flautario.com

commentary

It’s just a matter of time before Ubuntu is crowned “enterprise ready” by one of the major ISVs. Will it be able to maintain its popularity once it is popular with enterprise buyers?

Ubuntu plays an increasingly important role within the larger Linux market. According to a new white paper from IDC [PDF], Linux is big business and is ready for prime time, with IDC forecasting overall spending on hardware, software, and services for Linux to increase 25.2 percent annually through 2011, particularly at the expense of Unix:

Increasingly, deployments of the Linux server operating system are expanding from infrastructure-oriented workloads to more commercially-oriented workloads such as database, enterprise resource planning (ERP) software and other general business processing, workloads that historically have been the domain of Microsoft Windows and Unix. Where once Linux was seen by customers primarily as a low-cost infrastructure solution, it is now increasingly viewed as a solution for wider and more critical business deployments.

The question in my mind is therefore, “Which of the big-three Linux vendors is going to dominate the market?” Red Hat is the obvious first choice, but I think there’s a serious spoiler in the Linux market, and its name is Ubuntu.

Novell’s SUSE, with its rising Linux revenue, should perhaps be the contender for the throne, and I believe it will continue to leverage its Microsoft relationship to the maximum advantage. But Novell (and Red Hat, for that matter) is missing something that Ubuntu has in spades:

Community momentum.

This isn’t just about downloads. As I’ve seen within my own company, where Ubuntu is our fastest-growing Linux distribution, community momentum increasingly translates into revenue momentum.

If you’re Sun, IBM, Oracle, HP, or other big companies in or surrounding the Linux phenomenon, Ubuntu has to look mighty tasty. For Oracle, Ubuntu has proved that it can compete head-to-head with Red Hat. As movement from Sun, IBM, and others is starting to show, OEMs and ISVs are making room for Ubuntu in their certification plans. Given enough community momentum, the certifications will follow.

Oracle need not continue attempting to replicate Red Hat, the market leader. It could buy into Ubuntu and foster a new market leader.

Ubuntu has done all of this largely on its own. Consider what would happen if an IBM or Sun seriously threw its weight behind Ubuntu, banishing the “It’s not enterprise-ready” myth forever? I suspect we’d see widespread, immediate enterprise adoption of Ubuntu. Why? Because Ubuntu is already being used widely within enterprises. It’s just not being allowed to carry mission-critical enterprise applications yet, because those are reserved from the “enterprise Linux distributions” like RHEL and SUSE.

In short, the only thing Ubuntu is missing is a big brother that will officially declare, “Ubuntu is enterprise class.” It’s already there, but needs the certification by A Big Company. IBM, Sun, and others all have their own reasons for crowning it such.

It’s just a question of when and who. My money is on Sun, as IBM still wants to pretend it’s a neutral Switzerland in the Linux market. Oracle could also make a move, but I’ve not heard any rumblings on this one.

Would such an action make Ubuntu the winner in the Linux popularity contest? Of course not. If anything, it might stifle Ubuntu’s popularity, just as Red Hat’s transition of its Linux distribution into Fedora and RHEL caused consternation in its community (but not in its stock price or annual sales :-). Ubuntu “growing up” might be the worst thing that ever happened to its community.

But I think Ubuntu’s increased commercialization is not up for debate. It will happen. It’s just a question of whether Ubuntu can “grow up” on Mark Shuttleworth’s terms, rather than on crass materialistic terms. I think it can.

As more companies move to virtual machines and blade servers to reduce space and costs, there are few mentions of the downside or “dark side” of virtualizing hardware and operating systems. I spoke with Bob Waldie, CEO and founder of Opengear, an open-source out-of-band management services provider, about the problems associated with the adoption of blades and VMs.

Q: So why do you think that virtualization is such a problem?
Waldie: The dark side of virtualization is the complexity of the environment it creates. While server virtualization improves flexibility and asset utilization, it also adds complexity to the environment, specifically by adding an extra hypervisor layer to the operating and management load and creating complicated virtual appliances and virtualized I/O. For example, when the enterprise VPN network is routed through software VPN appliances running on virtual servers and then migrated to virtual machines, the connection of the virtual machines back to the LAN presents a new layer of management challenges.

Well, what are these new management challenges?
Waldie: For the vast majority of enterprises, virtualization coexists with physical deployments and data center managers need tools for managing both physical and virtual environments. However, in addition to needing new virtual management tools, the added network complexity means the tools that previously used to manage the physical infrastructure pre-virtualization aren’t appropriate. Adding to this management issue is the increase in disaster sensitivity that comes with consolidation. While prevalence of infrastructure outages may not increase, the consequence of a hypervisor or blade failure will, so managers have to find and implement a completely new set of tools while under incredible pressure to avoid any downtime or IT issues.

Even given additional management issues, isn’t there a real benefit to the consolidation that virtualization brings?

Waldie: Sure, there are lots of reasons that people give for virtualizing a data center with the big one being increased infrastructure utilization as well as load balancing and power management, simplifying scheduled maintenance, and improving disaster recovery. But we’ve seen significant barriers to delivering such promises, and there are real costs in moving to more virtual environments.

Do you mean costs beyond the problems raised by managing virtual and physical machines?
Waldie: Yes, organizations are actually cutting their ROI on previous infrastructure management tools, and increasing upfront energy costs. For example, there is currently around $3 billion in KVM switches installed in the racks in enterprise data centers, and these tools are being progressively disconnected from the systems they used to manage. Server virtualization isolates the KVM switch from operating system and application. Not only does the sys admin lose a management tool, the organization also loses its investment in the switch.

As far as energy consumption goes, as virtualization moves server utilization up from its historical floor of 10 percent to 20 percent, power consumption also rises. There is also more risk that these increasingly mission critical servers can become hot spots, which can degrade server life and performance. Blade servers are also a problem; built for density, they ramp the power dissipation in each rack but demand sophisticated direct cooling solutions to ensure the servers run smoothly.

So what kinds of solutions out there help solve these problems?

Waldie: There’s no integrated vendor-neutral solution that will monitor and measure power, direct cooling, control the physical and virtual servers that generate the load, and then enable the data center managers to load shed and balance power demand and power supply. On average, virtualized environments have 11 different platforms, technologies, and vendors present, and most proprietary tools can’t deal with this level or heterogeneity, making open-source tools like Opengear’s KCS6000 or Minicom’s KVM.net II a good, flexible fit.

Alltel is now selling the Samsung Muse. As we told you in January, the Muse is a slim flip phone in midnight blue. Though its rather generic design won’t stand out from the Samsung crowd, it offers more features than you might expect. Inside you’ll find a 2-megapixel camera, GPS capabilities with Alltel’s Axcess Mobile Guide preloaded, stereo Bluetooth, and a music player. And on an original note, it is compatible with Alltel’s new Celltop music-streaming application. The Muse is $269 if you pay full price, but you should be able to get it for $89 with service.

(Credit:
Alltel)

Samsung Muse

Kraus said that data center equipment vendors offer an option for direct current power supplies, which customers could choose for hardware upgrades or new installations. He said he expects to announce some partnerships with equipment companies in the coming months.

By going DC, a company can save up to 40 percent on its energy consumption on equipment and cooling, the company says.

Start-up Validus on Monday is scheduled to announce the availability of a DC-based electricity distribution system tailored specifically for data centers.

Can going back to direct current–favored by Thomas Edison–make data centers more modern and energy-efficient?

What Validus has done is make power distribution equipment suitable for “high density” data centers–that is, racks of servers or blade servers that consume a lot of electricity (and generate a lot of heat) in a relatively small space, he said.

(Credit:
Validus)

In practice, the lower power consumption may give corporate data center operators the ability to pack in more gear in the same amount of space, he said.

The idea of using direct current, rather than alternating current, to reduce power consumption has been around for a long time. The telecom industry relies primarily on DC power equipment, Kraus said.

The effect of using direct current is to reduce the number of voltage changes and conversions between AC and DC, which makes the overall system more efficient. Kraus said DC power is also more reliable and flexible.

Its product lineup includes a device that takes power from the grid and distributes it as high-voltage direct current. A distribution board acts as a point for wiring and to connect energy storage. And a power conversion unit steps down the voltage to 48 volts.

A power distribution model for supplying DC to servers and other data center gear.

The company, which raised $10 million in venture funding last December, has an initial Fortune 50 company customer that’s looking to reduce energy consumption at its corporate data center, according to CEO Rudy Kraus.

Ultimately, the company envisions on-site power generation at data centers, where solar panels, wind mills, or fuel cells supply direct current electricity into buildings with DC wiring.

For all its promise, a lack of DC equipment poses barriers to bringing DC power into data centers, according to a study by The Uptime Institute. The research firm last year listed a number of barriers to DC power in the data center, including monitoring equipment, appropriate IT equipment, and DC power-modeling software.

The E2 is still only good down to 5 feet and for only an hour at a time. It also has the same design and again records in MPEG-4 AVC/H.264. In fact the only really new bits are that it records at 60 frames per second, can now capture 8-megapixel still photos (up from 6) with 10fps sequential shooting capability (5fps), and it has face detection/tracking.

It’s available now in brilliant blue or pearlescent white at $399.99, just in time for the summer beach season winter sports season.

(Credit:
Sanyo)

We had mixed feelings about Sanyo’s first-gen underwater Xacti, the E1. While it showed some promise with its 5x optical zoom and occasionally good photo results, the overall performance was sluggish and the pistol-grip design wasn’t ideal. Sadly, it looks like the newly announced followup, the Xacti E2, doesn’t offer much of an improvement.

It’s when the
iPhone app store opens up next month that we’re really going to see geo-focused reviews sites and networks take off. Despite its lack of GPS (so far), a core component of the iPhone is location reporting (using a combination of Wi-Fi router mapping and cell tower triangulation). All of the Web 2.0 geo execs I’ve talked to are working on iPhone apps; many will be available on day one of the app store opening.

Being lazy, I favor the geo-focused sites that don’t require that I do any work. Everyblock (review) wins the lazy-geo award from me: It scans local news sources and public records and shows me what’s happening in my ‘hood. My participation with the site consists solely of entering my address. Outside.in (news) has a similar function, but its user interface is less clear.

Everyblock gathers hyper-local news and info.

The personal location-reporting sites (Loopt, Brightkite, Whrrl, etc.) require a change in behavior: I have to tell these apps who my friends are and where I am to get them to work right. Integration with existing social nets should help these products take off, but until people start hooking these apps into their network profiles, they are going to languish.

At least one geo site has a CEO who’s aware that you can grow your audience more by giving users a lot before you ask them to give anything to the site. Platial’s CEO, Diann Eisnor, recently relaunched her site with a new reader-focused interface, replacing a previous design that appeared to be made more for contributors. Platial, unfortunately, doesn’t have the rich data set of reviews that Yelp does, but it does a better job of displaying Yelp-like content. We can hope for a partnership.

Platial's new UI is great for browsing local info, but it needs a richer reviews database.

It’s entertaining to see all these new geo-focused sites trying to build out their social networks and their databases of local content. There’s still a huge disconnect between the sites that make data entry easy and the ones that do a good job of helping you find what and who you are looking for.

What I really want, though, is a geo-enabled Yelp, both on my desktop browser and in my mobile phone. Yelp has all the location data I could possibly want; it just doesn’t have a very good location-focused interface, or the capability to auto-locate me when I am on my mobile phone.

Now this is cool. I’ve been thinking lately that a web-based instant messaging (IM) service like Meebo would be all the better if it could be easily integrated into other applications by making itself available as open source. Perhaps my support team could then build it into their support portal, for example?

Well, the folks at soashable have done just that. They’ve basically built a Meebo clone and licensed it under the GNU General Public License Version 2.

commentary

We’ve got DimDim for web conferencing. Ringside Networks for social networking. SoashableIM for web-based IM. All the tools are there to build some killer, open-source based websites.

As data gets larger the slowdowns become more unbearable, at one terabyte you will notice data bandwidth slippage. At 10 terabytes, storage systems crawl. “At one terabyte the problem rears its ugly head and it gets worse every year,” said Ellison.

The hardware roll-out, which was cooked up in a partnership with Hewlett-Packard, is aimed at the emerging problem of moving data from hard drives to database servers. The storage server took three years of development with HP and has been tested for about a year with key customers such as Google.

SAN FRANCISCO–Oracle CEO Larry Ellison on Wednesday unveiled its first ever hardware product–a storage server with embedded software designed to work with the company’s databases and be used in a grid. The Exadata programmable storage server aims to put database intelligence next to each drive.

With the move, Oracle is copying Apple’s model to a degree. Ellison is arguing that combined hardware and software efforts can be more effective. Instead of the consumer market, where Apple’s secret sauce is tightly integrated in hardware and software for Macs and iPods, Oracle is coupling its database software with custom hardware to revamp data centers.

Ellison outlined query processing and how Oracle’s embedded software will handle query processing and other functions more efficiently. Oracle is hoping to sell its storage hardware as part of a grid. Drives will be searched in parallel also.

Originally posted in ZDNet’s Between the Lines.

Ellison, speaking at Oracle’s OpenWorld conference, said large databases are creating a fundamental problem: disk storage systems can’t cope with data that has to be moved off of drives to database servers. He called it a “data bandwidth problem.”

The Exadata storage server will be immediately available on Linux running on Intel, but Ellison noted other flavors for various platforms “are on the way.” Oracle’s move could be disruptive in the storage market and with players like EMC and IBM, since it can offer a joint-software hardware sale and leverage its HP’s partnership. HP and Oracle are also rolling out an “Oracle database machine,” designed for customers that don’t want to configure the systems. The initial machine has 168 terabytes of disk data and 64 Intel cores.

Intel countered that AMD and Nvidia are not willing to do the hard work that is necessary. “They could spend the time, engineers and money developing their own host controller spec,” the Intel source said. “In the past they have chosen to let us do the work and then benefit from the fruit of our labor.”

“We’re not doing anything differently now than we did with USB 2.0 and USB 1.1,” he added.

One possible reason for the frustration, the source said, is that Intel is “a little bit behind and that’s what might be causing some of the resentment. You could take the opinion that Intel is giving stuff out for free and people are complaining because (Intel) isn’t giving it out fast enough,” this person said.

In play is the USB 3.0 specification, a next-generation high-speed connection standard due in 2009. It is significant not only because all future PCs and devices will use connectors based on the standard but because it will offer 10 times the speed of USB 2.0–used in virtually all PCs introduced in the last few years–or roughly 5 gigabits per second.

Intel showed off a prototype USB 3.0 connector and an add-in card last year.

Intel formed the USB Implementers Forum in 1995 with other industry players, including Microsoft, “to support and accelerate adoption of USB-compliant peripherals,” according to an overview of the specification on the chipmaker’s Web site.

“The challenge is that Intel is not…giving the specification to anybody that competes with CPUs and chipsets,” said a source close to AMD who is familiar with the dispute.

The USB 3.0 specification is hammered out in the USB 3.0 Promoters Group, in which Intel is a working fellow. Other members include Microsoft, Hewlett-Packard, Texas Instruments, NEC, and NXP Semiconductors.

“If you have an incomplete spec and give it out to people, these people will build their chipsets and you’ll end up with chipsets that are incompatible with devices. That’s what (Intel) is trying to avoid,” the Intel source said.

“Intel only gives it out once it’s finished. And it’s not finished.” said the source. “If it was mature enough to release, it would be released.” (AMD and Nvidia claim that Intel has working silicon and thus the part of the specification they are seeking is mature enough for release.)

A source close to Intel said AMD and Nvidia are being disingenuous about what they’re actually seeking. In short, AMD and Nvidia are seeking technology–referred to as the “host controller” specification–that Intel says is beyond the USB 3.0 specification. “Think of it as a guide to building hardware for USB 3.0. This is the part that Intel invests dollars and engineering man hours in and then licenses to the industry (so far, for zero dollars). We will give this out as soon as it is finished (or close to finished),” the Intel source said.

As a result, AMD, Nvidia, Via Technologies and others (not yet specified) could be driven to create their own USB 3.0 specification. “We are going to be forced to create a secondary specification” that would be introduced along with the Intel spec, the source close to AMD said. “To create a new open host controller standard for USB 3.0.”

Intel, meanwhile, says it’s moving with all due speed.

AMD rejects this argument because people at the company were around when the earlier 1.0 USB specification came out and claim that Intel stonewalled back then too. Intel denies the assertion.

“We are starting development on it right now,” the AMD source added. The first meeting of members of the alternate “open” specification is slated to take place next week, a source close to Nvidia said. “We fully intend to productize this spec.”

The AMD source described USB 3.0 as “essentially PCI Express over a cable. And that intellectual property came from the PCI SIG”–the point being that Intel does not have a large intellectual property stake to defend. PCI Express is a data transfer specification for add-in card slots in desktop PCs today. The PCI SIG (Special Interest Group) promotes the Peripheral Component Interconnect specification, a standard used in all PCs today.

The problem, as AMD and Nvidia see it, is that Intel would virtually own the USB 3.0 market–a powerful competitive advantage–for many months if they waited for Intel to release the host controller specification. “Tack on six to nine months. Then we get USB 3.0,” the AMD source said.

(Credit:
Stephen Shankland/CNET News.com)

UPDATE: AMD and Nvidia aim to wrest control of a crucial PC specification from Intel, arguing that the chip giant is trying to box them out as they move to a new era of faster peripherals.

Nvidia and AMD are offering no official comment.

“Just as with previous generations of USB, Intel is working hard to get the complete spec to the industry with as little delay as possible in order to drive the wide adoption of USB 3.0,” the company said in a statement.

A separate specification–though designed to be compatible with the Intel USB 3.0 spec–has the potential to create incompatibilities, the source close to AMD said. “This is not good for users. But we have no choice.”

The reason for this is the Air’s aluminum construction and light weight. In other words, when you pick up the Air, you sense an almost perfect balance of sturdiness and weight–despite the fact that the Air exceeds the dimensions of a typical, more-compact subnotebook.

I finally got my hands on a MacBook Air. Though I’m sure I’m repeating what some others have said already, I needed to state one thing before I do a more extensive evaluation: this unbelievably thin notebook is rock solid.

2003 HP business notebook and MacBook Air

(Credit:
Brooke Crothers)

NOTE: This is not an official CNET review. And is not by any means a full review. Just a quick first-look. Official CNET product review is here.

Update: One other quick impression. No excessive heat to speak of. Another challenge for designers of ultra-thin notebooks is how to effectively dissipate the heat that is generated by the core electronics, including the main processor, the graphics silicon, and hard drive. Apple has succeeded admirably. Admittedly, I am using the solid state drive (flash memory) version of the Air so there’s no hard drive heat to worry about and I don’t play games (some gamers have cited heat issues). Impressive nonetheless. More later.

One of the concerns I had–and I’m sure I’m not alone–is whether a notebook this thin will be flimsy and overly delicate. The answer is a resounding no. It feels more solid in my hands than the rugged, well-built, 1.6-inch-thick HP Compaq nc8000 I have used for many years (since 2003).