantly different than either Google or Amazon."

"The actual per-unit pricing is totally uninteresting in my mind," Prashant Ketkar, director of marketing for Windows Azure, told Computerworld on Tuesday. "What will it cost me end-to-end?"

Ketkar says that Azure offers a number of standard features that, if purchased as add-ons for most competing platforms, cause their prices "to be substantially more expensive than us."

He cited Azure's automated service management as a "killer feature" that enables apps on a downed server to be reloaded onto another server with minimal interruption using Microsoft's "fabric controller".

Outcompete seems a stretch to me.

Elizabeth Montalbano reports Microsoft: Azure enterprise licenses will be simple on 7/17/2009 for ComputerWorld, but the following argues against her premise:

In particular, pricing for its hosted Business Productivity Online Suite (BPOS) -- which includes hosted versions of Exchange, SharePoint, LiveMeeting and Office Communications -- is causing customers some concern, said Paul DeGroot, an analyst with research firm Directions on Microsoft.

If a customer purchases a BPOS subscription for employees who will access only those services, the customer must still purchase CALs for those users, DeGroot said, even though they are not accessing the on-premise software as well. Microsoft gives customers a discount on other parts of their license in such scenarios -- on the Software Assurance (SA) maintenance program required for enterprise agreements, for example -- but they still end up paying for something they are not using, DeGroot said.

Depending on how it wants to give companies access to Azure beyond the pay-as-you-go pricing model, the company could run into the same trouble with its cloud-computing platform, he said. "With Azure it could get even more complicated," DeGroot said, though it remains to be seen until Microsoft unveils specific terms of Azure's integration into enterprise contracts.

Emma Stewart and John Kennedy offer their environmental The Sustainability Potential of Cloud Computing: Smarter Design analysis of 7/20/2009, which starts with this premise:

[F]or most financial, software and service sector companies, data centers are a major – and growing source – of greenhouse gas emissions. In 2006, the DOE estimates that U.S. data centers used 61 billion kWh of electricity, representing 1.5 percent of all U.S. electricity use, or the amount used by about 6 million US houses.

Emma and John work for Autodesk, Inc.

Cloud Security and Governance

•• Chris Hoff (@Beaker) proposes “embed[ding] a standardized and open API layer [of automated audit and security management capability] into each IaaS, PaaS and SaaS offering” in his Extending the Concept: A Security API for Cloud Stacks post of 7/24/2009:

This way you win two ways: automated audit and security management capability for the customer/consumer and a a streamlined, cost effective, and responsive way of automating the validation of said controls in relation to compliance, SLA and legal requirements for service providers. [Emphasis Beaker’s.]

Since we just saw a story today titled “Feds May Come Up With Cloud Security Standards” — why not use one they already have in SCAP to suggest we leverage it to get even better bang for the buck from a security perspective. This concept extends well beyond the Public sector and it doesn’t have to be SCAP, but it seems like a good example.

Of course we would engineer in authentication/authorization to interface via the APIs and then you could essentially get ISV’s who already support things like SCAP, etc. to provide the capability in their offerings — physical or virtual — to enable it.

Microsoft is promoting automated operation and has SAS 70 attestation and ISO/IEC 27001:2005 certification, so it sounds to me like a good model for the Azure Services Platform.

••• Beaker added the following update to the preceding post later in the afternoon:

Update: Wow, did this ever stir up an amazing set of commentary on Twitter. No hash tag, unfortunately, but comments from all angles. Most of the SecTwits dropped into “fire in the hole” mode, but it’s understandable. Thank you @rybolov (who was there when I presented this to the gub’mint and @shrdlu who was the voice of, gulp, reason.

•• Tim Green asserts Feds may come up with cloud security standards in this 7/24/2009 NetworkWorld Cloud Security Alert:

The federal government may step up with a set of cloud-security standards to meet government requirements for protecting sensitive data.

Federal CIO Vivek Kundra says he wants to certify cloud services that pass government muster so federal agencies can buy the computing or applications services they need and turn them on quickly. …

That requires establishing standards that officially meet 2002 Federal Information Security Management Act requirements that federal IT infrastructure must comply with. …

@Beaker tweets “Interesting given my discussions yesterday.” Was Chris referring to his anti-panelist duties for the Open Group’s The Cloud Security Debate: Is Cloud Computing More or Less Secure than Traditional In-house IT? discussion. In case you’re wondering, @Beaker’s official job title is Director of Cloud and Virtualization Solutions, Cisco.

•• John Pescatore’s Financial Friday: The Cost of a Security Incident Is Usually Much Greater Than Preventing It post of 7/24/2009 analyzes the total cost of a recent “loss of unencrypted CDs that contained on the order of 180,000 [HKSB] customer records” and concludes that it was US$20 million, without including soft costs, such as loss of business.

John is a Gartner analyst who specializes in IT security matters.

•• Stephen Foskett addresses the issue of business continuity planning in his Can You Leverage Cloud Services For Disaster Recovery? post of 7/24/2009, which asserts: “Cloud resources are inherently flexible, giving needed capacity on demand.” Another post with the same title claims that “The current economic climate is pushing many to look for current-year budget savings.”

•• Alan Wilensky’s Cloud insanity – the Shills come out of the woodwork post of 5/30/2009 (and just appeared in my feed reader today) takes on:

[A] shill for the cloud industry [who] says, in so many words, that the time to question the cloud hosted apps is over, they are established and able to deliver, and that self styled analysts, like me, have NO BID-NESS asking what if the service goes down, whaaaaaa! Self hosted solutions go down. And then commenter Russell says one of the most amazingly naive things I have ever seen in print, maybe in my entire life”:

“Many of the PaaS providers are in business with deep pockets (Force and Quickbase), well funded by professional investors (Bungee Labs), running with established management teams (Quickbase), or conservatively managed with established customer bases (WorkXpress).”

See the actual [comment] thread [on Jane McCarty’s blog] here.

• Chris Hoff (@Beaker) points out several cloud-security papers (including his) from the USENIX HotCloud 09 conference held in San Diego June 15, 2009 in his Tons Of Interesting Papers/Presentations From Usenix/HotCloud ‘09 post of 7/21/2009. Beaker’s post also includes many links to his own recent papers, articles, etc.

• The “Geneva” Team Blog’s 7/21/2009 Official Name for "Geneva" repeats the preceding week’s announcement and introduces Microsoft’s

Business Ready Security strategy to help both partners and customers 1) protect everywhere and access anywhere, 2) integrate and extend security across the enterprise, and 3) simplify the security experience and manage compliance.

For more information about these announcements and others that were made, check out the Microsoft Forefront Team Blog.

• Tom McHale asks Does Every Cloud Have a Silver Compliance Lining? in his 7/21/2009 post to CA’s Governance, Risk and Compliance blog:

While CIOs are considering cloud computing as a potential hedge for future IT infrastructure investments, the various compliance teams are saying/thinking: “You want to put what, where?” I have had discussions with a few members of the media on this issue (read more here, here, and here) and I thought I would summarize my few thoughts in a post.

• Ivan Lucas analyzes Password Recovery Speeds in this 7/10/2009 article on Lockdown.co.uk:

This document shows the approximate amount of time required for a computer or a cluster of computers to guess various passwords. The figures shown are approximate and are the maximum time required to guess each password using a simple brute force "key-search" attack, it may (and probably will) be possible to guess correctly without trying all the combinations shown using other methods of attack or by having a "lucky guess".

He offers times for passwords up to nine characters long consisting of numbers, letters, numbers and letters, and numbers, letters and special characters guessed by seven computer classes (from Pentium 100s to supercomputers.)

Erik M. Filterman asks Can Cloud Defend Against DDoS Attacks? on 7/20/2009 and claims:

If you've been thinking about moving your applications into the cloud but weren't sure how to best justify the investment, you can probably thank the North Koreans for helping to write your business case. …

Fiterman is a former FBI special agent and founder of Methodvue, a consultancy that provides cybersecurity and computer forensics services to the federal government and private businesses.

Ellen Messmer reports McAfee getting more aggressive on cloud-based security in this 7/20/2009 article for NetworkWorld:

McAfee Monday says it intends to expand its security-as-a-service offerings in recognition that customers are opting more and more to adopt cloud-based deployments.

"We already have a good foundation for this," says Marc Olesen, McAfee's senior vice president and general manager of the new software-as-a-service business unit. McAfee Total Protection Service, which has about 5 million users, is primarily cloud-based for endpoint and mail security scanning.

In addition, McAfee's Web Protection Service, wholly in the cloud, provides URL filtering and reputation analysis, while the company's Vulnerability Assessment service can scan Internet-facing systems to discover software vulnerabilities.

But McAfee anticipates a much wider push into security-as-a-service in the course of the coming year.

Cloud Computing Events

•• Leon Katsnelson describes a forthcoming IBM and xkoto Webinar, Scalability in the Cloud: Fact or Fiction?, in his The Great Cloud Computing Deception post of 7/22/2009:

… One of the most hyped and the least understood attributes of the Cloud is “elasticity” or in other words ability to get compute capacity when you need it on a moments notice and to pay only for the capacity that you actually used. This is indeed a fantastic feature and for some it will result in thousands or even millions in cost savings and more agile IT. That is – if it can be achieved – and it is one big “IF” and it is precisely the subject of this webinar. So come join us for an hour on July 28 in the conversation on this topic.

We will have Ariff Kassam, co-founder and CTO of xkoto, Rav Ahuja our IBM DB2 Cloud Computing Product Manager, Paul Lapointe, Solutions Architect from xkoto and me (Leon Katsnelson) share our opinions, do a demonstration and answer questions or engage in a debate if time permits. You can register free of charge here.

When: 7/28/2009 9:00 to 10:00 AM PDT
Where: Internet (Webinar)

• Bryce Cameron’s Welcome to [the Chicago] Azure July User Group meeting announcement of 7/22/2009 describes the 7/20/2009 meeting:

Our topic this month is Multi-Enterprise Business Applications (MEBAs), a new category of application for business collaboration that the cloud makes possible. We'll review what the current thinking on MEBAs is from Microsoft and the cloud community followed by an in-depth demo and code exploration of an Azure business collaboration application.

Dave Bost, Developer Evangelist from the Developer Platform and Evangelism team at Microsoft will be the presenter. He will discuss how MEBAs facilitate business processes that span enterprises, how they are enacted by the exchange of messages, and how complex, cross-organizational challenges are managed through these applications (e.g. Security, Data, Management and Governance).

When: 7/30/2009 6:00 PM to 8:00 PM
Where: Microsoft Downers Grove Office, 3025 Highland Parkway, Downers Grove, IL, USA

• Dan Logan reports on Cloud Presentations in the boonies of the central California coast for the week of 7/20/2009:

[T]he Cloud Computing Forum will offer a look at how cloud computing can be used to cut costs and lessen a business’ impact on the environment and [a] slide show … entitled “The Future of Computing: Green, Virtual, and Cloudy” will be presented by Winston Bumpus, president of an industry consortium called the Distributed Management Task Force.

When: 7/21/2009 4:00 PM - 8:00 PM
Where: Veterans’ Memorial Building, 801 Grand Ave., San Luis Obispo, CA, USA

Josh Fraser, vice president of business development at Santa Barbara-based RightScale, will give Softec, the Central Coast Software and Technology Association, an overview of cloud computing and discuss the benefits of deploying and automating in the cloud.

When: 7/22/2009 6:00 PM
Where: Pelican Point Restaurant, 2555 Price St., Pismo Beach, CA, USA

(Sorry for the short notice.)

• SYS-CON’s Cloud Computing Bootcamp Returns to Silicon Valley according to this 7/21/2009 post:

One-day immersive learning on November 3, 2009, at the Santa Clara Convention Center Santa Clara, CA

When: 11/3/2009
Where: Santa Clara Convention Center, Santa Clara, CA, USA

• Barbara Darrow analyzes Windows Azure pricing announced at the Worldwide Partners Conference in her Microsoft partners ponder Azure pricing complexities post of 7/14/2009. She notes:

… Perhaps most important for partners is that they will be able to bill their customers for any partner-developed Azure services and thus "own" that customer relationship. There had been concern that Microsoft would insist on controlling that transaction and relationship. [Carl] Mazzanti said that is very good for partners that they price and bill out their own Azure services. …

Carl Mazzanti is co-founder of eMazzanti Technologies, a Hoboken, N.J.-based Microsoft Gold VAR with a hosting business.

Microsoft’s Worldwide Partner Conference features 16 Azure-tagged sessions, including:

AP005 Extend Your Application to the Azure Cloud with SQL Azure Database (Mon 7/13 | 3:30 PM-4:30 PM | 220-222) Slides, ••• Video
SS001 Software-Plus-Services: Bringing it all together across MS Online Services, Partner Hosted and Windows Azure. (Mon 7/13 | 3:30 PM-4:30 PM | 217)
SS003 Lap around Windows Azure, Business Edition (Mon 7/13 | 5:00 PM-6:00 PM | 220-222) Slides (earlier slides by Dave Bost) ••• Video
AP002 Partnering with the Azure Services (Tue 7/14 | 2:00 PM-3:00 PM | 220-222) Slides ••• Video
SS006 The Azure Services Platform Partner Model and Pricing (Tue 7/14 | 3:30 PM-4:30 PM | 220-222) ••• Video
SS001R Software-Plus-Services: Bringing it all together across MS Online Services, Partner Hosted and Windows Azure (Tue 7/14 | 5:00 PM-6:00 PM | 220-222)
SS008 Embracing the Cloud: How ISVs Can Use New Microsoft Programs to Move into the Software-Plus-Services World (Tue 7/14 | 5:00 PM-6:00 PM | 215-216) ••• Video (Accompanying text for SS006 is incorrect)
US007 US Public Sector: Cloud Computing - the Government Perspective (Thu 7/16 | 3:00 PM-4:00 PM | 277)
CI011 Building the Foundation for a Cloud Computing Infrastructure (Tue 7/14 | 3:30 PM-4:30 PM | No Room Data)
Dynamic Datacenter: ••• Enabling the Hosted Cloud and Managed Services ••• Video

Slides linked from the “The Latest Resources” section of the Azure Partner Quickstart portal. This section will be updated as other slide decks appear. (Copied from Windows Azure and Cloud Computing Posts for 7/13/2009+ in the hope of adding more video links.)

Other Cloud Computing Platforms and Services

•• Steve Lesem describes Google Apps Sync for two-click synchronizing of Microsoft Exchange and Outlook data with App Engine projects in his Disrupting Microsoft: Google migrates Microsoft customers to the Cloud post of 7/24/2009:

The phrase “razor sharp focus” is a tired cliché in our field, but you have to hand it to Google. They have just announced a “two-click data migration tool which allows employees to easily copy existing data from Exchange or Outlook into Google Apps.”
By building a tool to make this migration a “point-and-click” experience, they are hastening the defection rate for businesses looking for an alternative to Microsoft’s office suite. What’s more, three service providers - NuVox, Netfirms and IKANO - have already begun offering this tool to their customer base.

Google Apps Sync, as the migration tool is called, has already been put to use at enterprises like Genentech and Avago.

It’s a case-study in business model disruption. The cost? One-sixth the price of Microsoft.

Of course we’re still in the “early days” and the jury is still out. Microsoft will surely counter with Azure, but you can see why Ray Ozzie is worried.

For Google, on the other hand, the state of cloud computing is promising. They claim around 1.75 million companies are running Google Apps. The enterprise, as Gray noted earlier, is ready for Cloud Computing. And why is this? We’ve mentioned the economics before, but here is Google’s take on the benefits of Cloud Computing.

Steve’s post contains a video of “some compelling Google propaganda” for Google Apps Sync.

•• Aliya Sternstein’s White House mulls making NASA a center for federal cloud computing post of 7/24/2009 reports:

Officials at the space agency and the Office of Management and Budget have "broached the idea of NASA becoming an IT service provider," said Mike Hecker, NASA's associate chief information officer for architecture and infrastructure. But, "NASA as an IT service provider takes us into a new realm. We're still debating if that's a good idea or not."

NASA is developing a cloud computing model, called Nebula, to support some of its projects. For example, the agency uses Nebula to share NASA images and statistics with international partners and academic institutions. The system provides high-capacity computing, storage and network connectivity

•• James Urquhart reports Sybase to enter cloud through mobility in this 7/23/2009 post:

A recent report noting some big wins in the mobile-platform space by former database powerhouse Sybase has some interesting observations about what this means for the company's role in the cloud-computing market. Since 1998, Sybase CEO John Chen has been redirecting the company's efforts toward mobility, which he claims is now paying off big time.

Most interesting to me, however, is Chen's claim that cloud computing means a big opportunity for his mobile-platform business. As the article notes:

“Sybase's mobile platform may provide a cloud-based lifeline for the likes of SAP, Microsoft, and Oracle, providing those legacy enterprise application vendors an entry into the mobile-computing world of the future.”

Another arrow in the back of Windows Mobile and SQL Server CE.

• John Foley’s Amazon Web Services Secrets Revealed InformationWeek expose of 7/22/2009 begins:

Amazon.com exercises tight control over information related to its cloud computing business, a source of frustration to anyone trying to get a complete picture of Amazon Web Services. So I went in search of information from other sources. Here's what I found.

First, Amazon does provide a few details about the size and scope of AWS. In a mid-year status report, the company said that "hundreds of thousands" of developers have registered for AWS and that the network bandwidth consumed by two AWS services – its Elastic Compute Cloud (EC2) and Simple Storage Service (S3) – now exceeds the bandwidth required for all of Amazon's global Web sites. It also disclosed that 52 billion objects are stored in S3 and that S3 requests regularly peak at 80,000 requests per second. …

And continues with more details gleaned from other sources.

• Chris Kanaracus claims Oracle Grid Update Tied to Emerging Cloud Trend in this 7/22/2009 PC World article:

Oracle this week shipped an update to its Coherence in-memory data grid, a member of a class of middleware that some say may be on the cusp of broader adoption for cloud computing.

In-memory data grids store information that applications need in memory across a pool of servers, instead of reading it off disks, resulting in major performance gains.

The Coherence product is one of the more mature in a space occupied by offerings from IBM as well as smaller companies like GigaSpaces and a number of open-source projects. Microsoft is also developing a system dubbed "Velocity."

• Tom Lounibos announces a NEW Web Services Performance Certification Program from SOASTA in this 7/20/2009 post:

I’m very excited to announce the new SOASTA Performance Certification Program designed to enable companies deploying software in the Cloud, at hosted data centers, or behind corporate firewalls to certify that their websites have been tested and have met or exceeded industry benchmarks for performance at peak levels of user traffic. For the past ten years, the dirty little secret in the web development community has been that whether due to cost, complexity or lack of resources, the vast majority of web applications and sites have not been tested at normal user volumes, much less for unexpected spikes in traffic. Which means our user communities have become the testers for virtually every website, a risk that has proven very costly time and again.

John Foley comments on Tom’s post in his Is Your Cloud App Ready For 100,000 Users? post of 7/21/2009 for InformationWeek.

• Jo Maitland interviews Mike Repass, product manager for Google App Engine, in her Google opens up on App Engine post of 6/29/2009:

Mike Repass: The question gets at, what are Google's core competencies? We know how to deal with hundreds of thousands of machines. All our hardware is custom built and not something we could easily serve up at a raw level in a way that makes sense to people. Infrastructure as a service would be a play against Google's core competencies.

We're saying, let's play with clouds and see if we can catch lightning in a bottle. Google does not say, "Let's build a product." The company doesn't work like that. …

Antone Gonsalves reports HP Buys Cloud-Computing Vendor IBrix on 7/20/2009 for InformationWeek:

Hewlett-Packard (NYSE: HPQ) has agreed to buy IBrix, a maker of file-serving software, to boost HP's infrastructure offerings to cloud-computing customers.

Founded in 2000, IBrix, in Billerica, Mass., has 53 employees and more than 175 corporate customers spanning the communications, media, entertainment, Internet, oil and gas, healthcare, life sciences and financial services industries. HP uses the company's technology in several products, including StorageWorks storage area networks, ProLiant servers, BladeSystems and ProCurve Ethernet switches and management software. …

Stacey Higginbotham claims PrimaCloud’s Virtualized I/O Takes Cloud Computing to the Next Level in her Giga Om post of 7/20/2009:

The folks behind PrimaCloud, a cloud computing and storage product that offers a service-level agreement that it claims delivers 99.99 reliability (that means it can go down 53 minutes each year), said today it will save $1 million by virtualizing its network and will spend 50 percent less to deliver its high reliability cloud. The company has installed boxes from Xsigo Systems that sit between the servers and switches and create a cloud through which the network traffic from the virtual machines loaded on the servers is routed. The network can handle traffic destined for other servers or for the storage network without requiring separate cables. …

Michael Wolf lists some of What Happened in Cloud Computing in Q2? in this 7/20/2009 post to Giga Om Pro. Topics are:

Old Guard Does Cloud Dance
McKinsey Rattles Cloud Crowd
VMware Launches “Cloud OS”
Networking Giant Cisco Jumps In
Oracle and EMC Make Big Buys

Alan Williamson’s “Quick Q&A with Geert Bevin, Evangelist, Terracotta Inc.” in his Scaling Java From the Enterprise To the Cloud post of 7/20/2009 begins:

Simple scalability - that's the Terracotta value proposition. In this Quick Q&A with SYS-CON's Cloud Computing Journal editor-in-chief Alan Williamson, Geert Bevin (pictured) - Evangelist at Terracotta Inc - explains that how Terracotta helps enable Java in the Cloud...and clarifies the difference between the commercial and OSS editions of Terracotta.

Ian Grant reports BT to host Microsoft cloud software - but not Azure on 7/20/2009 for ComputerWeekly.com:

Less than a week after Microsoft announced its Azure cloud computing initiative, BT said today it will market and deliver Microsoft Online Services to business customers, giving them access to integrated cloud computing and voice services.

Customers will have internet access to Microsoft's Exchange Online, SharePoint Online, Office Communications Online and Microsoft Office Live Meeting, all hosted by BT.

BT will embed the suite into its multi-protocol layer switching (MPLS) networks, giving customers a fully hosted service that includes power, performance management, maintenance and software upgrades. …

Neither firm responded to queries as to how this deal related to Azure, the Microsoft-hosted processing, storage and networking package launched last week.

Seems to me to be a bit premature to discuss commuercial Azure deployment in the UK when the product’s still in beta in the US.

Maureen O’Gara claims “Engine Yard Cloud platform along with Flex is a cloud service plan for production-level Rails applications” in her Ruby-on-Rails Apps Get Cloud Lift post of 7/20/2009:

Engine Yard, the Ruby-on-Rails doyen, is supposed to announce its Engine Yard Cloud platform today along with Flex, a cloud service plan for production-level Rails applications.

The company says its Engine Yard Cloud leverages hundreds of man-years of experience in deploying, managing and scaling some of the world's biggest Rails sites and will put that know-how to work for businesses that want to run Rails on Amazon's EC2 public cloud.

It automates the deployment and management of applications that fetch up to a million unique visitors a month. More and the company will advise shifting to its dedicated infrastructure.

Rich Miller’s Report: Microsoft Plans Data Center in Brazil Data Center Knowledge post of 7/17/2009 says:

Microsoft (MSFT) plans to add a new data center in Brazil to support its online services business in Latin America, company officials have told Business News Americas, which said the facility will be operational in the fourth quarter of this year, and will host Microsoft’s online services for the business market.

Although the report quoted several executives from Microsoft’s Latin America business unit, the company is not confirming any plans for a data center in Brazil.

“We don’t have anything to announce about new facilities at this time,” said a spokesperson for Microsoft Global Foundation Services, which builds and operates the company’s data centers. …

Microsoft’s Confirming Commercial Availability and Announcing Business Model release from #WPC09 says, inter alia.:

A few months after, in March 2010 to be more precise, Microsoft will target 16 additional countries with Windows Azure: Brazil, Chile, Colombia, Czech Republic, Greece, Hong Kong, Hungary, Israel, South Korea, Malaysia, Mexico, Poland, Puerto Rico, Romania, Singapore, and Taiwan. [Emphasis added.]