November 12, 2007 2:33 PM
Posted by: ITKE
For some time now, Linux has been the dominant operating system in high performance computing. For everything from IBM, with its rockstar status supercomputer Blue Gene, to NEC or U.S. HPC players SGI and HP–the bulk of the leading HPC clusters today are Linux-based.
Four of the top five HPC systems in existence today are based on Linux, according to Top500 Supercomputing Sites, an independent web site that tracks the largest, fastest HPC deployments in the world. In 2005, when Top500.com started calculating which specific OS was dominating HPC, it found that Linux was used in nearly 80% of the world’s the fastest HPC systems.
The next TOP500 list will be released Nov. 13th (that’s tomorrow) during the Supercomputing Conference (SC07) in Reno, Nevada, but why wait until then for more Linux HPC goodness? UPDATE below: The list arrived early.
This morning our sister site SearchDataCenter.com broke the news that Sun Microsystems would be releasing two new systems designed to address the extreme computation, scale and storage requirements of today’s high-performance computing (HPC) customers. Called the Sun Constellation System the supercomputers are open computing environments that combine ultradense, high-performance compute, networking, storage, and software into an integrated “petascale” general-purpose system. Running Solaris, Linux and Microsoft Windows, the Sun Constellation System is designed to scale from departmental clusters to the largest supercomputer configurations, enabling customers to solve complex computational problems, the company said.
And that’s the first big Linux news of the day. Apparently, Constellation’s first user is the Texas Advanced Computing Center (TACC), which will put the finishing touches on a 504 trillion floating-point operations per second (teraflop) compute cluster in December. This CentOS Linux Cluster, named Ranger, will have 3,936 nodes, 123 terabytes of memory and 62,976 processor cores from AMD Opteron quad-core processors. All system components will be connected via a full-Clos InfiniBand interconnect. Eighty-two compute racks will house the infrastructure, which will sit in TACC’s J.J. Pickle Research Campus in Austin, Texas. IDC seems to think Constellation will make a pretty big impact in the HPC space. We’ll have to keep an eye out for it on future Top500 lists.
But wait, there’s more! According to 451 Group analyst and CAOS Theory blogger Jay Lyman, last week supercomputer superpower Cray rolled out its super scalable XT5 supercomputer and wouldn’t you know it, the thing runs Linux. It’s apparently the company’s biggest use of Linux to date, with Cray using AMD quad-core processors in a configuration of more than 1,000 CPUs to top 40 teraflops of performance. “As far as I know this is the highest density of Opterons you can buy in a system,” said Jan Silverman, senior vice president of corporate strategy, in an interview with EFTimes.com.
But let’s not forget about Microsoft’s HPC endeavors in the midst of all this Linux HPC love. Microsoft’s Windows Computer Cluster Server (CCS) is designed for the lower end of the HPC market, and industry watchers said the technology has been well received, particularly for small computing clusters. Vendors too have jumped onto the Windows CCS bandwagon, including HP, which extended a multimillion-dollar investment agreement with Microsoft to drive HPC into the mass market. It is also selling CCS 2003 as part of its HP Unified Cluster Portfolio, and Fujitsu Computer Products of America Inc., which recently published a best-practices paper for HPC cluster deployment, is using Microsoft Windows instead of Linux. But I’m still willing to bet that Linux has a lock on the list tomorrow. Call it an educated hunch.
Like I said, the list hits tomorrow at SC07. Let the chips fall where they may (and that’s a lot of chips — lolz!)
UPDATE: Or we could just have the list today. It’s out at 2:45 p.m. EST. Blue Gene again takes the top spot, but apparently there were plenty of surprises. More below… Continued »
November 12, 2007 10:03 AM
Posted by: ITKE
News broke this morning that Oracle is entering the virtualization space with Oracle VM. Toes were stepped on, canned analyst comments were issued via press release, and the virtualization space once again proved that it is hot, hot, hot and has no intention of settling down just yet.
According to Oracle, its new server virtualization software supports both Oracle and non-Oracle applications. Oracle products like Oracle Database, Oracle Fusion Middleware and Oracle Applications are all certified with Oracle VM.
Consisting of open source server software and an integrated Web browser- based management console, Oracle VM provides a graphical interface for creating and managing virtual server pools, running on x86 and x86-64-based systems, across an enterprise. Oracle VM offers:
- Simplified Installation — with single install, patching and upgrading for both Oracle VM and Oracle Enterprise Linux
- Faster Deployment — through pre-configured Virtual Machine images of Oracle Database and Oracle Enterprise Linux
- Linux and Windows Support — support for Linux and Windows guest operating systems including: Oracle Enterprise Linux 4 and 5; RHEL3, RHEL4 and RHEL5; Windows 2003, Windows Server 2003 and Windows XP (on HV capable hardware)
- Single Source of Support — customers will have a single point of contact for their entire virtual environments, including the Linux operating system, and Oracle Database, Fusion Middleware and Application software
Oracle applications, middleware and database software currently certified with Oracle VM include:
– Oracle Database 10.2.0.3 and 11.1
– Oracle Application Server 10gR2 and 10gR3
– Oracle Enterprise Manager 10.2.0.4
– Oracle TimesTen 126.96.36.199
– Oracle Berkeley DB 4.6
– Oracle E-Business Suite 11.5.10 and 12
– Oracle’s PeopleSoft Enterprise 9.0
– PeopleTools 8.49.07 and above
– Oracle’s Siebel CRM 8
– Oracle’s Hyperion 9.3.1
I’m sure we’ll be hearing more about this very soon. The big questions now: Do we really need another hypervisor, and, what role will the OS play in the future, if any?
UPDATE@3:30 pm. EST: As far as specs for Oracle VM… it’s built off of Xen so the license itself is free. Oracle will only charge for support (either 2 proc for $499 or unlimited proc for $999 per instance). That’s from our man/woman on the inside, so look for confirmation later.
November 9, 2007 12:26 PM
Posted by: ITKE
This dropped earlier this month, but all of the breaking news about Red Hat kept me locked up in breaking news land and I couldn’t post until now. My apologies.
Here’s what’s new in OpenNMS 1.3.8. OpenNMS is the world’s first enterprise grade network management platform developed under the open source model. It consists of a community supported open-source project as well as a commercial services, training, and support organization.
The big news is that OpenNMS now runs on Windows. The OpenNMS wiki warns that this is initial Windows support, and “there are a lot of rough edges, but it is possible to install and run on Windows 2000 and up.”
Both Linkd and Syslogd have also been significantly improved and the “split-personality issues with configuration differences between running the embedded Jetty web UI and running in Tomcat have been improved.” The Jetty server is also more configurable, with support for being proxied. Rudimentary RSS feed support has been added to the Web UI.
And as always, the openNMS community would like you to know that a number of bugs, big and small, have been fixed.
November 9, 2007 11:04 AM
Posted by: ITKE
, Ubuntu Linux
It’s a day of blog subject line questions, apparently.
The subject line question is asked today because there’s apparently some concern over Gutsy Gibbon’s documentation. Gibbon is the latest release of Ubuntu, version 7.10, which came out last month to the usual fanfare associated with new Ubuntu releases.
That said, Carla Schroder, writing for Enterprise Networking Planet, has a bone to pick with Canonical and the Ubuntu development team over documentation:
Whatever anyone may think of Ubuntu, you can’t deny they’re busy little critters, stuffing all manner of new things into every release. Which is a splendid thing, and what would make it even better is if they documented all of these wonderful new things. And also the old things. I think it’s the worst of the major Linux distributions for documentation. I spent a considerable amount of time trying to find out what makes Ubuntu’s server kernel different from a desktop kernel, what exactly is the OEM installation, where is the online package search page, what’s new in this release, and what’s included in this release. www.ubuntu.com is poorly-organized and seems more marketing-oriented than informative.
In our overview of Linux support, then and now, from this earlier summer, we identified similar documentation concerns with Linux in general. To see that one of the signature distros (and by that I mean one of the most wildly popular ones outside of Red Hat and Novell) is suffering from poor documentation is troubling to say the least.
In 2003, systems administrator Pati Moss said some online Linux/OSS instructions are very “high level and difficult to decipher.” IT manager Rick Segeberg agreed: “Newbies to Linux (especially non-programmers) find it difficult to follow the very technical documentation and how-tos that are available,” he said. “Most of the technical documentation is done by technical people for technical people.”
So documentation can be too technical and difficult to decipher. Got it. In our overview this was still a “thorn in the side” of IT managers, but at least Red Hat and Novell had stepped up to offer paid, professional support. But with Ubuntu, however, things are apparently too sparse! Not a good sign when Mark Shuttleworth and company at Canonical are trying to get Ubuntu pre-installed on commodity hardware from vendors like Dell.
Schroder again (under the wonderful subject line, “Dammit, Jim, I need documentation!”):
The Ubuntu release notes are quite sparse, and they lump the server and desktop editions together. There are bug reports and workarounds, but where is the list of major and new features? AppArmor is a radically new inclusion, but the only mention of it is that it breaks printing. What is it, and what do you do with it? What are the kernel versions, and versions of major applications like Apache and … well, what exactly comes with this release? What hardware architectures are supported, and what are some of the specific issues for them? And so forth—just cruise the Debian and Fedora release notes to see how it should be done. In fact you can check out older Ubuntu release notes—the farther back you go, the more complete they are, though they’re still short of what they should be.
The problem here, it seems, is that the notes assume a lot on the part of the reader. Do I know what AppArmor is? Sure I do, but I write about OS’s like Novell SUSE Linux and I know that AppArmor has been included in that distro for quite some time.
A huge part of the Linux ecosystem is documentation. Ubuntu Server — at least for one columnist — comes up short: “It’s also typical to include batches of READMEs and CHANGES and other helpful documentation on installation CDs. Don’t bother looking on the Ubuntu Server CD for these, because there aren’t any. However, it does include the “Ubuntu Installation Guide”, which is actually the Debian Installation Guide with some minor modifications, such as changing “Debian” to “Ubuntu”, and adding useful links to online Ubuntu resources,” Schroder said.
Apparently the Ubuntu documentation is Debian stuff dressed up with Ubuntu branding. Sounds kind of like what CentOS does with Red Hat Enterprise Linux (in reverse, of course), but in this case it’s a bad thing. Enterprise users want enterprise level documentation with their Linux OS. If Canonical is truly serious about getting Ubuntu pre-installed on commodity hardware, they’d best be tweaking their docs … do you agree?
November 9, 2007 10:12 AM
Posted by: ITKE
, Red Hat
Over at InternetNews.com, Sean Michael Kerner picked the brains of several analysts to see which of them saw any effects, positive or negative, from the year-old partnership between Microsoft and Novell.
If you’ll remember, in November 2006 the two companies entered into a controversial partnership that — in theory — would allow interoperability between Windows and SUSE Linux as well as a patent covenant. The covenant ensured that Microsoft would not sue Novell’s SUSE Enterprise Linux Server users over alleged intellectual property infringement by open source applications.
Suffice to say, the analysts were mixed on whether or not the partnership was anything more that a temporary revenue boost for Novell (Novell noted that as of the end of the third quarter of its fiscal year, it had invoiced more than $105 million in SUSE Linux Enterprise certificates through business collaboration with Microsoft).
Our good friend Charles King, founder of Pund-IT, saw some glimmer of hope for Novell, but as one of two pro-Novell analysts (and I use that label very lightly in this case), even he wasn’t so sure. In fact, OS business gains are still very hazy for Novell, and King could only point to the firm’s 50% revenue gains in the Open Platform Group.
“This can’t be traced directly to the company’s deal with Microsoft, but the agreement may have had an impact,” King told InternetNews.com. “The financial effects for Microsoft appear negligible (or so company CEO Steve Ballmer has said), but the Novell deal arguably paved the way for the company to pursue other agreements with open source players.” (InternetNews.com)
Other analysts, including SearchEnterpriseLinux.com regular Gordon Haff, of Illuminata Inc., couldn’t identify any gains for Novell (or losses for Red Hat) that tied in directly with the Microsoft partnership.
Two corporations fire off a mean press release with 30 new customers and the opening of an interoperability lab — where’s the beef?
I realize it’s only one year in, but I think it would behoove Microsoft and Novell to flesh out what it is they mean when they talk about “interoperability frameworks” and the like. Revenue numbers from Novell for its OS business wouldn’t hurt, either.
November 9, 2007 9:52 AM
Posted by: ITKE
Administration, interoperability and integration
, Backup & recovery
, Clusters, grids and mainframes
, disaster recovery
, Enterprise applications for Linux
, green computing
, Hardware issues
, identity management
, Linux basics
, Open source applications
, Red Hat
, Systems Management
Working with vendors is tough. You need their help, they want your money. Hopefully, whatever it is they help you install works and the price meets you both somewhere in the middle (as in your side of the middle, right?).
Sometimes this process is a headache, but sometimes a project can really surprise you—things just work and upper management is just peachy keen with how the whole thing looks on the balance sheet.
In that vein, SearchEnterpriseLinux.com wants to help its readers discover the best of the best in Linux products for the enterprise in our prestigious SearchEnterpriseLinux.com 2007 Products of the Year awards. We’ve been asking readers and vendors over at SearchEnterpriseLinux.com to nominate a favorite product they’ve used or to nominate their own new product, and now we’ve opened it up to the Intertubes here at the Enterprise Linux Log. Regardless of where you fall — vendor, user or general Linux guru –the deadline is drawing near!
Our editorial team and a select panel of industry experts and analysts are currently accepting submissions online until 5 p.m. PST on Nov. 9, 2007 in a range of categories, including: Server Linux platform product (either a distribution release or a new, integrated server Linux offering); Security applications/tools for Linux on the server; Virtualization product for Linux on the server; and Linux administration tools. You can access the 2007 POY submission page in the link above.
To qualify, new or significantly upgraded products must have been shipped after October 31, 2006, and before November 1, 2007. Submit your entry today and let us know what you think are the top data center products on the market!
November 8, 2007 10:49 AM
Posted by: ITKE
, Systems Management
Yesterday, Red Hat offered a smörgåsbord to the press and the analyst community. Their main announcement: Red Hat is the greatest Linux and open source vendor on the face of the planet.
Actually, there were several announcements in yesterday’s conference call and webcast: within the typical sales and marketing noise was talk of virtualization at almost every level of the discussion, hosted by a trio of Red Hat executives.
The first of the announcements, regarding the official release of Red Hat Enterprise Linux 5.1, was made by Scott Crenshaw, Red Hat’s vice president of enterprise Linux business. In some prepared remarks, Crenshaw went after proprietary virtualization technologies, saying RHEL 5.1′s virtualization delivers broader server support and up to twice the performance that the competition.
The skinny on 5.1
There were no real surprises in this announcement, especially if you’re a regular reader of SearchEnterpriseLinux.com. Back in September we filed a preview article on 5.1 (RHEL 5.1 update tweaks virtualization, Windows interoperability), where we discussed the virtualizaiton updates with a few experts. Jan Stafford, our Senior Site Editor at SEL, had a 5.1 preview up as far back as May from the Red Hat Summit.
RHEL 5.0 was a success when it launched in March. The inclusion of Xen support was almost a full year behind Novell, which had baked in Xen paravirtualization back in June 2006, but it worked as advertised, albeit with a few tweaks here and there. “It’s not half-baked,” Illuminata analyst Gordon Haff told me at the time, “but it certainly doesn’t have the fit and finish we see with VMware.” Not many things do these days, as VMware loves to point out during their quarterly “ESX Server prints money!!!” press conferences. With 5.1 officially avaialble to Red Hat customers via the Red Hat Network, however, the consensus was that the gap got a little smaller.
Also back in September, Jim Klein, director of information services and technology at the Saugus Union School District in Saugus, Calif., told me that RHEL 5.1 is a “significant improvement over version 5 on the management side of things.”
In this regard, the Windows functionality in 5.1 is critical: IT managers are making decisions now about which platform to base their virtualized infrastructure on, Klein said. “If Red Hat can get their Windows drivers out soon, I think they will be well positioned to pick up significant market share in the coming year,” he said (Read Jim Klein’s Enterprise Linux Log guest blog post on Xen and Fedora 7 – J.L.).
Chance of clouds
Moving on, things got a bit cloudy during the press conference as Crenshaw and company (Paul Cormier, v.p engineering; and Brian Stevens, CTO) announced that beta availability of Red Hat Enterprise Linux on Amazon Elastic Compute Cloud (Amazon EC2), a web service that provides re-sizeable compute capacity in the cloud.
In a statement that accompanied the press call, Red Hat said the combination of RHEL and Amazon EC2 “changes the economics of computing by allowing customers to pay only for the infrastructure software services and capacity that they actually use. Red Hat Enterprise Linux on Amazon EC2 enables customers to increase or decrease capacity within minutes, removing the need to over-buy software and hardware capacity as a set of resources to handle periodic spikes in demand.”
As part of this partnership, Red Hat Network will offer a common set of management and automation tools across on-premises deployments and the Amazon EC2 cloud computing environment. Red Hat will provide technical support and maintenance of Red Hat Enterprise Linux on Amazon EC2. This is the first commercially supported operating system available on Amazon EC2.
As far as pricing and availability are concerned, RHEL on Amazon EC2 is available as a private beta today, with public beta availability planned for the fourth calendar quarter of 2007. Base prices are $19 per month, per user and $0.21, $0.53 or $0.94 for every compute hour used on Amazon’s EC2 service, depending on whether customers choose a small, large or extra-large compute instance size, plus bandwidth and storage fees.
Red Hat Appliance OS, HO!
The final piece of the pie was the pending release of Red Hat Appliance Operating System, or AOS for short. This ISV-themed OS means that in the very near future (first half of 2008, execs told me), ISVs will be assembling appliances for their customers that run on AOS and work with every certified RHEL application under the sun. Hint: That’s a lot, and was exactly the angle Red Hat executives took on the Wednesday call.
“The Red Hat Appliance Operating System will allow applications that are certified on Red Hat Enterprise Linux to be deployed as software appliances on the broadest range of servers in the industry, including those running Red Hat Enterprise Linux, VMware ESX and Microsoft Windows Viridian. Red Hat’s Linux Automation strategy, also announced today, delivers a standardized development, deployment and management infrastructure for the entire Red Hat Enterprise Linux ecosystem,” a statement said. Look for an industry reaction piece from us on SearchEnterpriseLinux.com later in the day.
The Red Hat Appliance Operating System (AOS) is built from Red Hat Enterprise Linux, with which it shares full ABI and API compatibility. It includes the Virtual Appliance Development Kit (vADK) that will allow ISVs to configure the operating system along with their middleware and applications to produce a complete system image.
Red Hat also announced that a range of software solutions on Red Hat Exchange are available for trial and purchase as pre-configured software appliances. Customers can now purchase and deploy an integrated solution consisting of third-party software, JBoss middleware and Red Hat Enterprise Linux. The total time necessary to purchase, install and use these solutions is “just minutes,” Crenshaw said.
A lot of PR in this announcement, so we’ll have to see where it goes in 2008. Stay tuned.
November 6, 2007 2:35 PM
Posted by: ITKE
, Linux basics
IBM’s DeveloperWorks site has an up-to-date, concise explainer about the Linux file system up today that’s worth a quick read, for refresher’s sake.
When it comes to file systems, Linux is the Swiss Army knife of operating systems. Linux supports a large number of file systems, from journaling to clustering to cryptographic. Linux is a wonderful platform for using standard and more exotic file systems and also for developing file systems. This article explores the virtual file system (VFS)—sometimes called the virtual filesystem switch—in the Linux kernel and then reviews some of the major structures that tie file systems together.
It covers all the basics (and I mean BASICS; example of a header found within: What is a file system?), and then goes into the technical stuff after that. Like I said, worth a quick read when you get the chance!
November 6, 2007 1:23 PM
Posted by: ITKE
, Red Hat
, sun microsystems
It took them about a year, but Red Hat finally got on board with Sun Microsystems and Java this week when it announced that it signed Sun’s contributor agreement that covers participation in all Sun-led open source projects by all Red Hat engineers.
We covered the news angle at SearchEnterpriseLinux.com, and today the Business Review Online’s Jason Stamper gives the news a fresh spin at the CAOS Theory blog:
The only question really is why it took Red Hat so long to make this move. As The451 Group’s Raven Zachary noted (in our article — J.L.), until now Red Hat has been a little coy about fully backing Java, choosing instead to work with BEA Systems on JRockit for optimization on Linux.
What’s changed of course is the open sourcing of Java, which has made it simpler for Red Hat to use it in its Linux distribution and related tools, Zachary said. The question remains though, what took them? Sun officially open sourced Java a year ago (emphasis mine).
I’m sure it was a variety of things: pride, business acumen, the intricacies of working collaboratively with a competitor who wants to bury your OS and replace it on the server with his own. But the question kind of lingers in the air like a whiff of freshly brewed coffee, doesn’t it? (indirect pun totally intended)
One Slashdot commenter also wondered about this partnership earlier this morning, saying:
“With all the ‘openness’ going on with Java these days will things get even more complicated? I have three important commercial apps that run on Java, all three have their own run time environments that are incompatible with each other. I have no end of trouble with jre and firefox. I can’t count how many times I’ve had problems with classpaths trying to run Java stuff. Will the OpenJDK mean another runtime? As in Blackdown, Sun, Open?”
There’s no doubt that this partnership is a good thing for Red Hat Linux and for Java. Nevertheless, these persistent little questions remain. I don’t know if RH dragging its feet has too much effect on the end user, but still, we’d like to know what the delay was all about.