Kroah-Hartman announced on his blog that the 3.10 Linux Kernel would be the next long-term kernel.

He said, "Despite the fact that the 3.10-stable kernel releases are not slowing down at all, and there are plenty of pending patches already lined up for the next few releases, I figured it was a good time to let everyone know now that I’m picking the 3.10 kernel release as the next long term kernel, so they can start planning things around it if needed."

Why 3.10?

Kroah-Hartman explained, "I'm picking this kernel after spending a lot of time talking about kernel releases, and product releases and development schedules from a large range of companies and development groups. I couldn’t please everyone, but I think that the 3.10 kernel fits the largest common set of groups that rely on the long term kernel releases."

This idea of formally anointing an "official" Linux kernel has been around for years, but it only really got going in 2011. It was then that Kroah-Hartman, who'd been supporting Linux 2.6.16 as a de facto long term kernel, said that since that had worked out well, "a cabal of developers got together at a few different Linux conferences and determined that based on their future distro release cycles, we could all aim for standardizing on the 2.6.32 kernel, saving us all time and energy in the long run."

This worked out well, and all of the major 'enterprise and "stable" distro releases became based on the 2.6.32 kernel. Kroah-Hartman then found that it wasn't only the big Linux distributions' developers that craved long-term stability. Many other Linux developers, especially those in the embedded consumer-device space, wanted such a Linux kernel too.

So, Kroah-Hartman publicly proposed that a new long-term kernel be picked every year and that it be supported for two years. After some discussion on Google+ and the Linux Kernel Mailing List (LKML), the idea was refined and gained popular support in the development community.

Now, two years later, the plan has come to fruition. For the next few years, enterprise and commercial Linux distributions will be based on the 3.10 Linux kernel.

Related Stories:

• Linux 3.11: Linux for Workgroups
• Linux 3.10 brings accelerated video playback and new SSD caching
• Linux 3.9 kernel release offers SSD caching and server performance improvements
• SUSE to support SUSE Studio customized Linux distros

The new 4.3 features are, besides adding restricted profiles:

Android sandbox reinforced with SELinux:

Android 4.3 now includes SELinux, a mandatory access control (MAC) system in the Linux kernel to augment the Unique Identification Number (UID) based application sandbox. This makes almost all apps  with the Android sandbox much more secure.

Some users are wary of SELinux, since the NSA had a large hand in creating it. Since SELinux, just like all of Linux, is open source that seems foolish to me. After all, the code is right in plain sight for anyone to look for security holes.

KeyChain enhancements:

If you're still worried about the NSA snooping on your messages you'll be happy to see Google's new KeyChain API provides a method that enables applications to confirm that system-wide keys are bound to a hardware root of trust. This means that carrier and OEM developers can add private keys that cannot be copied off the device, even if it's otherwise completely compromised.

This won't stop the NSA -- or most major Internet companies -- from using big data, metadata, and traffic analysis to keep an eye on you, but it will eventually help to keep the contents of your messages and apps secure.

Android Keystore Provider:

At the same time, Android 4.3 also introduces a keystore provider and APIs that allow applications to create exclusive-use keys. What that means is that apps can create or store private keys that no other app can see or use.

Restrict Setuid from Android Apps:

Your device's /system partition is now mounted "nosuid" for Zygote-spawned processes. This helps prevent Android applications from executing setuid programs. In turn, this reduces root attack surface and likelihood of potential security vulnerabilities. In English, this means malicious apps will have a much harder time trying  to take over your device's superuser/root privileges.

It sounds good, and it is good. Unfortunately, it's also already obsolete. Chainfire, creator of SuperSU, an Android rooting program, has found a way to root Android 4.3 "by using an "su daemon," which is started from init [A vital Android boot-up program] and not from a Zygote process."

Wi-Fi support for WPA2-Enterprise networks:

New application programming interfaces (API)s can now be used configure the Wi-Fi credentials needed for connections to access points using WPA2 enterprise with Extensible Authentication Protocol (EAP) and Encapsulated EAP (Phase 2). With this, developers will be able to create apps that can join business Access Points (APs) that use EAP and Phase 2 authentication methods.

Beyond Android 4.3:

Google has also been adding improved security features for older versions of Android as well.

First, Verify Apps, a security feature introduced in Android 4.2, is no longer part of the operating system. Instead, it's been incorporated into Google Play Services, which is incorporated on Android 2.3 and higher. This service is client-side process that scans apps for malware as you install them. This works even if you're side-loading your new apps as Android application package files (APKs) from a third-party Android app store and not Google's Play Store.

Second, and boy have we waited a long time for this one, Google has finally added a lost phone finder to Android. Like Verify Apps you don't need a new smartphone to use it. This service will also be available to anyone using an Android device using Android 2.2 or above. With it you can make your little lost phone ring at maximum volume, signal you from a map, or, if worse comes to worse and it's been stolen, you can erase all your data from it remotely.

So much for the good news. The bad news is that OEMs are still being slow as an old dog on a hot day about rolling out Google's Bluebox Security patch. Adding insult to injury, the first malware-infected apps using this security hole have started appearing.

Even with this Android security has been improving this summer. Now, if only Google started forcing vendors and carriers to push security updates to users, I'd be a lot happier and you'd be a lot safer.

Related Stories:

• Google setting up 'lost phone' feature for Android devices
• ;Android should embrace a Windows-style security update model
• Android OEMs slow to roll out Bluebox Security patch
• Proof of concept for Android flaw found, patches start rolling out
• Google releases fix to OEMs for Blue Security Android security hole
• Five simple ways to avoid Android malware

T-Mobile, the smallest of the major US carriers, completed its merger with MetroPCS in May. Since then T-Mobile has been hyper-aggressive in trying to improve its sales position.

In early July, the company announced its new "JUMP!" program. In it, T-Mobile customers can upgrade their phones up to twice per year. Later in July, T-Mobile starting offering the company's entire phone lineup for no upfront payment.

It thus makes perfect sense for T-Mobile to look to Ubuntu as yet another way to separate it from AT&T and Sprint. Canonical said that "T-Mobile USA is the newest member of the Ubuntu Carrier Advisory Group. T-Mobile USA reaches almost 300 million American consumers and business people today. As a member of the CAG, T-Mobile USA will join discussions to influence the development of Ubuntu for smartphones."

Canonical, unlike Google with Android, won't allow T-Mobile, or any carrier, too much control over the operating system's look and feel. As Jono Bacon, Ubuntu's community manager, said at OSCon, Ubuntu is striving to prevent the interface fragmentation that plagues Android. "My wife and I both had Android phones and they gave us two entirely different experiences," said Bacon. "We're avoiding that."

Related Stories:

• Verizon backs Ubuntu smartphone
• Ubuntu: One OS, one interface, all devices
• Ubuntu Edge might just change the computing world
• The future of Ubuntu on mobile: Canonical forms carrier group to shape OS
• Ubuntu Linux enters the smartphone wars

This new Power server, which IBM advertises as a "high-performance/high-end" machine, is the next step up in its Power Systems PowerLinux line. It comes with four sockets and 32 CPU cores.

The PowerLinux 7R4 server is built on the same Power Systems platform running IBM's famous Watson cognitive computing solution and Jeopardy champion.

"As the Linux operating system and open source applications continue to mature, more clients are choosing IBM’s higher value hardware systems designed to handle mission critical and complex cloud and big data workloads in an open environment," said Doug Balog, General Manager for IBM Power Systems in a statement. "Responding to this need, we are aggressively continuing investments in our open Power Systems ecosystem -- including new products, applications and partnerships -- that support today's emerging Linux workloads.”

Like the rest of PowerLinux server line, the 7R4 runs with either Red Hat Enterprise Linux (RHEL) or SUSE Linux Enterprise Server (SLES). If you'd rather use AIX Unix and/or IBM i operating system you can use IBM's PowerVM virtualization tools to partition any Power Systems server into separate virtual machines with some running Linux-based applications while others running AIX or IBM i.

IBM is adding IBM Cognos Business Intelligence and EnterpriseDB database software. According to IBM, "In addition to IBM DB2 database software for Linux, which offers an average 98 percent compatibility with Oracle Database applications, IBM announced that EnterpriseDB's enterprise-level PostgreSQL-based database solution is now available on all Power Systems servers running Linux."

Ed Boyajian, EnterpriseDB's President and CEO, claimed in a statement that "Switching databases has traditionally been costly and risky due to limited application compatibility and lack of comprehensive migration tools and resources. EnterpriseDB's Postgres Plus Advanced Server and IBM Power Systems solve this problem by providing extensive Oracle compatibility functionality, migration tools and expertise that deliver significant cost savings while allowing many Oracle based applications to run virtually unchanged.”

Oracle would, I'm sure, beg to differ. With this move, IBM seems to be distancing itself ever further from Oracle. 

Related Stories:

• IBM to bring Linux KVM virtualization to its Power server line
• IBM Rounds Out Its Linux Offerings With Power Linux
• IBM targets emerging markets, offers cheap Power servers
• Red Hat and SUSE join IBM in new Linux system, Canonical opts out
• Power Systems: A better virtualization and big data platform than x86

When the first Top500 supercomputer list was compiled in June 1993, Linux was just gathering steam. Indeed, in 1993, the first successful Linux distributions, Slackware and Debian were only just getting off the ground.

What happened next, as reported in The Linux Foundation's report, 20 years of Top500.org Supercomputer Data Links Linux With Advances in Computing Performance, was that "after first appearing on the list in 1998, Linux has consistently dominated the top 10 over the past decade and has comprised more than 90 percent of the list since June 2010."

Before Linux made its move, Unix was supercomputing's dominant operating system. Since 2003, the top operating system by performance share on the Top500 List underwent a complete flip from 96 percent Unix to 96 percent Linux. By 2004, Linux had taken over the lead for good.

According to The Linux Foundation, "Linux [became] the driving force behind the breakthroughs in computing power that have fueled research and technological innovation. In other words, Linux is dominant in supercomputing, at least in part, because it is what is helping researchers push the limits on computing power."

The Foundation believes that this has happened because of two reasons. First, since most of the world’s top supercomputers are superscalar research machines built for specialized tasks, each supercomputer is a standalone project with unique characteristics and optimization requirements. Thus, it's not affordable for anyone to develop a custom operating system for each system. With Linux, however, research teams can easily modify and optimize Linux to the one-off, groundbreaking designs that characterize the modern generation of supercomputers.

And, just as importantly, "The licensing cost of a custom, self-supported Linux distribution is the same, whether you’re using 20 nodes or 20-million nodes." Thus, "by tapping into the vast open-source Linux community, projects had access to free support and developer resources to help keep developer costs on par with, or below other operating systems."

The result of this has been supercomputers that are going faster than ever. By total RMax, a supercomputer's maximum achieved performance on the Linpack benchmark, supercomputer performance has outpaced Moore’s Law (The number of transistors incorporated in a chip will approximately double every 24 months.) by doubling roughly every 14 months. At the top end, supercomputing is progressing at even more rapid rate. The RMax of the fastest supercomputer on the Top500 list has increased by a factor of three to reach the Tianhe-2’s 33.86 petaflop/second in 2013 from the CM-5’s 59.7 gigaflop/s in 1993."

Therefore, The Linux Foundation concluded, "By isolating RMax by operating system using the past 20 years of Top500 data, it’s clear that Linux is not only responsible for supporting the majority of supercomputers today, but is a driving force behind the disproportionate growth in supercomputing capacity over the past decade. In continuing to drive progress and innovation in computing, Linux is also helping to explore the mysteries of the universe and solve our toughest problems."

I can only agree with these conclusions.

Related Stories:

• Build your own supercomputer: First $99 Parallella boards ship
• PaRSEC: Designing software for the exascale supercomputer generation
• Build your own supercomputer out of Raspberry Pi boards
• TOP500 leaders announce new supercomputer benchmark
• Linux continues to rule supercomputers

The recent headlines have been about Canonical's crowd-sourcing of its hybrid smartphone/PC, Ubuntu Edge, but Canonical's plan of one integrated operating system and interface for all platforms predates it by years. While Unity is known better as an easy-to-use Linux desktop interface for beginners, Canonical has been aiming Unity at the smartphone and tablet market since it was introduced in October 2010.

It's only now that this plan is coming into focus for those who don't follow Ubuntu like a hawk. As Mark Shuttleworth, Canonical and Ubuntu's founder said at OSCon, the major open-source convention held in Portland, OR, "Convergence is the core story. Each device is great, but they should be part of one family. On any device you'll know what you're doing. One device should be able to give you all the experiences you can get from any one of them."

That's easy to say, but how do you do that? Jono Bacon, Ubuntu's community manager, explained how Canonical is making this happen at an OSCon session.

Bacon said, "Unity is all about getting rid of the computer. Its focus is helping users focus on content."

That's one reason why Unity has proven so unpopular with Linux power-users. For these users it's all about the power to customize the operating system and interface at as low-a-level as possible for the job at hand. That is not how ordinary people see an operating system and it's a major reason why Ubuntu Unity is more popular with new users.

What does Bacon mean exactly? Well, for starters, the interface should "show controls only when necessary and those controls should be responsive to what you're doing at the time." For example, if you're watching a video, the only visible controls should be ones such as pause, fast-forward, and rewind. Even then, these controls should only be visible when you want them.

Next, says Bacon, "Unity uses the same patterns across different devices. We want to make sure that those patterns can be used by our app developers across platforms." In other words, when you write an app, whether you use Ubuntu's native Qt Modeling Language (QML) or HTML5 for its interface, it's going to look and act the same no matter where it's running.

Bacon continued, the idea is to "focus on elegance. We don't want to clutter it up with buttons and widgets. We want beautiful, elegant devices." For example, "on a smartphone or tablet, you only need three buttons—power and audio up and down and you really don't need the volume controls."

Instead of buttons or on-screen icons, Unity uses the edges of the display. Specifically, the top of the screen is used for indicators and settings . The left edge holds the Launcher, which is a bar of icons that's similar to the Mac OS X dock. On the bottom edge you'll find the controls for the app that's currently on the screen. Finally, the right edge gives you access to multi-tasking functionality. To access any of them from a touchscreen you simply swipe from the edge to the display's center. With a mouse, you move to the edge of the screen, click, and pull to the center of the screen.

You see the point? It all looks and works the same regardless of the platform.

To help developers work with this Ubuntu version provides not just a software developer kit (SDK) but an app design guide as well. In addition, Ubuntu provides its own app font set, global patterns to make sure all apps behave consistently, and design building blocks to help make certain all your apps look like they're all part of the Unity family.

Besides providing a consistent look and feel for Ubuntu users, Bacon also said the idea is to avoid the kind of interface fragmentation that's afflicted Android. "My wife and I both had Android phones and they gave us two entirely different experiences," said Bacon. "We're avoiding that."

Can Canonical pull this off? Well, technically, they already have. The bigger question is: "Will the device vendors and carriers let them do it?" According to Bacon, they will. "They can have branding designs and their own selection of apps in the Launcher, but the edge interface look and work will remain the same."

The ultimate question, of course, is will you buy into this? Well, you'll get your chance. The US's biggest phone carrier, Verizon, will be offering Ubuntu smartphones and there's a carrier-group of international phone carriers that are also backing Ubuntu.

I'd say Ubuntu has a decent shot at becoming the number three mobile OS vendor after Android and iOS. And, who knows, it may finally get more of a share of the desktop operating system world since Windows 8 is still stumbling.

Related Stories:

• Ubuntu Edge might just change the computing world
• Ubuntu launches crowdfunding effort to manufacture a PC-level Edge superphone
• Verizon backs Ubuntu smartphone
• Can any tablet OS challenge Android and iOS?
• Ubuntu for Android: Linux desktop on a smartphone

First, for Joe and Jane user:

1) Support for Restricted Profiles:

This feature is for users who have kids. Android has allowed you to have multiple users for some time now, but with this version you can finally have restricted profiles.

What that means in English is you can keep junior out of your, ah, questionable apps or Web sites. Technically, it means that you can set up separate environments for each user with fine-grained restrictions in the apps that are available in those environments. According to Google, "Each restricted profile offers an isolated and secure space with its own local storage, home screens, widgets, and settings. Unlike with users, profiles are created from the tablet owner’s environment, based on the owner’s installed apps and system accounts. The owner controls which installed apps are enabled in the new profile, and access to the owner’s accounts is disabled by default."

While ideal for kids, restricted profiles are also ideal for guest users, kiosks, and point-of-sale (POS) devices.  This last point will give Android tablets a chance at the retail POS market that's recently been a strong point for iPads in stores.

2) OpenGL ES 3.0 for High-Performance Graphics:

I know what you're thinking. "How the heck will something called OpenGL ES 3.0 ever matter to an ordinary guy or gal with their smartphone or tablet?"

Easy, they'll never know the tech but they'll enjoy the far higher quality graphics in their games and videos. To really get the most out of it, of course, you'll need the hardware to back it up. Still, I see much better video experiences ahead for high-end Android tablet users.

Today, this is only supported on the new Nexus 7, Nexus 4, and Nexus 10 devices. More will follow.

And, now for the developers. Of course, they'll be busy implementing the above into their programs but what I see as attracting their attention are the other following features.

3) Bluetooth Smart Ready support

You may not know it, but a whole new family of Bluetooth devices have been arriving. What makes them different from their predecessors is Bluetooth Smart Ready. These are designed as sensors. So, for example, one might check if all windows are locked, while another might measure your heart rate. You get the idea.

In Android 4.3, with application programming interface (API) support for Bluetooth Generic Attribute Profile (GATT) services, you can create Android apps that will support these devices. This represents a new and potentially very profitable market for Android developers and their Bluetooth hardware partners.

4) Notification Access

People love those notifications at the top of their Android display. I know I do. I'm constantly checking them. Until this new version of Android appeared developers couldn't access this data stream. Now they can. That is, if you, the user, allow them to.

What developers can do is register a notification listener service that, with your blessing, will receive all the data notifications when they're displayed in the status bar. Developers can then launch applications or services for a new class of "smart" apps.

5) Better Digital Rights Management (DRM)

OK, go ahead and boo. I know you want too. I hate DRM too. But, here's the painful truth, DRM is here to stay and we might as well try to make the best of it.

That's exactly what Google has done with its new modular DRM framework. This will enable developers to more easily integrate DRM into their own streaming protocols such as MPEG Dynamic Adaptive Streaming over HTTP (DASH) (PDF Link).

Google has also added new media DRM framework APIs and improved the existing ones to provide  an integrated set of services for managing licensing and provisioning, accessing low-level codecs, and decoding encrypted media data.

The net effect of these changes is it will make DRM easier to manage and it should make video streams with DRM, which are pretty much all of them these days, look and play better. Like I said, Google is making the best of an annoying commercial video necessity.

Now, let's cut to the chase. When will you see it? Will you see it at all? Hugo Barra, vice president of Galaxy Nexus smartphones, will get the upgrades over the air. After that, the Google Play editions of the Galaxy S4 and the HTC One will get the upgrade.

As for everyone else... good question. As usual it will depend upon your phone's OEM and your carrier. If you can't stand to wait, possibly forever, for them, you should start looking into alternative Android Jelly Bean ROMs such as Cynaogenmod. I have no doubt they'll be porting Android 4.3 as fast as they can to a wide variety of Android devices.

Related Stories:

• Google teams with Asus on upgraded Android 4.3-based Nexus 7
• Android plus Chrome OS equals Google's future operating system
• Microsoft's most profitable mobile operating system: Android
• AMD unveils plans to support Android and Chrome OS

Before getting into why he'll win not just the bet that gadget fans will want one but the far bigger bet that the world is ready for a single handheld device that can double as a full PC, the Ubuntu Edge.

Jono Bacon, Ubuntu's community manager described the Ubuntu Edge as a device that will "dual boot Ubuntu and Android, and will transform into a PC when docked with a monitor—with the full Ubuntu desktop and shared access to all the phone’s files. For this it needs the power of a PC, so Ubuntu Edge will be equipped with the latest, fastest processor, at least 4GB of RAM and a massive 128GB of (SSD) storage."

Specifically, its technical specs, at this time, look like this: 

• Dual-boot Ubuntu Edge into either Ubuntu or Android

• A fully integrated Ubuntu desktop PC when docked

• Fast and powerful device with multi-core CPU and at least 4GB RAM

• 128GB of Solid State Device (SSD) storage for photos, music, content

• A 4.5-inch 1,280 x 720 HD display with a pure sapphire crystal screen, the hardest natural substance after the diamond

• Cameras made for low-light, fast response and close up pictures: 8mp rear camera, 2mp front

• Faster connections all over the world with dual-LTE, dual-band 802.11n Wi-Fi, Bluetooth 4, Near Field Communication (NFC)

• Connect to HDMI TVs and monitors easily with MHL connector, 3.5mm jack

• GPS, accelerometer, gyro, proximity sensor, compass, barometer

• Stereo speakers with HD audio, dual-mic recording, Active Noise Cancellation

• Silicon-anode Li-Ion battery

• 64 x 9 x 124mm size.

So, why will Canonical get the $32-million it needs from crowd sourcing via Indiegogo over the next 30 days to finance a limited production run of 40,000 phones? First, because gadget lovers won't be able to resist it.  Second, they're already buying into it.

In the first 24-hours of the fundraiser, Canonical raised over $3-million. According to the company, "We hit $1 million in under five hours, and $2 million not long after that. We’ve now passed the 10 percent line, which shows just how much demand there is out there for the Edge’s radical new approach to mobile technology."

Pretty good right?

Here's what you may not know.

Canonical has been working on combining an Androd smartphone and the Ubuntu desktop since early 2012. This idea of a hybrid smartphone/PC has been in the works for months. Mark and company know exactly what they're doing.

And, what are they working on you ask? Well, it's not simply throwing out yet another cool gadget idea to float or sink. It's the result of a long, hard look at the PC market and deciding that it was time for a radical new idea.

Unless you've been under a rock, you know that the PC market has started a long slow death. Earlier this year, IDC reported the worse PC sales drop in a generation.

It's not that PCs are going to go away. They're not. We're still going to be using them in offices and homes for years. They let us do work that we can't easily do with a tablet or a phone. But, and it's a big but, what we really want these days are smaller, hand-portable devices.

One of my ZDNet colleagues argues that RT's failure may herald the death of Windows. Another thinks that it shows that the Windows PC is becoming irrelevant. Google would agree but maintain that desktop computing's future belongs to the cloud-based Chromebook. The Chromebook's sales numbers would seem to bear this out.

Canonical looked at this same market and came up with its own take. Yes, they say, people want a handheld device not a PC. But they also know that many of us will always need a keyboard-based device as well.

Their answer: Give people a single device. Make it so people don't need to carry a laptop, even a Chromebook, and a smartphone. Make it so that you can carry all your needed computing power in a shirt pocket. They foresee a post-PC future where you simply plug your super-smartphone into your office, hotel, conference center, or home keyboard and monitor and you'll be ready to go.

Will people buy into this bet? I think they will.

 I don't know if single-unit, everything-in-one, fit into your pocket PCs/smartphones will become the form factor for the near future. I do know that it won't be the traditional PC or laptop. Their day is done. The future belongs to cloud-based devices, tablets, smartphones, and, quite possibly, devices that look and work a lot like Canonical's Ubuntu Edge.

Related Stories:

• Ubuntu launches crowdfunding effort to manufacture a PC-level Edge superphone
• Windows isn't dying, it's just becoming irrelevant
• Did we all just witness Windows start to die?
• Can any tablet OS challenge Android and iOS?
• Ubuntu for Android: Linux desktop on a smartphone

Of course neither OpenOffice nor LibreOffice are Linux only. Both are available on multiple platforms, including Linux, OS X and Windows, with additional third-party ports to other operating systems. Neither, however, have delivered on a long-promised, software-as-a-service (SaaS) cloud version.

That said, OpenOffice 4.0 comes with a new, more modern user interface, improvements to Microsoft Office interoperability, enhanced graphics support and many other enhancements. For office workers, the most significant of these may be its new Microsoft Office Open XML (OOXML) support. It now includes support for docx outline levels, support for table background color from table style in docx files, more bullet and numbering support in docx, and support for font color in pptx files.

"Microsoft Office interoperability is a very high priority for the project," said Juergen Schmidt, Apache OpenOffice Release Manager, in a statement. "We are working hard to ensure that our users can successfully exchange documents and document content with colleagues who continue to use Microsoft Office. Moving forward we plan to continue our focus on OOXML document interoperability."

The user interface also now includes IBM's Lotus Symphony Sidebar user interface. While IBM is no longer developing its Symphony OpenOffice fork, its coders are still at work on the main OpenOffice.

Apache claims that "the new Sidebar makes better use of today's widescreen displays. Users may easily edit their document properties in-context, with the most-frequently needed controls available in panels in the Sidebar.  Panels may be expanded or collapsed as needed."

"IBM is proud to see its source code contribution of IBM Lotus Symphony coming to fruition with the release of Apache OpenOffice 4.0," said Kevin Cavanaugh, VP of IBM Collaboration Solutions. "The time is right for wide-scale enterprise adoption, especially with the upcoming end-of-support for Microsoft Office 2003. By choosing Apache OpenOffice, enterprises will free up resources for their cloud and mobile infrastructure investments."

 "With Apache OpenOffice  we are making major improvements to our user experience by introducing the Sidebar, the first radical improvement to the OpenOffice user interface in years," said Andrea Pescetti, Apache OpenOffice's VP.

Pescetti continued:

"Together with major new improvements in Microsoft Office interoperability, enhancements in graphics and color palette management as well as improvements in Calc, Chart and Draw editor modules, Apache OpenOffice 4.0 adds up to a compelling new release. With a rigorous quality assurance testing process, we wholeheartedly recommend our user community begin to upgrade.  Innovation happens at Apache and is immediately available to everybody who wishes to build upon the OpenOffice source code." 

In addition, OpenOffice now has a new framework that enables application developers to build extensions to the Sidebar interface.  This is meant to allow programmers to integrate business application data,  seamlessly integrate with cloud and mobile document editing environments, and automate common document workflow tasks. These extensions, like all of OpenOffice, are under the Apache License 2.0. They can be downloaded from the new version of the Apache OpenOffice Extension Repository.

Related Stories:

• Does OpenOffice have a future?
• LibreOffice 4.0: The big changes will be under the hood
• Cloud-based Apache OpenOffice aims to woo mobile users
• IBM to close down Symphony, its OpenOffice fork

On July 23rd, Adapteva announced that it has started delivering its first Parallella computers to users who ordered the board through its Epiphany multicore processors.  

"The Kickstarter community took a chance that we could deliver an entirely new type of computing platform for only $99," said Andreas Olofsson, Adapteva Founder and CEO in a statement. "We could not have done this without the passion and backing of this unique community and I’m excited to see where the developer community will take this platform in the future."

The first model comes with the following hardware features:

• Zynq-7020 dual-core ARM A9 CPU

• Epiphany Multicore Accelerator (16 or 64 cores)

• 1GB SDRAM

• MicroSD Card

• USB 2.0 (two)

• Four expansion connectors [option]

• Ethernet 10/100/1000

• HDMI connection

These initial boards use Ubuntu Linux.  Support for other operating systems will depend on community involvement, but Ubuntu is the recommended operating system.

 The mini-supercomputer board also ships with the open-source Epiphany development tools (PDF link). This includes a C compiler, multicore debugger, Eclipse IDE, OpenCL SDK/compiler, and run-time libraries.

There are also several already ongoing Parallella community software projects. These include support for the Go language; GNU radio, a software defined radio; and support for high-performance computing (HPC)'s Message Passing Interface (MPI). In the case of the latter, there's also an ongoing project to build a Beowulf-style supercomputer using Parallella boards.

 After some final refinements, Adapteva promises that the 6,300 Parallella’s ordered via Kickstarter will be delivered by summer's end. Delivery to the entire Kickstarter community will be fulfilled first and general availability orders will ship later this fall. Parallella boards will be available in different build configurations with a starting price of $99.

Pre-orders of the 16-core Parallella platform for the general public can be made at the Adapteva shop . The standard versions for non-Kickstarter backers will have the Zync-7010 dual-core instead, with the 7020 offered as an upgrade.

Related Stories:

• Parallella: The $99 Linux supercomputer
• Build your own supercomputer out of Raspberry Pi boards
• TOP500 leaders announce new supercomputer benchmark
• Linux continues to rule supercomputers

Jim Curry, one of OpenStack's founders, and today Rackspace's general manager of private cloud, explained how OpenStack exploded on the cloud science in a ZDNet interview.

"A couple of things came together," said Curry. "First, cloud technology and its form factor was hitting an infraction point. After several years, Amazon Web Services was just moving into the mainstream and people were looking for not just for an open-source alternative, but for any AWS alternative."

In addition, of course, OpenStack is open-source software and that also attracted many partners.

"When we launched, cloud technology was still immature." Besides, "We were all building our own clouds and there was no interoperability between clouds." Everyone knew this wouldn't work as a business model.

In many ways, OpenStack was simply the right technology at the right time. It also helped enormously, Curry thinks, that "Rackspace and NASA already had great reputations in servers and research respectively."

Last, but in no way least, "We marketed the heck out of it. Much of what OpenStack was when we launched was a promise. We worked hard on the marketing and partnering side. It then became a self-fulfilling prophecy. Lots of users and vendors came in and wanted to work with OpenStack. We, as a community, did a good job of finding like-minded people and we built something that worked really well."

Even with all that, Curry never expected it to become as big as it is today. "I'm shocked at where it is today. We didn't have global ambitions for it. We were trying to solve specific cloud problems but everyone wanted to get involved. Today, it helps not only business IT, but humanity as a whole, with projects like CERN and genetics research. We're making the world a better place with open source."

"On the commercial end," Curry still sees "Amazon is who we compete with," rather than other open-source cloud efforts such as AWS-friendly Eucalyptus or CloudStack.

It's not just the public cloud that concerns Curry. He's found that OpenStack is being used to set up "Private clouds as internal competitors for AWS. OpenStack has exploded in the enterprise."

Some analysts still think Rackspace has a way to go before it has really taken off in the enterprise. RightScale's State of the Cloud 2013 survey (PDF Link) found that Rackspace is second only to AWS on the business public cloud and leads Eucalyptus and CloudStack on private cloud deployments.

One thing that is slowing down OpenStack is what Curry believes is slowing down all cloud deployments: "The enemy to how to move to DevOps. Switching to the cloud represents a fundamental change in IT architecture. Sure, you can treat the cloud as if it was just shared hosts, but if you do, you're missing the point that the service model is what makes the difference."

Looking ahead, Curry doesn't see big changes for OpenStack. "The next OpenStack version, Havana, which comes out in October, will not be adding much in the way of new features." Instead, "it will be much more about continuing to stabilize it and make it more manageable and even better for real deployments."

Related Stories:

• Red Hat bets its cloud future on OpenStack
• Why CERN chose OpenStack: A conversation with Rackspace's John Engates
• HP builds out OpenStack infrastructure, adds public cloud services
• OpenStack Summit Report: Real Customers Building Better Products Faster With Open-Source Cloud
• VMware, Canonical team up on OpenStack

The first move GitHub made to make their site more truly open-source friendly was to reset the site. Now when you start up a new project with the Git distributed version control system (VCS) file repository you're  given a choice to add an open-source license to your project immediately.

If you can't tell the differences among the patent-aware Apache License, the anything goes MIT License, or the restrictive GPLv3 license, worry not. GitHub now has a simple site, ChooseALicense that gives you some direction.

Of course, if you have hopes for your project to go commercial you'll need more legal guidance than this site provides, but still it's a good start.

If you choose to create a new public repository and not to use a license be aware that you're going to be waiving some of your copyright rights. Specifically, according to GitHub's new Terms of Service, Section F, "By setting your repositories to be viewed publicly, you agree to allow others to view and fork your repositories."

The key word is "fork." Unless you want to share your work without having any control over what happens to it next, you're going to run to either keep your GitHub project private or put it under an appropriate open-source license.

It's a good thing that GitHub is doing this because its users' various project licensing was actually far messier than anyone knew.

Black Duck Software, the leading open-source legal software and consulting company, turned out its new Black Duck Suite 6.5 with its Deep License Data feature on GitHub and other public open-source file repositories. This version can examine source code for embedded licenses that exist within projects having no declared license.

Black Duck did this because “The lack of a declared license for an open source project can cause an enterprise to steer clear of it, limiting the projects organizations can use,” explained Mark Driver, Gartner's VP and Research Director in a statement. “The ability to access embedded license information and obligations up-front during the code selection process opens a sizable opportunity for enterprises and could have significant impact on their bottom line.”

What Black Duck found was distributing for anyone who planned on using many GitHub projects. Black Duck found  "that 77 percent of projects on GitHub have no declared license, compared to only 7 percent of projects from all other open-source forges."

Worse still, "Further analysis from Black Duck’s Deep License Data shows that of this 77-percent, 42-percent of GitHub projects actually have embedded licenses that carry specific obligations for use. Providing up-front visibility into embedded licenses and their associated obligations gives organizations the insight needed to make informed decisions about using such projects, and dramatically expands the open-source adoption opportunity."

What all this means for developers is that, if you want your code to be used by anyone else than you and your buddies, you must check to see if the code you've borrowed from elsewhere already has a license attached to it. In short, you must  pay attention to open-source licenses. If you get your licensing ducks all in a row, it will make it much more attractive to other developers and businesses.

For enterprises considering using open-source programs from GitHub, or other public sites, Black Duck President and CEO, Tim Yeaton thinks using Black Duck Suite to find licensing problems before they can bite you only makes sense.

“We’ve analyzed one million projects for embedded license information so our customers can access it at the beginning, and also throughout, the development process," said Yeaton in a statement. "As a result, more projects can be considered for use during the code selection stage of development – enabling developers to make informed component choices, and allowing enterprises to build better software faster, confident from the start that they can meet the code’s license obligations for use."

Related Stories:

• GitHub celebrates fifth birthday, 3.5 million users and six million repositories
• Google woos developers by releasing cloud platform code to GitHub
• Microsoft adding support for Git open-source version control to its developer tools
• Open source: Its true cost and where it's going awry by Monty Widenius

The two organizations claim that with ReKey, Android users can immediately protect their Android phone from Bluebox Security's "Master Key" vulnerabilities, without waiting on security updates from their mobile carrier.

"ReKey is the latest of our research projects designed to make the Internet a safer place," said Collin Mulliner, a postdoctoral researcher at NEU SecLab in a statement. "We hope that ReKey will provide a practical tool for users to protect themselves and, at the same time, raise awareness of the challenges in the mobile security space."

Jon Oberheide, CTO of Duo Security, added, "The security of Android devices worldwide is paralyzed by the slow patching practices of mobile carriers and other parties in the Android ecosystem."

ReKey isn't for everyone though. It will only work on rooted devices.

In the ReKey FAQ, they explain, "In order to patch the vulnerabilities on your device, ReKey requires escalated privileges. Normal unprivileged applications on stock Android devices do not possess such privileges, hence the need for a rooted device with the Superuser (or similar) application."

The fix program itself "is based on a dynamic instrumentation framework for Dalvik bytecode. Both Master Key vulnerabilities are present in software that is written in Java and is executed in the Dalvik VM. ReKey injects a small piece of code into the running Android framework. The code dynamically patches the ZipEntry and ZipFile classes to interpose on the vulnerable routines and thereby fix the root cause of the bugs. In addition to fixing the bugs, ReKey installs a warning system that alerts the user when they attempt to install an APK [Android application package file] that abuses the vulnerabilities."

In addition, the Bluebox scanner that checks for the security hole doesn't register the ReKey fix. So even after you install ReKey, the scanner will still report that your phone has the vulnerability. The ReKey team claims that Bluebox scanner "does not appear to be accurately checking whether the vulnerability is actually present or not."

If this makes you wary of ReKey, I can't blame you. That said, the two organizations have a good reputation and the program currently has a decent rating of 3.8 on the Google Play Store. So, if you have a rooted smartphone or tablet and are nervous about their security, you may want to try ReKey. Users running stock Android on their devices, however, will not be able to use it.

Related Stories:

• Android should embrace a Windows-style security update model
• Android OEMs slow to roll out Bluebox Security patch
• Proof of concept for Android flaw found, patches start rolling out
• Google releases fix to OEMs for Blue Security Android security hole
• Security firm claims 99 percent of Android apps open to takeover
• Five simple ways to avoid Android malware

Eucalyptus is an open-source program for building AWS-compatible private and hybrid infrastructure-as-a-service (IaaS) clouds. It pools together existing virtualized infrastructure to create cloud resources for computers, network, and storage. The program also works with AWS, thanks to its support of AWS application programming interfaces (APIs) for Amazon Elastic Compute Cloud (Amazon EC2), Amazon Simple Storage Service (Amazon S3), and other AWS services.

The project and its backing company, Eucalyptus Systems, don't just use the AWS APIs. Eucalyptus and AWS have an agreement to facilitate moving cloud workloads from Eucalyptus-based environments in and out of Amazon's AWS.

Eucalyptus 3.3 brings to the table the following new features:

Auto Scaling
 allows application developers to scale Eucalyptus resources up or down based on policies defined using Amazon EC2-compatible APIs and tools. With Auto Scaling, cloud resources can be seamlessly increased or decreased to maintain performance and meet SLAs [Service Level Agreements].

Elastic Load Balancing is an AWS-compatible service that distributes incoming application traffic across multiple Eucalyptus instances to provide greater fault tolerance for applications.

CloudWatch
 is an AWS-compatible service that monitors cloud resources and applications running on Eucalyptus clouds. It provides a reliable and flexible monitoring solution, which allows application developers and cloud administrators to programmatically collect metrics, set alarms, identify trends, and take action to ensure applications run smoothly.

In addition, the latest version of Eucalyptus includes resource tagging. This enables developers and cloud administrators to assign customizable metadata to Eucalyptus resources. Users, in turn, can use these to categorize cloud resources. For example, you tag cloud resources by purpose, owner, or environment. Developers and administrators can then filter by these tags to manage and monitor specific resource collections.

Eucalyptus also now supports an expanded set of instance types that more closely align with Amazon EC2 instance types.

Want to check it out for yourself before buying into the service? You can try the commercial version of Eucalyptus for free on your own systems with Eucalyptus' FastStart automated installer. Since it's open-source software, you can also download the code and work with the Eucalyptus Community Cloud and rely on community support.

Related stories

• Open-source cloud Eucalyptus gets visual with 'toughened' version 3.2

• Amazon reaches for datacentres with Eucalyptus deal

• Eucalyptus drops paid-for version to become single cloud platform

• Eucalyptus and Amazon work together to expand cloud adoption

Now, with a new logo sporting Tux, the Linux penguin, holding a flag that's remindful of the old Windows 3.1 logo, Linux 3.11 is ready for testers. This is far from the first time that Torvalds has given early builds of the Linux kernel whimsical nicknames. Earlier ones have included "Unicycling Gorilla, "Holy Dancing Manatees, Batman," and "Jeff Thinks I Should Change This, But To What?"

On a more serious note, Torvalds announced a variety of small changes in 3.11.

Most of these come from AMD and amount to better support for the Radeon graphics card family. This will include a new Direct Rendering Manager (DRM) driver.

Intel was also represented with improvements for the Haswell processor line. It also includes better support for Intel's Valley View Atom processor and Bay Trail, its Atom system-on-a-chip design.

In addition, Xen and KVM virtualization will work on 64-bit ARM architectures with 3.11. The kernel will also include a Lustre-distributed file system client. Lustre tends to be used in cluster-computing.

Finally, zswap is a lightweight compressed memory cache for swap pages. Its purpose is to improve performance when a Linux system is in danger of running out of memory.

Last, but not least, Torvalds has not given any hints about any future version of Linux being entitled "Linux XP."

Related Stories:

• Linux 3.10 brings accelerated video playback and new SSD caching
• OS X apps on their way to Linux courtesy of Darling project
• Linux 3.9 kernel release offers SSD caching and server performance improvements
• SUSE to support SUSE Studio customized Linux distros
• Fedora 19, Schrdinger's Cat, lives

Unlike the $25 bare-bones Raspberry Pi PC, the Utilite will come in a paperback-book sized, 5.3-inch 3.9-inch 0.8-inch container.

It also boasts far more processing power. The Utilite, which will start shipping in August, is powered by a Freescale i.MX6 system-on-chip with a single, dual, or quad-core ARM Cortex-A9 processor running up to speeds of 1.2GHz.

This miniature computer can hold up to 4GBs of Double data rate synchronous dynamic random-access memory DDR3 RAM. For storage it can come with up to a 512GB mini-Serial ATA, Solid-State Drive (SSD). The front-panel micro-SD socket supports Secure Digital extended Capacity (SDXC) cards with up to 128GBs of room.

For video output, the Utilite uses High-Definition Multimedia Interface (HDMI) 1.4 and a second Digital Video Interface-Digital (DVI-D) port. Both display outputs support resolutions up to 1920 x 1200 and can be operated in dual-head mode. 5.1 channels audio is available through HDMI and through Sony/Philips Digital Interconnect Format (S/PDIF). In addition there are standard 3.5mm jacks for stereo line-out and line-in.

To connect it with your network and other devices, the Utilite comes with a pair of Gigabit Ethernet ports and built-in 802.11n Wi-Fi and Bluetooth. It includes four USB 2.0 ports: two on the back and two on the front. It also has a mini RS232 connector.

Users will have their choice of Android or Ubuntu Linux for their operating system. CompuLab has experience in delivering Linux-based systems with its Mint-based MintBox line. No further operating system details are available at this time.

With its combination of power, size, dual Gigabit Ethernet ports, and case, businesses might want to consider using Utilite as well. It appears as if it would make a fine low-end server or, with the right Linux server software, a dedicated router, firewall, or file-server.

Related Stories:

• Build your own supercomputer out of Raspberry Pi boards
• PC homebrewing and white-boxing: Dead or alive?
• 'We thought we'd sell 1,000': The inside story of the Raspberry Pi
• Raspberry Pi gets photo and video capabilities with 20 camera module
• Parallella: The $99 Linux supercomputer

Google does a decent job of fixing Android security holes. For example, the Bluebox Security hole was fixed three days after it was publicly announced. That's great as far as it went, but the Android OEMs and carriers have released the patch for only a handful of smartphones and none of the tablets.

This is unacceptable.

True, you'd need to ignore Android security basics to pick up an infected program, but there's a security fool born every minute. Besides, while today most Android malware infects devices via third-party Android app stores and questionable malware-laden Web sites, it's only a matter of time before hackers adopt more subtle ways to introduce malware into Android devices.

In short, Google needs to tighten Android's security. True, Google has introduced such advanced security features as Address Space Layout Randomization (ASLR) and Data Execution Prevention (DEP) in Android 4.1, Jelly Bean. That still doesn't protect you from all malware.

On top of that, only 37.9 percent of Android users are running 4.1 and higher. Over 60 percent are running earlier, more vulnerable, versions of Android. In addition, just like Windows, there are always new Android security holes being discovered and exploited even in the latest and newest versions.

Security is a never-ending battle.

While Microsoft's answer has its problems--for every Patch Tuesday, there's an Exploit Wednesday--at least Microsoft's approach ensures that careful users will be protected from most security holes regardless of whether they're running a Dell laptop, an HP PC, or a Lenovo ThinkPad.

Google needs to take the same approach. Just like Microsoft releases patches for XP from Windows 8.1, Google needs to push security patches from at least Android 2.1. Eclair, which still has 1.4 percent of the market, to market-leading Android 4.1 and up.

Microsoft doesn't depend on the big PC vendors to deliver patches and Google shouldn't either. As this latest episode shows, neither the OEMs nor the carriers can be trusted to keep their users secure.

Google needs to sit its Android OEM customers down and tell them that since they can't, or won't, deliver security patches, it will do it for them. Microsoft did it with Acer, Asus, and all the other PC vendors, Google must do it with HTC, Samsung, and all its smartphone and tablet partners.

The alternative is for Android's users to be permanently vulnerable to both old, long-fixed security holes and the latest malware.

Related Stories:

• Android OEMs slow to roll out Bluebox Security patch
• Proof of concept for Android flaw found, patches start rolling out
• Google releases fix to OEMs for Blue Security Android security hole
• Security firm claims 99 percent of Android apps open to takeover
• Five simple ways to avoid Android malware

In a document detailing the support policies for this virtualization software and hardware (Link requires Oracle support account), Oracle said, "In an effort to more tightly align Oracle's future desktop virtualization portfolio investments with Oracle Corporation's overall core business strategy, we have ended new feature development for Oracle Virtual Desktop Infrastructure Software, Oracle Virtual Desktop Client (OVDC) Software, Oracle Sun Ray Software (SRS), and Oracle Sun Ray Client hardware (including Sun Ray Operating Software)."

These products enable customers to manage, deploy, and provide users with access to server-hosted desktop operating systems on nearly any client device. They were picked up by Oracle as part of its 2009 acquisition of Sun.

In the same document, Oracle said, "Going forward, Oracle's desktop portfolio investments will be focused on continued development and new enhancements to both Oracle Secure Global Desktop and Oracle VM VirtualBox software." Secure Global Desktop is another server-based VDI program, and VirtualBox is a popular desktop virtualization program.

Oracle partners don't care for this move. According to one disgruntled partner, Oracle could have handled this change better. "Announcing the death of yet another product via a hard-to-find support document ... I would have appreciated a more honest and open approach."

In addition, the partner expressed unhappiness about how Oracle had handled this virtualization line since Oracle's takeover of Sun. "Despite being the owner of Oracle VDI, Sun Ray, and Oracle VM, Oracle refused to integrate those three. So up to this day you cannot use Oracle VM as a hypervisor. You can use Hyper-V, you can use VMware, you can use that special version of VirtualBox that shipped with Oracle VDI ... but Oracle VM? Nope. How ridiculous is that??"

Still, he added, "The idea behind Oracle VDI and Sun Ray is excellent: You have a device on your desk without any mechanical parts and MBTF [mean time between failure] is around 22 years ... All you need now is plenty of servers with enough CPU cores, RAM, enough disks for IOPS [input/output operations per second] and storage space ... and all your desktops live there, centrally in the server room."

Another Oracle partner said, "This is very disappointing and troubling for the healthcare and banking verticals, which has heavily invested and integrated Sun Ray desktops for its security and mobility."

For the time being, Oracle will continue to support the existing software and hardware and renew licenses.

"We would like to assure new and existing customers that technical support for these products will continue uninterrupted as they are today. Customers may also continue to renew existing support contracts or purchase new licenses. Exact time-frames for a last order date for Oracle Sun Ray client devices will be announced shortly."

Still, this is a blow for Oracle system integrators, value-added resellers (VARs), and any enterprise that had committed to this VDI stack.

Related stories:

• Oracle enhances its desktop virtualization offering

• Virtualisation suites compared

• "The new VDI Reality" - a quick book review

• New Wipro VDI faces perception challenges

• Dell unveils new virtual desktop portfolio catering to mobile workers

Worse still, there's now a proof of concept for the security hole. This proof of concept means that as surely as the sun will rise in the east in the morning we'll soon see real malware using it.

What's a user to do? Well, for starters, there's no real need to panic if you just follow a few simple, security rules with your Android device to avoid apps that have been compromised with this exploit.

What the OEMs should be doing, and for the most part aren't, is releasing the patch so there will be no reason to worry about it. At this time, the only Android smartphones and tablets I'm certain have have the patch are the Samsung Galaxy S4, the HTC One, and hardware using the latest version of the alternative Android firmware CyanogenMod.

According to Gina Scigliano, Google's Android Communications Manager, Google has "not seen any evidence of exploitation in Google Play or other app stores via our security scanning tools. Google Play scans for this issue - and Verify Apps provides protection for Android users who download apps to their devices outside of Play." Verify Apps in a security program in Android 4.2 and higher. It scan any apps you want to download and install against Google’s database of safe apps.

Scigliano also said that "Nexus devices will receive the fix in an upcoming software update."

In the meantime, if you want to make darn sure that there are no compromised apps on your system Bluebox Security has released an Android program, Bluebox Security Scanner for apps that try to take advantage of this security flaw.

In addition, Bluebox Security Scanner checks to see if your device is vulnerable or patched for the Bluebox "Master key" security flaw. The scanner also checks to see if your system is set to allow non-Google Play application installs. Non-Google Play Android markets are the most likely vector for any corrupted Android apps.

To sum up, if you're careful about where you download your Android software you should be safe whether your system is patched or not. That said, it would sure be nice for the OEMs to get on with integrating Google's patch into their customized versions of Android so we can all have safer devices and we wont need to worry about the problem anymore at all.

Related Stories:

• Proof of concept for Android flaw found, patches start rolling out
• Google releases fix to OEMs for Blue Security Android security hole
• Security firm claims 99 percent of Android apps open to takeover
• Five simple ways to avoid Android malware

While some experts, like Horst Simon, the Deputy Director at the Lawrence Berkeley National Laboratory’s National Energy Research Scientific Computing Center (NERSC ) is willing to bet we won't have exascale computers by 2020, other supercomputer pros are sure we will.

Exascale computers will be about 1,000 times faster than today's top Linux-powered petascale supercomputers. How big is that? Well if you had a quintillion pennies, it would make up a cube about five miles to a side. 

Leaving silliness aside, Simon pointed out that exascale supercomputing will fundamentally break "our current programming paradigm and computing ecosystem." In other words, even if we build the hardware, we won't be able to use it efficiently.

Jack Dongarra, distinguished professor of computer science at the University of Tennessee, Knoxville and creator of the TOP500 supercomputer list, thinks we will have exascale computers and doesn't disagree with the second part of Simon's position. That's why he's working on designing software that will make the next generation of supercomputers operational.

Dongarra recently received a million-dollar grant over three years from the US Department of Energy to find answers for these programming problems. Called the Parallel Runtime Scheduling and Execution Controller (PaRSEC), this aims to address the critical situation that is facing the supercomputing community due to the introduction of more complex supercomputer designs.

"You can't wait for the exascale computers to be delivered and then start thinking about the software and algorithms," said Dongarra in a statement. "The exascale computers are going to be dramatically different than the computers we have today. We have to have the techniques and software to effectively use these machines on the most challenging science problems in the near future."

According to Dongarra, "Today's supercomputers have processor counts in the millions. Tomorrow's exascale computers will have roughly a billion processors. In addition, the general makeup of the machines will differ dramatically through the use of multiple central processing units and hybrid systems to overcome barriers limiting today's supercomputers. These barriers include large amounts of heat and power consumption, leaking voltage and a limited bandwidth of data through the pins on a single chip."

The PaRSEC site states that PaRSEC is a generic framework for architecture aware scheduling and management of micro-tasks on distributed many-core heterogeneous architectures. In it, applications will be expressed as a Direct Acyclic Graph (DAG) of tasks with labeled edges designating data dependencies. DAGs are represented in a compact problem-size independent format that can be queried on-demand to discover data dependencies in a totally distributed fashion. PaRSEC assigns computation threads to the cores, overlaps communications and computations and uses a dynamic, fully-distributed scheduler based on architectural features such as Non-Uniform Memory Access (NUMA) nodes and algorithmic features such as data reuse.

Dongarra is also developing an algorithm to overcome a reliability problem associated with the increasing number of processors. If it works when one processor fails, the calculation may then have to be repeated partially or in full. The project aims to develop software that can survive failures.

Additional work on exascale computing will be done at meetings hosted by the National Science Foundation, are held around the world annually. 

Related Stories:

• TOP500 leaders announce new supercomputer benchmark
• Linux continues to rule supercomputers
• Build your own supercomputer out of Raspberry Pi boards
• Parallella: The $99 Linux supercomputer
• China takes supercomputing crown from US