Betanews - IT Systems

Intel's marriage of CPU and GPU not ready for prime time

Scott M. Fulton, III — Mon, 07 Dec 2009 12:54:22 -0500

A spokesperson for Intel confirmed to Betanews this morning that the company's highly anticipated initial release of a commercial processor product based on its CPU+GPU architecture, code-named "Larrabee," will not come within early 2010 after all. This despite the first public demonstration late last September that seemed to indicate all was on track for a release in the first half of 2010.

"Larrabee silicon and software development are behind where we had hoped to be at this point in the project," stated spokesperson Nick Knupffer. "As a result, our first Larrabee product will not be launched as a standalone discrete graphics preoduct, but rather be used as a software development platform for internal and external use."

The project represents Intel's first attempt at a commercial product architecture that implements pipelining architecture for parallelism. Pipelining is what makes graphics architecture possible, essentially taking the same series of instructions with different data and distributing them through a very long chain. For massively parallel applications like graphics (drawing billions of filled triangles in real-time, for instance), pipelining is far preferable to multi-core, which doles out multiple, dissimilar threads that run in parallel but that are distributed sequentially.

Still, pipelining can be useful for applications other than graphics, which is one reason Intel and other manufacturers, including Nvidia and AMD division ATI, are actively experimenting with hybrid architectures. Some very good signs had been emerging from Intel's experiments, including not only a beautiful demonstration at the last Intel Development Forum in September, but a demonstration at a supercomputer demonstration last week by no less than Intel CTO Justin Rattner.

There, Rattner closed his keynote speech with a Steve Jobs-like "One More Thing" demo: a live test of a Larrabee architecture chip breaking the one teraflop barrier (one trillion floating point instructions per second). That's not a first, even for a commercial graphics card -- ATI was demonstrating teraflop cards in mid-2008. But it would put Intel in a position to compete directly against AMD's ATI division, as well as no-love-lost rival Nvidia, at the high end of the market.

Maybe not next year, though, as Knupffer's comments indicated that we'll hear an update about a Larrabee graphics card's progress "some time in 2010." At that time (whenever that really is), Intel may go into more detail as to the development of a "throughput computing development platform" based on Larrabee -- probably a blueprint for more supercomputing applications, but not consumer products.

There will be some sort of consumer hybrid computing product from Intel this month, Knupffer said: "Our plans to deliver the world's first CPUs with integrated graphics this month are unchanged." But that will evidently not be on the high end of the scale, but more of a value play. Intel has long been a producer of integrated graphics components, but in the performance department, they've never really won awards, especially from consumers.

An alternative to Research in Motion's enterprise e-mail? There's an app for that

Tim Conneally — Mon, 07 Dec 2009 12:46:27 -0500

By Tim Conneally, Betanews

Hardly a day goes by when our inboxes and feeds aren't flooded with messages from companies announcing that they have created a new iPhone application. They range from the disappointingly simple to the disconcertingly arcane, but as a whole skew more toward the consumer. So when a significant player in enterprise services releases an iPhone app, it's worth looking into...especially when it's from an "underdog" company trying to challenge a market leader.

Good Technology, a long-running provider of mobile e-mail solutions, today debuted its Good for Enterprise iPhone app, which provides secure access to corporate e-mail, calendars and contacts with companies using the Good for Enterprise mobile e-mail solution.

This means Good's e-mail service -- by many accounts, the strongest competitor to BlackBerry -- is now available on nearly all of the most popular handset operating systems: iPhone, Android, Windows Mobile, Symbian S60, PalmOS.

Good Technology has been competing as the "little guy" in corporate mobile e-mail for several years, offering an alternative to powerhouse Research In Motion. It's even commenced litigation against both Research in Motion and Microsoft concerning its remote e-mail patents.

But now that Good for Enterprise is on the iPhone, the company can present a much stronger threat to Research In Motion's enterprise e-mail dominance.

"The iPhone is far and away the most requested device for accessing corporate networks," Good Technology's chief marketing officer John Herrema said in a statement this morning. In the company's survey of 500 IT departments in the US and UK, it found that nearly 82% of respondents wanted to hook their iPhone into their work accounts.

Five improvements for IT managers in 2010

Thu, 19 Nov 2009 20:51:32 -0500

By Ed Moyle, TechNewsWorld

Every year around this time, everyone from antimalware companies to analyst firms line up to tell us about the top IT and security trends -- what they are and why we should care. This year, chances are they'll tell us all about cloud computing, virtualization, and social networking, and why these technologies are the new best (or worst) friends for security folks in 2010.

Now if you're sensing a bit of snarkiness here, you're right -- I find these lists a bit frustrating. That's not because of inaccuracies in the lists themselves (to the contrary, many of them are dead-on), but instead because they sometimes inappropriately drive how IT managers make budgeting decisions. Don't get me wrong, keeping abreast of the new areas is always valuable -- and I'm always fully on board with keeping us and our staff up to date and capable of reacting to new types of threats. But it's also important to keep in mind that what's new isn't always what's most critical. Where should you be investing budget dollars? At critical areas, not just what's new and shiny.

To illustrate this, consider a firm that doesn't use AV (antivirus) and also allows users to access social networking sites. The trend predictions are likely to clue us in about why social networking is something we should care about, but they might not mention malware at all (after all, that's been around forever). But if your firm doesn't yet have a cohesive antimalware strategy ... well, you've got bigger fish to fry than how, when, or what your employees tweet. In other words, when it comes time to allocate budget for new projects, you need to consider both the new and the old -- both the upcoming trends that Big Analyst Firm says are emerging, as well as the "tried and true" fundamentals that don't get as much play.

In the field, it's all about the basics. When you stop to think about it, how many of us are really where we need to be when it comes to the fundamentals? Which position would you rather defend: that your firm was hacked because of some newly emerging threat, or that you got hacked because you weren't doing the generally accepted minimum practice?

So in the spirit of keeping one eye on the practical, here's my New Year's list -- or, more accurately, my "reminder list": a highly unscientific breakdown of the top five basics that are often overlooked in enterprise. These are things that you probably should be doing, but might not be -- and things that you could probably do more with, but maybe aren't.

1) Vulnerability Assessment. There are several reasons why you might not be doing as much vulnerability assessment as you could be: It can bring down critical systems, it requires some specialized knowledge to vet false positives, it has a high overhead in terms of care and feedback by staff members. As such, there are quite a few organizations that just don't use it at all -- and for companies that do, it's often inconsistently deployed.

However, aside from this complexity, it's also one of the most valuable areas of feedback that you can get about how your organization performs. Data about the effectiveness of your patch management processes, your password policy, and your system-hardening procedures are all directly and practically observable through the data coming out of vulnerability assessment results. If you haven't deployed it yet, the technology is cheap, mature and commoditized.

2) Asset Inventory. How many of us have a detailed inventory of all the "stuff" on our networks? Organizations tend to grow their IT organically, so many of us are in the situation where going back to inventory what we have fielded is a huge, expensive undertaking. Even when we do have some idea of what's out there, there are very often "gaps" in our understanding of our environments. For example, we may have a pretty clear idea of what desktops are fielded but not have tremendous insight about applications.

If you don't have a clear idea of what you have fielded in your organization, now is the time to put together that inventory you've been putting off. Leverage existing tools like VA reports or business impact analysis documentation to put together a rough "map" of what you have fielded and keep it updated as changes occur. You don't have to have a fancy system to do this. Start small and grow your inventory the same way you grew your network -- organically.

3) Provisioning. We all know the ideal end state of user provisioning: defined roles that govern access to network resources and applications. But in practice, when the topic of provisioning comes up we wind up going down a path that involves deploying complicated systems or that involves significant effort parsing out users based on vague or poorly defined roles. While we wait for the dust to settle, the day to day business of assigning new users to applications moves ever forward -- often with little or no assist from security staff and even less organization.

However, a solid map of roles doesn't have to be complicated. Start by defining roles at the very highest levels, and get more granular over time. Don't have a provisioning system deployed? Delegate responsibility for creating roles to subject matter experts who use the application all the time. Check in with them periodically to make sure they're doing the right thing.

4) Audit and Monitoring. IDS systems are chatty, and the individuals who review alerts are often under significant stress and workload already. At the OS and application levels, staffers are often too overwhelmed to review log and activity reports as much as they should. So who has time to keep up? In many organizations, the day-to-day monitoring of audit and activity logs tends to fall by the wayside -- for folks that even have auditing features enabled.

However, compliance mandates like PCI, HIPAA and others specifically require review of audit logs, so failure to make this happen is not an option. Step up what you review and how often you review it. Institute spot-checks to make sure that staffers are doing their jobs when it comes to reviewing this critical data.

5) Business Continuity Planning. Let's face it, BCP is a lot of work. It involves participation from all areas of the organization -- from subject matter experts to business to management. Because of the number of folks involved, very often we don't have time for formal models, or when we do, very often our analysis goes without updates for long periods of time.

However, planning for contingencies is beyond critical, and all that data you're getting about applications, systems and business processes can be recycled for other purposes within your security program, such as triage during an incident response exercise, risk analysis, and even asset inventory. So maybe now is a good time to do a refresh on this valuable data -- or start doing it if you haven't already.

Now maybe your company has already hit all these topics, and you don't need me to remind you to "eat your vegetables." If so, nice work -- count yourself among the well-positioned minority.

However, if in reading through these items you see areas where you could be doing better, remember that boning up on the basics is just as important as looking for new ground to cover. After all, the basics might be "old hat," but that doesn't mean they're not important.

Originally published on TechNewsWorld

PDC 2009: Windows Server's plan to move customers back off the cloud

Scott M. Fulton, III — Wed, 18 Nov 2009 02:21:47 -0500

By Scott M. Fulton, III, Betanews

Much of the value proposition for Windows Azure -- the star of the show Tuesday at PDC 2009 in Los Angeles -- has been its ability to open up new business avenues for customers who had not been able to envision hosting high-intensity data center operations before. Azure could give these customers a leg up, a new and more affordable way to get off the ground.

But once they're off the ground, the question becomes, why stay up in the air? What's to keep those customers grounded -- to mix metaphors like an old editor of mine -- in the cloud? The surprise answer to that question is coming from a senior product manager for Windows Server, not Azure. Scott Ottaway told Betanews today that provisions are being planned for customers to move their deployed applications back off the Azure cloud, onto on-premises data center servers.

"Our goal is to provide a common application platform, just the way that we support ASP.NET in both places, PHP in both places. We want to supply a common platform so customers don't have to make the hard choice up front about where they're going to run something."

During the October 2008 PDC conference when Windows Azure was first introduced, it wasn't exactly clear who Microsoft was targeting as its customer. Thirteen months later, we have a much clearer picture of cloud services customers comprising three discrete classes: one that is made up of SMB businesses investing in affordable data center architecture for the first time, and building entirely new cloud applications that have never been tried; another made up of applications and services hosts that are simply mirroring their existing apps to the cloud space for affordability and scalability; and a third class in-between comprised of businesses of all sizes, who aren't looking to the cloud as a migration platform, but as a way to backup, complement, or augment their existing customer services as necessary.

The common factor between each of those classes is the need for a bridge -- perhaps now, perhaps later -- between their on-premise and off-premise cloud platforms. That's the reason for Microsoft's latest brand, announced Tuesday morning.

"If you write your code for Windows Server AppFabric, it should run on Windows Azure," said Ottaway, referring to the new mix-and-match composite applications system for the IIS platform. "What we are delivering in 2010 is a CTP [community technology preview] of AppFabric, called Windows Azure AppFabric, where you should be able to take the exact same code that you wrote for Windows Server AppFabric, and with zero or minimal refactoring, be able to put it up on Windows Azure and run it." AppFabric for now appears to include a methodology for customers to rapidly deploy applications and services based on common components. But for many of these components, there will be analogs between the on-Earth and off-Earth versions, if you will, such that all or part of these apps may be translated between locales as necessary.

"Right now, if I have an ASP.NET app running on Windows Azure, and it's using SQL Azure, I can pull that off with minimal refactoring, in most cases, and run it on premises on Windows Server, because Windows Server supports ASP.NET via IIS, and our .NET Framework, and it supports SQL Server," remarked Ottaway. "So I can move a certain class of apps that I have up there, off -- not all." Applications that utilize binary large object (BLOB) storage, and other Azure-specific features, would still require significant refactoring. "But if it's just an ASP.NET app through SQL Azure, you should be able to pull it back on-premises without much effort at all."

Some of the more recently deployed, non-Microsoft language platforms, including PHP and MySQL, will also aid customers in that transition.

"Whether there's a total cost-of-ownership advantage to being off-premises or on-premises, depends on so many factors. Are you highly virtualized? Do you have really good management tools? Do you have affordable staff? Regulatory concerns? Privacy, security concerns?" remarked the Windows Server senior product manager. "There's many, many factors that may indicate to you that...you want to keep it on-premises, or that it's okay to go all the way off-premises. But it's probably going to be more of a non-enterprise decision; it's going to be project by project and app-by-app."

Next: Virtual machines migrate to Azure...

Virtual machines migrate to Azure

One of the distinguishing factors between Microsoft's cloud platform and Amazon's has been the distinction in what's being served. Specifically, Amazon's EC2 gives customers a way to deploy entire server images on its cloud, while Azure provides an active cloud-based .NET runtime for the deployment of applications rather than servers. Still, that distinction will grow fuzzier, first with last week's announcement by Amazon of an SDK for cloud-based applications deployment, more similar to Windows Azure.

Then there's Microsoft's announcement Tuesday -- effectively made by Ottaway during an afternoon press conference -- of a system for deploying Windows Server virtual machines to its cloud.

"What we're announcing is that we will have what we're calling Windows Server Virtual Machine Roles on Windows Azure," Windows Server senior product manager Scott Ottaway told a press gathering Tuesday afternoon at PDC 2009. "What these will be, are pre-configured images of Windows Server -- one might have .NET [Framework] 3.5 on it, the next one might have .NET 4.0 when that's available. And you can bring your existing app, that you're running on-premises now on Windows Server, to Windows Azure, deploy it in this virtual machine image, and there you have existing applications support or migration -- you don't have to necessarily write new apps to take advantage of the Windows Azure service."

Notice the distinction still exists, if lessened a bit: VM Roles will provide Windows Server admins a way to deploy apps in the cloud as though they were being deployed on virtual machines locally. It's still not deploying servers, like the Amazon model, but it's closer. One big reason, Ottaway said: Customers already have applications that they paid for; they don't want them redeveloped.

But the other reason (and here's the basis for Microsoft's new value proposition for Azure) is that it makes less sense for customers to deploy servers in the cloud (including from a licensing perspective) when their goal is really to add scalability to their applications and public-facing services, not to Windows.

"Certain companies out there say everything's going to move off-premises, and everything's going to be a service. But there are two issues here: Network latency, the sweet nectar of having a big pipe and fast response time, is a big inhibitor for a lot of things moving off, that might make a lot of sense. And then, you want Print Server nearby, right? You don't want to send your print requests to Dublin [the Microsoft service, not the country] before it gets to your printer. Then for security servers and your management servers for managing all your clients, you're definitely going to want them to be on-premises."

Ottaway pointed to a compelling demonstration during the Tuesday morning keynote, of an online automobile shopping service provided by Kelley Blue Book using Silverlight. That service was devised to take advantage of a feature called cloudbursting -- using the cloud where necessary, such as in periods of heavy traffic, but not necessarily as a principal deployment platform.

"The scenario they were talking about on there was, in general, most of the time, their application and business presence runs on Windows Server and SQL Server and Silverlight. But currently, they've had to have a large amount of investment in server hardware, for a peak capacity that occurs very rarely across their year. And what they're essentially using Windows Azure for [instead] is for cloudbursting, so when they do have unexpected peak capacity, they can very quickly spin up a bunch of new instances of their Silverlight application and be able to meet the demand that's occurring.

"The key thing that's really interesting here -- and where Microsoft, again, is differentiated from other cloud providers -- is that Kelley Blue Book also wants to be able to scale down, and bring everything back on-premises when that demand isn't as great. It's a hybrid scenario; they're using off-premises Windows Azure to meet their increased capacity demands. That's a unique advantage of having a Microsoft infrastructure on-premises, as well as Windows Azure."

One reporter asked Ottaway today, is there a way for Azure customers to utilize any kind of capacity planning tool -- to estimate what it might cost to cloudburst under varying traffic circumstances? The long answer from Ottaway -- who seemed to take some inspiration from this question -- was no. Online calculators do exist to help customers set caps on their expenditures, and to turn off service once those caps are exceeded to avoid tremendous billing. Azure can also provide customers with warnings in advance of heavy charges. Even with those stopgaps in place, he acknowledged, it may become cheaper for certain customers to keep their applications on-premises.

Customers will need to consider those possibilities and scenarios for the future, Ottaway said. But for now, the templates and policies and best practices -- the "solution accelerator" for the cloud, to borrow a phrase -- is very much in the works.

Microsoft's Top 3 advances in Exchange Server 2010

Scott M. Fulton, III — Mon, 09 Nov 2009 18:16:34 -0500

By Scott M. Fulton, III, Betanews

The biggest change to Microsoft Exchange Server 2007 was supposed to have been the introduction of something called Unified Communications -- the introduction of a singular console for the handling of all forms of digital communication, wrapping voice mail, instant messaging, and e-mail into a single delivery system. History may yet vindicate UC as the product's singular achievement.

But in the near term, administrators credit Exchange more for what it gives them than the world at large. In that light, the inclusion of PowerShell as not only the underlying language of the system but as its engine as well, changed everything for the admin. It may very well be why the product has surged to a two-thirds market share, by some estimates, over once formidable competition such as Lotus Notes.

So learning a lesson from history, the message from Microsoft with regard to Exchange Server 2010, which went on sale this morning, is about new levels of control. The idea that e-mail, or any kind of communication, once sent unto the vast Internet is out of the sender's hands -- like a paper sailboat launched from a river pier -- is what the Exchange team has been working to combat. During a beta program which Microsoft says involved dozens of universities, signing up some ten million participants worldwide, the company has completed development of a browser-based endpoint for ES 2010-delivered e-mail that is not only more manageable than Outlook 2007, but that has beaten Outlook 2010 -- the product it's supposed to be derived from -- to market by perhaps eight months.

What that means is, hopefully for a short time only, there will be a functionality gap between what the new Outlook Web App -- hosted by ES 2010 -- can deliver compared to what Outlook 2007 provides. If Julia White, Microsoft's marketing director for Exchange, has anything to say about it, that gap will be shorter rather than longer, but it's not unnoticed.

Microsoft Outlook Web App previews the textual contents of a voice mail. [Courtesy Microsoft Corp.]

White spoke with Betanews this afternoon from Berlin, where she had just completed a TechEd Europe demonstration along with Corporate Vice President Stephen Elop. "Obviously Outlook Web Apps comes with Exchange, so they can use that today; when Outlook 2010 comes out, they can use that," said White, "and we are absolutely planning support for Outlook 2007 in the roadmap here. So it's on the agenda, and we will actually be getting to it."

Much of what Exchange 2010 will deliver absolutely depends on this upgrade to Outlook 2007, as you'll see. We asked White for her take on what she would consider the top three enhancements to administrator functionality in ES 2010.

#3: Transport Protection Rules

Number three on this list is the Transport Protection Rules system, which we described earlier today. It enables the administrator to designate the extent to which the recipient of a message can utilize its contents, based upon rules that enable Exchange to analyze the content itself. "In the demo this morning, I set a Transport Protection Rule based on a keyword. But actually another aspect of that is, those rules can be set based on the sender, the recipient, or even contents of an attachment," White told Betanews.

"Any of those things can be triggered; and having the ability to centrally decide what gets encrypted and what doesn't, is a really powerful tool. With end users, it's hard for them to keep up with corporate policy, pay attention to it, or know about it. So oftentimes it's unintended, versus intended, when information isn't protected. Having that essentially managed brings peace of mind, for the users as well as the IT pros."

The ability to analyze an attachment takes place on a granular level, White told us. If a PowerPoint presentation, for example, were to contain the words "Microsoft Confidential," that fact alone would trigger a rule that automatically encrypts the message outgoing, and that restricts the recipient from being able to pass it on.

#2: Role-Based Access Control

One of the least loved features of Exchange, or anything Microsoft has ever done, disappears in ES 2010: The Access Control List is a Registry-based system for designating which identified and authenticated user had permissions to control specific objects. It has often been a ridiculous concept that starts one off with the assumption that everyone has rights to everything, and that ACLs provide the exceptions.

Exchange Server 2010 replaces this entirely with a concept that is much more rooted in Active Directory. Now, the administrator starts off in a universe where nothing is allowed until groups of users are added into the pool of permissions. Those groups that are added in are called management role groups, with the concept being that a predefined set of roles exist (a concept made popular by Windows Server 2008), and that groups of users or individual users are delegated those roles.

This morning, Julia White demonstrated how Role-Based Access Control enabled an otherwise unprivileged user to search for e-mails through multiple mailboxes on the company's behalf (in this case, Microsoft's usual fictitious firm, Contoso). Her system was delegated a role that let her perform the search, without having to delegate other responsibilities and privileges of a much higher administrative order. "A compliance officer might get that level of capability," White explained to us, "but a help desk might have the rights to increase mail box quota size. Maybe HR would be given the ability to update contact information on behalf of employees. Extending all the way down to end users, even that same roles-based administration capability -- end users can now create and manage their own distribution groups within Exchange. That no longer requires a call to the IT pro...usually that's a lot of overhead."

Next: The best thing ever to happen to old e-mail...

#1: Integrated archiving

During the late 1980s and into the '90s, Microsoft liked to centralize things, thinking that if everything were in one big pile -- as Arlo Guthrie put it -- that would beat two or more little ones. The System Registry is, and remains, one big pile. Another -- which can stink just as bad -- is the .PST file, the single personal folder file that is created on the client side by Outlook.

It is every Outlook user's nightmare, especially since Office buries this file typically in a black hole within a hidden directory inside each user's Documents folder. For individuals who receive hundreds of thousands of e-mails per year (I'm on that list, believe me), the archiving process has cost users many a weekend.

With Exchange Server 2010, Microsoft marketing director Julia White told Betanews today, is the ability to perform this process completely in the background. But in addition, the archived items remain indexed and available, still listed as part of "Personal Folder" but stored separately.

"Today, the vast majority of e-mail actually sits on the local hard drive on those .PST files," White remarked.. The end users love it because they can file as much as they want in there, and they have access to it when they're on their PC. But from an administrator's perspective, they don't like them because they're very expensive to discover, they get lost, they get corrupted, it's a liability and a lot of overhead for the IT organization.

"So with integrated archiving...it doesn't have any change to the end user experience," she continued. "That Personal Folder appears, but the archive shows up and it looks just the same, it's another folder in your file directory, it looks like a secondary Inbox...The benefit is, it's all sitting on Exchange, so it's not going to get corrupted or lost. It's very easy to discover -- that time comes down dramatically. And as a user, you get access to it through Outlook Web App, [as opposed to] on the local hard drive."

Here, White took the bold step of proclaiming OWA as superior to Outlook, in that users still get full access to their mail (albeit with transport restrictions), but without having to keep those multi-gigabyte .PST files locally:

"What got us into this in the beginning was when we talked to our Exchange customers as we were planning [ES] 2010, and we found out that 20% of Exchange mailboxes have an archive on them today, but over 60% said it was important to them. It's scary, because there's not a mailbox out there that shouldn't be archived for one reason or another. What we heard from them was, 1) the cost and overhead of maintaining and managing another system -- new tools to learn -- was too expensive; and 2) the end-user experience. Oftentimes you have an archive today, you have to go to a different UI to retrieve the mail, or the performance is really poor on the archived mail. Because what they do is called 'stubbing,' which means they literally just leave a little bit of the e-mail in the Inbox, and the rest of it sits out on a third-party system. So the performance has to go bounce between multiple systems, so it's very slow.

"If end users don't adopt it, it doesn't work," White remarked. "So this clears the hurdle of both the end user experience as the IT pro cost and management perspective."

That 70% cost savings claim

During this morning's presentation at TechEd in Berlin, Microsoft CVP Stephen Elop made the staggering claim that within a group of 100 companies testing Exchange Server 2010 over the last year, some were able to cut their administrative costs over earlier versions of Exchange by as much as 70%. As is Betanews' custom (and as is the custom of Betanews readers who see anything in double-digits beside a percentile mark), we asked how that figure was obtained. For instance, we've seen companies in the past that said the expenditure to do something this year was X% lower than the expense to do something in the past, and that typically refers to the fact that memory or storage or processor power is just that much cheaper. That's not really savings; that's a factor of the economy.

So what is this 70% savings a factor of? "A big cost driver is storage," responded White. "We know the storage aspect of e-mail, it's a lot of information and it can get expensive. Traditionally, Exchange was deployed always on a storage-area network, which was fine back in the day when you had a 200 MB mailbox. Obviously, that's not sufficing anymore, and 10 GB is becoming more of a standard. Supporting that kind of mailbox storage size on a SAN becomes cost-prohibitive.

"So what we've done in Exchange 2010 is two things: First, we dramatically improved performance, tenfold over Exchange 2003. When I say that, I mean the time it takes to read and write information to the disk. What that enables is world-class support of low-cost storage options -- direct-attached storage, SATA, even in a JBoss configuration. So big, slow disks, you can run Exchange without any performance or reliability impact." NEC Philips, for example, was able to increase its storage capacity by a factor of eight, while simultaneously reducing costs by a factor of four, White said; and Germany-based hosted service provider Elabs was able to reduce its storage costs by 70%.

Isn't that saying that the expenditure this year is 70% or so less than the expenditure for a similar service in 2003? Yes, according to White, but that's in terms of operating cost run-rate, which is figured according to time and not total investment, especially since companies don't always purchase storage capacity all up-front.

Betanews also learned today that Microsoft's SMB Windows Server bundles, Small Business Server 2008 and Essential Business Server 2008, will not be updated immediately with Exchange Server 2010. Those bundles may continue to be sold with Exchange Server 2007 for at least several more months down the road.

Amazon lowers EC2 cloud service fees, adds MySQL relational instancing

Scott M. Fulton, III — Tue, 27 Oct 2009 18:07:26 -0400

By Scott M. Fulton, III, Betanews

Come November 1, Amazon's Web Services division will be lowering the per-hour prices for all of its current five instance types (AMIs), while adding two new AMI types on the high-end, according to a multitude of announcements from Amazon today. At the new high end of the scale will be a "quadruple extra-large" AMI with 68.4 GB of dedicated RAM, and the virtual computing power of a 1 GHz, 26-core Intel Xeon processor (albeit a 2007 model).

The new high-end instances won't come cheap -- they'll carry a premium of $2.40 per instance-hour for Linux editions, and $2.88 per instance-hour for Windows Server 2003. The previous high-end AMI, still called "extra large," had been priced at nearly one-third that amount.

However, revenue from the new super-high-end will help drive down prices for everyone else, starting November 1. At that time, the per-hour price for the smallest and cheapest instance available, running generic Linux, will be reduced by 15% to $0.085 per hour. Windows Server instances will be trimmed a bit, but not by as much percentage-wise -- the "extra large" price, for instance, will drop only 4¢ to $0.96 per hour.

The price cuts come as Amazon looks to offer more competitive buildouts, for customers that continue to prefer to deploy entire machine instances -- rather than just applications, such as Salesforce.com and Microsoft offer -- in the cloud, administered using everyday software. Already Amazon has been offering pre-configured AMIs with Microsoft SQL Server (at about a 10% premium per-hour), and IBM DB2 (somewhat higher at $0.38 per hour). Now Amazon is committing to offering its own brand of database server, called Amazon RDS, in lower-priced instances that will compete with its SQL Server, Oracle, and DB2 instances, using MySQL as the underlying engine.

"For customers whose applications require relational storage, but want to reduce the time spent on database management, Amazon RDS automates common administrative tasks to reduce complexity and total cost of ownership," reads a statement from Amazon AWS to Betanews this afternoon. "Amazon RDS automatically backs up a customer's database and maintains the database software, allowing customers to spend more time on application development. With the native database access Amazon RDS provides, customers get the programmatic familiarity, tooling and application compatibility of a traditional RDBMS. Customers also benefit from the flexibility of being able to scale the compute resources or storage capacity associated with a Relational Database Instance via a single API call."

Customers will still be expected to maintain their own databases, Amazon's statement tells us, although instancing in the cloud will enable them to re-provision resources as necessary on a more granular basis. Multiple statements today managed to make mince meat of Amazon's quoted rates for transfer, but a blog post this afternoon managed to straighten the matter out: RDS customers will be charged 10¢ per gigabyte per month for storage, and another 10¢ per month for every one million I/O requests. Bandwidth charges should then be the same as for Amazon's existing, non-relational SimpleDB instances: The first gigabyte of data in or out is free, then fees rise to 10¢ per gigabyte in and 17¢ per gigabyte out, declining to 10¢ per gigabyte out after 150 TB.

This will not be the first appearance of MySQL on Amazon's cloud; the commercial MySQL Enterprise has been available under Sun Microsystems' branding since this time last year.

Flashback 1990: The debut of Windows 3.0

Scott M. Fulton, III — Wed, 21 Oct 2009 16:24:00 -0400

By Scott M. Fulton, III, Betanews

This is most likely neither the first nor the last article you will read on the subject of Microsoft Windows 3.0. The attention being given the new product is not only deserved, but in many cases carefully orchestrated. The weeklies and fortnightlies have already extolled the merits of Win3's "three-dimensional" buttons, proportional text, and now-boundlessly managed memory. Their gold-star awards have no doubt been bestowed upon the product for being the best in its class, albeit the only product in its class. The "pundits" have already laid blame upon someone for Win3's alleged tardiness to market. The entire story is so well-patterned, it may be read without ever having laid eyes to the printed page.

Yet if we follow the pattern, we miss the real story...

It is May 1990. For several months, reporters had been prepared by Microsoft to cover what was being billed as the most important event in the history of software. It was the beginning, we were told, of the end of DOS, and the birth of a new software "ecosystem" that enabled independent developers to build graphical applications for the first time, without having to jump through the many hoops and stroke the countless egos of Apple. Microsoft would have a hands-off policy in the development of software that supports what was being called, for the first time, the Windows Operating Environment.

Sure, it still used the MS-DOS bootstrap, but don't tell anyone that. And sure, that bootstrap still required 640K of conventional RAM, but don't tell anyone that either. The real benefits were to be seen in something Macintosh itself couldn't do: run more than one application at once, with true multitasking and pipelining for the very first time...and all in color.

The prospects for applications were boundless, and Microsoft wanted to be seen as opening all the doors and not stepping through them first. The first question in journalists' minds was, would there be a counterpart to Hypercard? Without a Hypercard, Windows may as well be broken. Rest assured, we were told, a company called Asymmetrix would provide the toolkit that would revolutionize programming, with a bit of Microsoft's funding. The next generation metaphors for Windows were being created not by Microsoft but by Hewlett-Packard, for a product called NewWave -- again, Microsoft made certain journalists knew, with its help but not its supervision. And the world would know Windows was for real when it used an everyday spreadsheet with a name familiar to everyone: Lotus 1-2-3 G.

In the spring of the turn of the decade, I had a regular series in a magazine that was widely considered to be "Computer Shopper in exile," called Vulcan's Computer Buyer's Guide, staffed by many former Shopper regulars who would, like myself, become regulars there again once a dispute with the new owners, Ziff-Davis, was resolved. I had the lead role in covering the biggest software release in history, for a magazine whose editors told me flat out, "Use as much space as you need."

Months earlier, Microsoft had granted me some of the first demonstrations of Dynamic Data Exchange ever shown outside its laboratories. It was astounding to me, and I was proposing to write a book on it all, except that none of the book editors at the time knew, or appeared to care, about running two applications at once. "We want a book about Excel or a book about Word," one editor told me toward the close of a conversation. "No one wants to read a book about Excel and Word."

It was uncharted territory, as every editor I worked with kept reminding me. One of these days, my former Shopper editor told me, you'll be writing this story in May and someone on the other side of the screen will read it in May. But for now, it was the August issue we were working with, and complete with interviews with everyone we thought would matter -- Asymmetrix, HP, and Lotus included -- I headed forward for 33 pages of draft copy, with a full head of steam...

Yet if we follow the pattern, we miss the real story. There is a real development taking place between the authors of and for Windows 3.0, which concerns the remodeling of the computer application. We are familiar with the application as a program and its associated data, which is entered and exited like a jewelry store or a bank. We sometimes see ourselves "in" an application, just as we often see ourselves "in" the subdirectory pointed to by the DOS prompt. The data we need while we're "in" the program is much like the diamond necklace behind the display case; we're allowed to look at it and touch it, but unless we're very crafty, we're not allowed to take it outside. It doesn't belong to us, even if the data's very existence is due to our having typed it in.

The entire contraption of the DOS environment -- along with the guilt feelings it so subtly leaves us with -- are being shattered by Windows 3.0. There is a movement under way by Microsoft and its independent software vendors (ISVs) to abolish the structure which grants exclusive ownership rights of a set of data to an application. Having done that, the movement will also seek to dissolve the programmatic barricade which surrounds the once-exclusive application, allowing for the equal distribution of correlated tasks within an arbitrarily-defined computing job, to other programs non-specifically.

It's a difficult concept to discuss in the orchestrated fashion with which we have become accustomed, so instead I offer a hypothetical situation: Assume you have an inventory list card file. You want to compare gross profit percentages, so you demand such a list from the computer. Your spreadsheet -- whatever that may be -- shows you the list. You didn't need to save the card file, translate it, export it, and re-import it -- the list simply appeared. You want to see how these figures look graphically, so instantly you see a detailed pie chart. You'd like to make this chart part of your report to your superiors. This is quite simple to accomplish. Since what you're reading is the report to your superiors, the word processor saw your chart and automatically composed a standard form. This was sent to your typesetting program, which is providing the image you're seeing now.

Your superiors are in six different countries and two of them are out on the road somewhere. you tell your machine to send them all a copy of the report you're looking at now. The machine already knows two of them have fax machines, two are available via WANs, and the other two have cellular phones connected to laptops. You neither know this nor care; you just have your computer "send" the report to them, regardless of the media of transmission. The report is received in six different places, even if the recipient computers' operators weren't using their machines at the time. A mere three minutes of your life have been expended in the processing of the weekly profit report.

You have just been witness to an example of the model for the meta-application -- one smoothly-flowing, correlated process combining the resources of several programs from different vendors. This is the model Windows 3.0 seeks to gradually implement. Actually, this is what OS/2 was supposed to implement at first; its muddled and haphazard development agenda has prevented it from leading the way.

The meta-application is not an inevitable fact of computing; the marketing debacles of cross-vendor cooperation it imposes may render it as ineffective as OS/2 in changing our computing habits. Still, it is something to be wished for. And it is a far more important facet of the Windows 3.0 story than faceted buttons and little pictures. The way in which world industry and commerce works is not affected in the least by faceted buttons and little pictures.

A Sidekick crisis post mortem on cloud confidence

Tim Conneally — Wed, 21 Oct 2009 16:15:41 -0400

By Tim Conneally, Betanews

I was sitting in the dentist's chair getting my teeth drilled, while the technician complained about her now-worthless Sidekick. With no way to access her contact lists, she couldn't get in touch with her family due to arrive in DC for a reunion, and had to rely on the frequently failing device as a simple inbound line for family members to contact her. When that failed, she had to use a payphone.

It was a pretty sad story, and thousands of users were faced with a similar communication breakdown...for more than a week.

People threaten to sue Google when Gmail goes down for mere hours. But the poor Sidekick customers lost use of their phones, lost their personal data (in some cases permanently), and even lost the ability to turn off or restart their devices for upwards of nine days.

Microsoft and T-Mobile claim to have finally restored much of the lost data, and now offer a recovery tool on the T-Mobile site, but the incident was just too tremendous to simply walk away from. The whole concept of Web-based, shared, and distributed cloud services now has a huge black mark smeared across it, even though the Sidekick's contact, calendar, photo, and info sync only vaguely constitute a "cloud" service.

The resulting cloud doubt was never as clearly portrayed as it was by ABC News columnist Michael Malone, who said, "What counts is that we never really ever trust the cloud again."

But "the cloud" -- the marketing buzzword -- represents much of what the computer industry has been working toward for years: small, fast, and lightweight consumer end devices relying heavily on their persistent connection to a network (whichever one that may be) to provide information, storage, and processing on demand.

It's not something one can particularly trust or distrust.

I liken it to home PC security. You can't simply say the whole Internet is untrustworthy because it contains viruses, scams, and information of dubious authenticity. You take the good with the bad, protect yourself, and tread lightly.

There's only so much your antivirus software can do to protect you if you're recklessly clicking through everything, installing whatever fake plug-ins you're told to, and voluntarily submitting personal information when asked. Similarly, if you're uploading information to a single hosting service without backing it up somewhere, you're putting yourself at risk.

Unfortunately, Sidekicks are a risk.

Regarding the fiasco, Microsoft CEO Steve Ballmer recently said, "Non-Sidekick users, we are not earning their trust back, but I think people are going to say, 'Hey, look, show me what you are doing to ensure this does not happen to me.'"

The devices have been compared to thin clients, where there is no mechanism in place for a client-side backup of all the data stored on the Danger/Microsoft servers, so users are given only minimal control over their data.

This degree of control is one of the critical elements in establishing confidence in cloud-based services.

"All risk cannot be removed," said Erik Laykin, leader of the Global Electronic Discovery and Investigations group at independent financial advisory and investment bank Duff & Phelps. "After all, we are still relying on hardware and software systems, both of which can fail."

"Both organizations and individuals need to consider some factors [when] turning their data over to a third party...What are the risks we opening ourselves up to? Where is this data going to be (US, India, China)? What kind of backups and redundancies are in place? Who are the third party's subcontractors?" Laykin said.

Risk-benefit analysis, therefore, is another critical element.

"My clients are corporations and law firms, but consumers have to be aware of the risks too. For example, as an individual, I outsource my backups to online storage services like Carbonite. Do I accept that as the final solution? No. I make a physical backup because I can't always rely on my third party. Sidekicks unfortunately lack flexibility in this area."

But because we cannot rely solely on a cloud service, does that mean that ABC News was right to suggest we can't trust the cloud?

Laykin said no: "I'm getting on a plane tomorrow, and we all know the plane can drop out of the sky for a thousand different reasons. But does that mean I can never trust airlines to get me there safely? Of course not. We trust that the airline has mitigated those risks, but protect ourselves and our families with insurance in the unlikely event that something happens."

It's situations like the Sidekick data loss that could ultimately restructure the way we deal with data protection.

"For the last ten years, the insurance industry has provided coverage to mitigate data loss, which usually addresses hosting data," Laykin said. "Data insurance for individuals may not be something widely available today, but I envision policies like that being part of the average user's portfolio twenty years down the road."

Today, groups like Laykin's can assess the risks for companies looking to move their data into the cloud, but incidents like the Sidekick outage could trigger the demand for stricter regulations. "We may see a day in the future where organizations and companies like Microsoft or Google are rated on how safe and accessible their data is. Right now, it's still a brave new world in terms of third-party data management."

Office 2010, SharePoint public betas for November, VS 2010 Beta 2 Wednesday

Scott M. Fulton, III — Mon, 19 Oct 2009 14:56:58 -0400

By Scott M. Fulton, III, Betanews

During an industry event whose original purpose was to concentrate on SharePoint 2010, Microsoft's collaborative server product, CEO Steve Ballmer revealed that his company is making ready an official "public beta" of Office 2010, the applications suite for Windows.

The most likely timeframe for such a release would be during PDC 2009, Microsoft's annual developers' conference now scheduled for the third week of November in Los Angeles.

No further details on Office 2010 were revealed, but this much we do know: It is Microsoft's plan now to build O'10 in such a way that OEMs can install a full product on PCs for sale, which will come up in a limited "Office Starter" functionality mode until customers purchase the product online or through retail outlets. This reduced functionality mode will not be the same as Office Web Apps, which will be available to all users for free, even if they're not licensed Office users and even if they're not using Windows.

What we will probably learn at PDC next month is how fully-featured Office Web Apps will be compared to Office Starter. Given the company's recent developments, it's seeming more likely that OWA will be geared toward users who already have access to Office documents, even though they may not have Office itself. This way, editing functions may be limited to making changes to content that has already been created -- especially in the case of Excel, whose spreadsheets in the early public beta enabled users to view charts but not create them, and enabled them to use cells that already had conditional formatting but not create new ones with the same feature.

The current Office 2010 Technical Preview is a relatively stable product; extensive Betanews tests involving the Word, Excel, and Outlook components have revealed numerous cosmetic problems, but surprisingly few crashes. However, there are relatively few new features in the Technical Preview; what Outlook 2010 appears to support that Outlook 2007 did not, appears to be focused on new functionality that will be provided by SharePoint 2010.

SharePoint's functionality this round appears to be extensively reworked, especially around the idea of letting businesses create rich, socially active networks for their employees. These networks will be better integrated with Office applications, by making SharePoint sites into destinations directly accessible through the apps themselves, rather than through Internet Explorer.

New services available through SharePoint servers will effectively make them into hosts for Office Web Apps. Demonstrations at today's conference in Las Vegas included the presentation of an Excel spreadsheet through a SharePoint site, whose contents were viewable directly through the Web browser. The SharePoint server, in this case, is the one hosting the Excel display functionality, eliminating the need for many employees to have direct access to Office or Excel.

The two buttons on the demonstration site's toolbar, "Edit" and "Open in Excel," suggest two levels of functionality for editing purposes. The "Edit" functionality appears to be a way to launch the complete Excel Web App through the SharePoint site, which may offer some (perhaps not all) online counterparts for Excel editing capabilities. Microsoft is about nothing if it isn't about leveraging; and in this case, Silverlight is being used as the trestle for connecting platforms. The rich media support for new SharePoint 2010 sites will be provided by Silverlight -- that will apparently be the component that enables higher-definition video than is currently capable, as well as richer controls. The ability for developers to build SharePoint sites is being built into Visual Studio 2010; and developers' first taste of this ability is the primary focus of VS 2010 Beta 2. That build went live on MSDN this afternoon, along with .NET Framework 4 Beta 2. Both betas will become generally available Wednesday afternoon.

How do you define performance?

Carmi Levy — Thu, 15 Oct 2009 16:26:55 -0400

By Carmi Levy, Betanews

System performance is an interesting concept; everyone seems to define it differently. To some, it involves chewing through a complex spreadsheet. To others, it's how fast a 3D video sequence can be rendered, or how easily Web pages are served up.

Call me a rebel, but after years of living off of a BlackBerry, my thinking has evolved. As much as I focused on megahertz and gigahertz for much of my computing life, the most important criteria for me these days are how fast the thing turns on, and how long it stays on before I have to recharge it.

It goes against everything we grew up with, of course, as the PC industry has always been obsessed with the electronic era equivalent of brute force horsepower, rather than eco-friendly efficiency metrics. But when I'm sitting in a deserted airport terminal with an hour's worth of work to power through and only 45 minutes of remaining battery life, the last thing I'm thinking about is processor performance.

Machines (and uses) have changed

However we choose to define performance, we've always assumed that performance measurement benchmarks were run on machines that had already been powered on and booted up. We also assumed that these machines were plugged into a power outlet. These days, these assumptions are laughable, because outside of cubicle-dwellers sweating over dusty desktops for their corporate overlords, few of us leave our computers on 24/7/365. We take them with us to places like Starbucks, where finding a power outlet is about as challenging as locating the lost city of Atlantis, and we don't have the luxury of time to sit and wait while our machines boot up.

This shift in performance metric priority is being driven by two major changes in attitude and technology:

Green matters. Conserving power didn't matter much when 30-pound desktops ruled the planet and no one paid attention to how much all that electricity cost. Now that we've mobilized, disconnected, and learned the value of power-performance curves, the cost and availability of power are suddenly much more important to us. It isn't feasible to leave a laptop or other mobile device perpetually powered on. Even if we didn't run the risk of frying our laptops inside their padded cases, their three-hour battery life would leave us high and dry before lunchtime.
Software deployment has evolved. Corporate software distribution methods that required conventional desktop machines to be powered on at night, are yesterday's news. Today, commodity chipsets, firmware, and monitoring software allow machines to be remotely -- and automatically -- woken up, updated, and shut down. Web-based remote management solutions keep mobile machines in top form as well. In many cases, cloud-based software infrastructure is largely obviating the need to update client code in the first place. One reason that everyday users neither know nor care when Google Docs adds a new feature is because the software no longer makes a big deal of it. Windows tends to announce when it's been patched or updated, but Web apps typically don't.

Beyond eco-friendliness, better, smarter client hardware, network control, and an increasingly thin software stack, lies another simple truth: Today's mobile machines lead very different lives than more sedentary desktop hardware we all grew up with. We shut them down and hibernate them before shlepping them from one place to another, and when we get to our destination and open the lid, we expect things to happen quickly. The more dynamic nature of the modern workplace, coupled with the transition of mobile handsets from basic voice devices to full-featured computing devices, has radically altered our expectations of how these devices -- mobile or not, pocketable or not -- should work.

Thankfully, the industry is tentatively taking notice. The turtle-slow speed with which the average computer wakes itself up, freshens its desktop, and presents its happy face to its owner, is finally getting some attention from hardware vendors. HP's QuickWeb and Dell's Latitude ON, for example, boot into reduced-scale environments for quick access to simple apps like e-mail and contact management. Beyond launch performance, once these machines are on, the amount of time they actually remain on is also finally becoming a key concern for hardware makers.

I'd like to ask them what took so long, but that would be cliché. What matters is that this real-world performance metric is no longer being ignored. And as operating systems initially developed for handsets such as Google Android start showing up on netbooks and other lower-end machines, end users will increasingly expect these larger devices to offer the same kind of instant-on, battery-friendly experience provided by more traditional handhelds.

Not so fast

As heartening as it is to see vendors finally wake up to power-on and battery life performance, we're not quite up to speed yet in terms of standardizing and normalizing the messages they send to the broad base of consumer and enterprise buyers.

For example, Nokia claims its new Booklet 3G will run for 12 hours before it needs to be recharged. In the absence of standardized power-based performance metrics, however, it's difficult to take Nokia at its word. Is it 12 hours of watching a movie? Is wireless on or off? If so, is it Wi-Fi, 3G or both? Since vendors have no established baseline for these performance claims, they're free to say pretty much whatever they want.

As a result, they'll continue to use rather useless metrics (like how many cells their batteries have) to flog their machines. But computer vendors making a big deal out of six- or nine-cell batteries aren't much removed from automakers hawking the size of a given vehicle's gas tank. In both cases, we still don't know how far the thing will go. In the real world, we continue to lack industry-accepted parameters akin to the EPA's fuel economy standards that allow computer buyers to make apples-to-apples comparisons.

Change often comes to an industry when consumers demand it. And as we spend ever greater stretches of time using battery-powered devices far from home, the onus is on us to challenge vendors to get straight with realistic methods of measuring such real-world performance parameters like battery life and power-on availability. I'm sure I'm not the only one who looks forward to evaluating comparable machines based on performance figures that weren't first invented by the marketing department.

Carmi Levy is a Canadian-based independent technology analyst and journalist still trying to live down his past life leading help desks and managing projects for large financial services organizations. He comments extensively in a wide range of media, and works closely with clients to help them leverage technology and social media tools and processes to drive their business.

Danger signs: Now how secure does the cloud look?

Carmi Levy — Mon, 12 Oct 2009 17:57:06 -0400

By Carmi Levy, Betanews

There are service outages, and then there are service outages. T-Mobile customers who carry the Sidekick smartphone are learning the hard way that there's a major difference between having no access to a service for a little while and losing every contact, calendar entry, and related shred of personal data they've got.

In the not too distant past, Google, Twitter, and Facebook have all experienced basic, quaintly simple service outages. Despite the headlines and general chaos associated with each incident, the bottom line impact was never all that onerous: When service returned, so did their users' data. For the most part, users were given an easy excuse to take a few hours off. And with the exception of Google's subscription services, most were free, so folks couldn't argue that they weren't getting their money's worth.

More than a free service

Microsoft's experience isn't turning out as charmed, and it wouldn't surprise me if some of the folks behind its 2008 purchase of Sidekick maker Danger might be rethinking the $500 million deal in the wake of last week's worse-than-usual outage. When service was restored and countless users (Microsoft and T-Mobile still aren't fessing up to actual numbers) realized their devices had been wiped clean, Microsoft was forced to release an unprecedented mea culpa admitting data had been lost and would in all likelihood not be recovered. More embarrassing for the companies involved, Microsoft implored users to keep their batteries in place and avoid resetting their devices or allowing them to lose power.

However you slice it, this is not a happy place for anyone. While it's easy to assume Microsoft's and T-Mobile's customers are the real victims here, the sad truth is these very clients shoulder at least part of the blame for losing their stuff. It may sound harsh, but users who rely so heavily on a vendor that they neglect to implement their own disaster recovery plan shouldn't complain too loudly when said vendor drops the ball. Although in this case Microsoft and T-Mobile were accountable for the service itself, data ownership always resides with the customer. While the peculiarities of the Sidekick dictate that much of the data resides in the cloud, end users remain ultimately accountable for their information.

Sadly, many of them are learning a hard lesson about the value of local syncing. Whatever mobile device or OS you're using, this should be a wakeup call if you're not doing the same.

A cloudy question

This debacle doesn't just force this particular service into question. More ominously, it challenges the very notion of cloud-based services at a time when their takeup rate is accelerating. The fundamental trust that we have in such services -- that a provider that specializes in large-scale deployments like this could absolutely never lose our precious data -- has been thrown into question. Suddenly, keeping things stored on our rickety old hard drives, or at least backing them up there, may not seem like such a bad idea. Any way you slice it, it's a backward step in the march toward the cloud.

To its credit, Microsoft is doing everything it can to make the best of an unfortunate situation. It's apologized for losing customer information, it's scrambling to recover what it can, and it's offering up a free month of data service. While customers who have lost it all may disagree, this is a textbook response to this kind of situation. And as the vendors involved strive to save whatever face they can, it's fair for current and prospective customers to feel burned by a service whose monthly subscription fees implied a certain trust relationship. More than a free service like Twitter, which when it inevitably goes dark users can simply shrug their shoulders in response because they're simply getting what they've paid for (namely, nothing), a service like T-Mobile's that comes with a monthly bill can't simply rely on shoulder-shrugging users when the worst happens.

[Original photograph by Carmi Levy]

A case of bad timing

All this must weight heavily on Microsoft as it prepares to release its core cloud-based environment, Windows Azure Platform. While Microsoft was hardly involved in the architecture decisions made years before it acquired the Danger unit, the brand association is anything but positive as Microsoft takes its biggest step yet toward a Web-enabled services model. Convincing customers still comfortable with the notion of physical servers in tangible data centers that they should toss their data into infrastructure owned and managed by some unseen entity just got a lot harder.

Google, Salesforce.com, and other cloud-based vendors -- free or not -- are doubtless also feeling Microsoft's pain, because they all know full well that this kind of thing can happen to them, too. The industry clearly has a long road ahead of it as it seeks to balance the compelling capital and operational advantages of Web services with the never-ending need for customers to take an active role in securing their data.

That road will be difficult indeed if vendors ignore the need for this form of partnership. Despite their passion for making their new generation of Web-based services as worry-free as they possibly can, no amount of technology can ever remove the need for personal and corporate accountability from the equation. Vendors that market themselves as the answer for customers who can't be bothered to pay attention to their own data need a not-so-slight attitude adjustment.

For their part, customers also need to begin challenging cloud-based services vendors with specific questions revolving around how data is secured, backed up, and restored. Before signing on the dotted line, they should ask about what tools and processes the vendor makes available for customers to self-serve their own backups. Even if it's as simple as a basic export to a .CSV file, with the right support from their vendors, customers can set up automated processes that ensure they can keep going even if the service itself does not.

Vendors that don't help customers help themselves will be quickly eclipsed by those that do. And when the worst happens and a vendor-caused meltdown takes data with it, customers that don't step up to the plate will have no one to blame but themselves. Welcome to the new reality of the cloud.

'Amateur' Linux IBM mainframe failure blamed for stranding New Zealand flyers

Scott M. Fulton, III — Mon, 12 Oct 2009 11:20:29 -0400

By Scott M. Fulton, III, Betanews

12:05 pm EDT October 11, 2009 · The president of a design firm that specializes in data center power efficiency, and that was working on a new design last year for the Auckland-based data center that failed Friday morning, told Betanews today that even if changes were being made to that data center, if both the original design and the changeover plan were implemented properly, the data center failure would not have happened.

"What seems strange about this incident is that they are blaming it on a generator failure during testing," stated California Data Center Design Group President Ron Hughes, whose organization was not responsible either for the data center's current design or the changeover. "If this failure did occur during testing, the question I would ask is why didn't the redundant generators assume the load or why didn't they just switch back to utility power."

Though Hughes has no specific knowledge of last Friday's incident, his insight does shed more light on the situation.

"A properly designed Tier 3 data center -- which is the minimum level required for any critical applications -- should have no single points of failure in its design. In other words, the failure of a single piece of equipment should not impact the customer," Hughes told Betanews. "A generator failure is a fairly common event, which is why we build redundancy into a system. In a Tier 3 data center, if you need one generator to carry the load, you install two. If you need two, you install three. This is described as N+1 redundancy. It allows you to have a failure without impacting your ability to operate...In a Tier 3 data center, it should take 2 failure events before the customer is impacted."

The CEO of Air New Zealand -- one of the few major CEOs anywhere to have been elevated to the top post from a CIO position -- expressed his disgust last weekend over what he describes as the poor handling of a data center failure at his airline's outsourcing partner, IBM. Rob Fyfe's e-mail, made public by IDG's Randal Jackson, excoriates IBM for its handling of a systems outage that took place at 9:30 am local time Friday morning, and that lasted for at least six hours.

During the entire time, ticketing, baggage handling, and traffic rerouting procedures for the entire airline were at a standstill, causing chaos for airports there. This at a time when Air New Zealand was engaged in a public showdown with its chief rivals there, Pacific Blue and Qantas subsidiary Jetstar, challenging them to meet ANZ's standards for flight punctuality. "In my 30-year working career," Fyfe told his colleagues, "I am struggling to recall a time where I have seen a supplier so slow to react to a catastrophic system failure such as this and so unwilling to accept responsibility and apologize to its client and its client's customers...We were left high and dry and this is simply unacceptable. My expectations of IBM were far higher than the amateur results that were delivered yesterday."

Indeed, even as of this morning, IBM New Zealand has issued no public statements. The data center failure apparently affected all of IBM's customers in the region, not just the airline, although there is no word yet as to the identity of those customers or the extent of damage to their operations.

The move to outsource data center operations to IBM appears to have happened partly under Fyfe's watch as CIO, and was heavily touted by the time by IBM's marketing literature as a "design win" for mainframe-based Linux. Though some mainframe database operations for ANZ came online as early as 1999, the most lucrative move came in August 2002, when the airline replaced its mid-range Windows NT-based in-house network made up of 150 Compaq z800 workstations, with a single eServer zSeries Linux outsourced mainframe hosted by IBM Global Services. The airline's CIO at the time of the move, Andrew Care, said maintaining the outsourced zSeries would cost his airline 30% less in maintenance fees, and save $600,000 in software licenses.

The migration was seen as a huge loss for Microsoft, whose NT operating system was already well on its way to having been branded a failure for mid-level networks.

IBM established the global airline industry standard software for transaction processing as far back as 1960, in a joint project with American Airlines called the Airlines Control Program, which made possible the original, groundbreaking Sabre system. Since 1979, IBM has sold other airlines a commercial version of this system, called Transaction Processing Facility (TPF).

To this day, the transaction format used by airlines everywhere is based on ACS' half-century-old protocol. It isn't the format that has needed evolution, but rather the software that runs it; and IBM itself has been the key innovator here, developing a new class of software this decade called the Airlines Control System (ALCS). Originally seen as a mid-level alternative to a higher-class TPF system for smaller airlines that couldn't afford big iron, ALCS -- a TPF emulator -- now runs on bigger iron, thanks to the evolution in hardware as well.

Air New Zealand was one of ALCS' biggest customer wins in August 2002. Up to now, the airline has been one of ALCS' more active supporters, contributing a big chunk of new requirements for the software's latest version, according to literature from the UK-based ALCS User Group.

At this point, Air New Zealand may have too much investment tied up in the software to be in any position to migrate its applications to an IBM competitor -- if there even really is one in this field. But the airline's problem may not be with so much with the software but with its current host.

According to an ANZ group general manager cited in local radio news reports, the offline incident was traced to a single generator failure at IBM's Newton Data Center in Auckland. Usually data centers have redundant power sources, and normally the Newton center would not be an exception. An August 2008 article in Data Center Journal by the designer of new energy-efficient data center power generators with redundant sourcing, specifically mentioned the Auckland center as one of his customers at that time.

"I've seen numerous references recently to reducing the amount of redundancy as a way to achieve higher energy efficiency," wrote engineer Ron Hughes, president of California-based Data Center Design Group, referring to his Auckland data center project. "While I have no doubt that it is true, it may not be in the long term interest of the client. Data center outages can be career changing events. That extra redundancy may be the difference between a component failure with little impact and a system-wide outage."

Current ANZ CIO Julia Raue has been overseeing an innovative new information systems project at her airline, which has involved the creation of customizable self-serve ticketing kiosks, which customers themselves can change online using selectable widgets to suit their airport demands. In an interview with CIO Magazine last month, iGoogle was credited as a design inspiration for the self-serve system. But the entire system revolves around the zSeries mainframe, whose uptime last week appeared to have revolved around a single faulty generator.

While CEO Fyfe certainly has understandable reasons for wanting to abandon IBM, with his entire information strategy dependent on the move ANZ made in 2002, he may not have many alternatives open.

Top 10 Windows Server 2008 R2 Features #9: Processor core parking

Scott M. Fulton, III — Fri, 25 Sep 2009 13:06:12 -0400

By Scott M. Fulton, III, Betanews

When you think of kids on the playground playing Star Trek make-believe, you see the guy who plays Scotty inevitably being shouted at to increase or decrease the power, and then the guy putting on his best (or worst) Scottish accent and complaining back to the captain about how it canna be done, she can't take this abuse much longer or we're all genna bloe! Powering up and powering down is the most common task that amateurs think of when they consider the role of an engineer running a big machine.

And yet up until very recently, servers have existed in a perpetual "on/off" state -- they're either turned on and consuming the energy they've been designed to consume, or they're off and your data center is offline. Only in the last few years, with the introduction of the multicore era coupled with the sudden ubiquity of virtualization, has there been the notion that you can move the entire serving job at any one time to the most efficient processor available. New CPU technologies like Intel SpeedStep have created the opportunity for administrators to eliminate the problem of processor latency by turning off entire cores when they're not in use.

Beginning with both Windows Server 2008 R2 and Windows 7, Microsoft will build that capability directly into the operating system, with the addition of new parameters to the powercfg command-line tool. Believe it or not, powercfg made its debut in Windows XP SP2 (yes, the client, not the server), as a tool for reporting the power profile of systems such as laptops that frequently suffer from low battery life.

In WS2K8 R2, powercfg takes on the role of a PowerShell cmdlet (though it can still run from CMD.EXE; if Microsoft's engineers had really gone all out with the cmdlet idea, they'd have created an alias "set-power" in accordance with PowerShell's grammar guidelines). This tool will now communicate directly with the operating system's new processor performance management (PPM) engine, which has the capability (assuming the underlying CPUs support it, and both new AMD Opterons and new Intel Xeons do) to turn down or off minimally used, underutilized, or unused processor cores.

BETACHECK
For more:

"Powercfg and Power Efficiency Diagnostics Reports" by Daniel Nerenberg. Article on TheLazyAdmin.com.
"Improved Data Center Power Consumption and Streamlining Management." Marketing literature from Microsoft.com.
"Seeing Core Parking in Action" by Matt McSpirit, from his VirtualBoy blog.
"WMI Via PowerShell, Remotely and Repetitively" by Scott Fulton. From the Windows Server Reference Guide on InformIT.
"Performance Tuning Guidelines for Windows Server 2008 R2." Documentation in .DOCX format from Microsoft.com.

For now, in lieu of a management console plug-in, and also instead of group policies (which Microsoft had promised, but which conceivably a third party may write using an SDK), R2 will provide administrators with a simple syntax for setting limits on when and how PPM can park unused cores or reduce processor frequency. You'll find it easier to use powercfg from PowerShell because you'll need to use the get-processor cmdlet to locate the GUID of the CPU to whose settings you'll be referring; you'll be referring to WMI, and it's just easier to create a variable that grabs the WMI object, rather than retyping the GUID from scratch.

As an obscure, but obtainable, new piece of Microsoft documentation explains it, "The PPM engine chooses a minimum number of cores on which threads will be scheduled. Cores that are chosen to be 'parked' do not have any threads scheduled on them and they can drop into a lower power state. The remaining set of 'unparked' cores are responsible for the entirety of the workload (with the exception of affinitized work or directed interrupts). Core parking can increase power efficiency during lower usage periods on the server because parked cores can drop into a low-power state."

Using powercfg, you create a power scheme (which is separate from a group policy object), and you give that scheme a name like ordinary_scheme. You may have already been doing this before for regular system power management. But now, the scheme adds a sub_processor parameter with which you designate the integer percentage of cores in your system (remember, a single R2 instance can now support 256 logical cores simultaneously) that can be regulated through PPM. If it just so happens that all the cores being powered down reside on the same die, PPM can take the next logical step and power down the whole chip, if necessary. In this type of regulation -- again, if your CPUs support it -- the various range of supported frequencies, known as ACPI p-states, can be tapped into to reduce the frequencies of underutilized cores at will, and power them back up again when the load gets heavier.

Early beta testers who had not been informed of this new feature had actually been alarmed by the discovery that several of their processor cores were not registering heartbeats in Resource Monitor. What's the matter with Resource Monitor, they asked? Nothing at all, it turns out...and with the shipping version, Resource Monitor will allay admins' concern by explicitly reporting shut down cores as "Parked," with grey text.

FOLLOW THE WINDOWS SERVER 2008 R2 COUNTDOWN:

#10: Boot from virtual devices by Scott Fulton

For AMD, keeping it too simple may be too stupid

Carmi Levy — Mon, 21 Sep 2009 16:59:01 -0400

By Carmi Levy, Betanews

The "Keep It Simple, Stupid" rule works in business as well as in everyday life, primarily because it forces us to focus on the one or two basic issues we need to make the right decision. My kindergarten teacher shortened it to "KISS," perhaps in the interest of simplicity. When you're driving a car: Know where you are, know where you're going. Leader of the free world: Keep your hands clean, know what you're aiming at. Reprogramming your iPod/in-vehicle audio system interface: Pull over, let the semi pass you.

More data, please

The computer market never got the memo -- or if it did, maybe no one could make sense of it. This is a good thing, though, because when you're investing in a technology, more data is always preferable to less. Whatever parameters you're measuring -- processor architectures and speeds, types of memory, hard drive spin velocities and capacities, or display resolutions and refresh rates -- the more you know about what's going on under the hood, the more likely you are to make a sensible buying decision that you can live and work with.

Now all of a sudden, AMD wants to dumb this process down because some people find it too confusing. And that scares me more than a little. The company's just-announced Vision program aims to replace the information-stuffed process of buying a computer with one based on three categories: Vision, Vision Premium, and Vision Ultimate, each of which corresponds to a particular end user need. Plain old Vision machines define the low end, and will do the trick for Web browsers and basic game players. Vision Premium might be suitable for gamers interested in something more challenging than Solitaire or Sudoku, as well as HD video. Vision Ultimate addresses the most demanding gamers and content creators. Early next year, Vision Black-based machines that target high-end users will join the party.

There's an eerie resemblance here to "Vista Premium" and "Vista Ultimate," but I suppose AMD didn't get that memo, either.

A slightly blurred vision

After a generation of PC marketing being driven by and targeted at technically inclined folks who used the often arcane parameters of hardware capability to measure bottom line performance and calculate value propositions, my mother-in-law is now calling the shots. If AMD's strategy plays out, consumers who don't understand the difference between level 2 and level 3 cache will now find it easier to choose the right solution.

I get that. I've seen my mother-in-law's eyes glaze over when I try to explain the benefits of 7,200 RPM drives versus 5,400 RPM ones. She doesn't much care that a quad-core processor lets her edit photos and video more smoothly than a dual-core one, and she wishes I'd stop trying to get her to upgrade her webcam from her current, lame VGA device that washes out in the low light of her home office to an HD-capable one that doesn't make her and my father-in-law look like stop action ghosts. While I quote numbers to illustrate the benefits, she waits for Wal-Mart to drop the price on a cheap machine she'd like to install in the kitchen.

What about the rest of us?

So she'll appreciate AMD's gesture. It's the rest of the world I'm worried about. Taken to its ultimate extreme, simplified computers will make it virtually impossible to make real, side-by-side comparisons. Want to know if $699 is too much? Tough. No longer will savvy customers be able to laugh off the know-it-all salesperson's claim that no one else can match their price. How can you appreciate a product's capabilities when its basic guts are deliberately hidden from view? Simplified marketing may make life easier for the uninitiated, but does the message have to be so diluted that folks with more than a little knowledge can't dig a little bit deeper on their own?

When otherwise educated customers are forced to buy based on vague notions of role- or task-matched performance, the entire process begins to look suspiciously like the one from the most recent silent auction at my children's school. In amongst the various items up for bids -- the gift certificates, logoed clothing, and small electronics -- was an envelope with a single word printed neatly across the seal: "Surprise." Based on similar experiences from past years, there was an even chance that the envelope contained a booby prize, such as a yellowed old bumper sticker or a collection of discarded scripts from last year's school play. Well-meaning parents, hoping this was a non-booby-prize year, nevertheless bid the mystery envelope up to $50 before the night ended, and ended up with a $25 gift certificate at a local coffee shop for their donation.

I'd hardly expect someone laying down hundreds or thousands of dollars at a profit-seeking technology store would be as understanding as these involved parents were. If the recession has taught us anything, it's the value of a dollar. Computers are not sold in sealed envelopes, and customers aren't supposed to guess, even vaguely, at what might be inside.

A better kind of simple

Intel already gets the message. While it arguably kicked off the simplification parade with the Centrino brand, and has been gradually moving away from feeds and speeds ever since, shoppers who understand GHz and Gbps can still pull out info sheets at the store and geek out to their heart's content. To its credit, Intel recognizes that today's PC buyers come in all sorts of shapes, sizes, and levels of knowledge. And a one-size-fits-all, universally-simplified marketing strategy fails to give the tech-savvy buyer sufficient background to make the right decision.

I appreciate why AMD is going simple. Data released last week by iSuppli shows its market share down by 1.4% to 11.5%, while Intel's is up by 1.5% to 80.6% -- its highest since 2005. Companies in decline need to do radical things to change their fortunes. But ticking off the large percentage of the market that still has a clue isn't a great way to start. Even my kindergarten teacher would agree.

Like HP, Dell also acquires a Perot empire for enterprise services

Tim Conneally — Mon, 21 Sep 2009 14:09:00 -0400

By Tim Conneally, Betanews

Though Dell has extended its brand to consumer electronics of all sorts, the company's latest drive is straight into enterprise services, a segment of the IT market which has helped HP retain a competitive edge on Dell in hardware sales.

Today, Dell announced it will be acquiring Perot Systems in a $3.9 billion all-cash transaction expected to be completed in January.

"We've really been transforming Dell in a number of ways," CEO Michael Dell told CNBC this morning. "One of the ways we're doing that is by becoming a solutions integrator, providing the best value solutions to help companies run a very efficient enterprise architecture...and the addition of the Perot Systems team here will really allow us to extend that to a much broader set of customers."

Indeed, in the last year and a half, Dell has grown its IT services portfolio by acquiring IP Storage Area Network company EqualLogic, and then later enterprise e-mail service company MessageOne. This morning, Michael Dell said the market for services like those from EqualLogic has grown more than four times in just over a year.

With the addition of Perot Systems, Dell is expanding its IT consulting business; and the company today said this acquisition will let Dell offer a broader range of IT services and solutions by optimizing how they're delivered, extend Perot Systems' capabilities and customer segments, and supply Dell systems to more Perot customers.

This is important to note because nearly 80% of Dell's revenue comes from enterprise hardware sales; and the company will need a diversified revenue stream to outgrow chief competitor HP.

HP, currently the world's leading computer manufacturer, made a similar acquisition last year when it picked up EDS for $13 billion. EDS was founded in 1962 by none other than H. Ross Perot, who 26 years later founded Perot systems in the very same town of Plano, Texas.