Language Selection

English French German Italian Portuguese Spanish

Server: Supercomputing, Kubernetes and More

Filed under
Server
  • Supercomputing under a new lens: A Sandia-developed benchmark re-ranks top computers

    A Sandia National Laboratories software program now installed as an additional test for the widely observed TOP500 supercomputer challenge has become increasingly prominent. The program’s full name — High Performance Conjugate Gradients, or HPCG — doesn’t come trippingly to the tongue, but word is seeping out that this relatively new benchmarking program is becoming as valuable as its venerable partner — the High Performance LINPACK program — which some say has become less than satisfactory in measuring many of today’s computational challenges.

  • Bright Computing adds support for OpenHPC

    Today Bright Computing announced it has joined the Linux Foundation and will participate in the OpenHPC Community project. The latest release of Bright Cluster Manager provides the ability for Bright customers to easily integrate OpenHPC libraries and packages for use within a Bright cluster.

  • Kubernetes Becomes The First Project To Graduate From The Cloud Native Computing Foundation
  • Usenet, Authentication, and Engineering (or: Early Design Decisions for Usenet)

    A Twitter thread on trolls brought up mention of trolls on Usenet. The reason they were so hard to deal with, even then, has some lessons for today; besides, the history is interesting. (Aside: this is, I think, the first longish thing I've ever written about any of the early design decisions for Usenet. I should note that this is entirely my writing, and memory can play many tricks across nearly 40 years.)

  • The true costs of hosting in the cloud

    Should we host in the cloud or on our own servers? This question was at the center of Dmytro Dyachuk's talk, given during KubeCon + CloudNativeCon last November. While many services simply launch in the cloud without the organizations behind them considering other options, large content-hosting services have actually moved back to their own data centers: Dropbox migrated in 2016 and Instagram in 2014. Because such transitions can be expensive and risky, understanding the economics of hosting is a critical part of launching a new service. Actual hosting costs are often misunderstood, or secret, so it is sometimes difficult to get the numbers right. In this article, we'll use Dyachuk's talk to try to answer the "million dollar question": "buy or rent?"

  • Memcached DDoS Attacks Slow Down as Patching Ramps Up

    Days after the largest distributed denial-of-service attack in internet history, the attack size of memcached DDoS attacks is now on the decline.

    On March 5, Netscout Arbor Networks reported a 1.7-Tbps DDoS attack that was driven by the amplification of misconfigured memcached servers. While there were some initial fears that the attacks would continue to grow in size, the opposite has happened.

    "We're still seeing lots of them, but their average size is considerably smaller due to ongoing cleanup and mitigation efforts," Steinthor Bjarnason, senior network security analyst at Netscout Arbor, told eWEEK.

More in Tux Machines

Linux 4.18 RC2 Released From China

  • Linux 4.18-rc2
    Another week, another -rc. I'm still traveling - now in China - but at least I'm doing this rc Sunday _evening_ local time rather than _morning_. And next rc I'll be back home and over rmy jetlag (knock wood) so everything should be back to the traditional schedule. Anyway, it's early in the rc series yet, but things look fairly normal. About a third of the patch is drivers (drm and s390 stand out, but here's networking and block updates too, and misc noise all over). We also had some of the core dma files move from drivers/base/dma-* (and lib/dma-*) to kernel/dma/*. We sometimes do code movement (and other "renaming" things) after the merge window simply because it tends to be less disruptive that way. Another 20% is under "tools" - mainly due to some selftest updates for rseq, but there's some turbostat and perf tooling work too. We also had some noticeable filesystem updates, particularly to cifs. I'm going to point those out, because some of them probably shouldn't have been in rc2. They were "fixes" not in the "regressions" sense, but in the "missing features" sense. So please, people, the "fixes" during the rc series really should be things that are _regressions_. If it used to work, and it no longer does, then fixing that is a good and proper fix. Or if something oopses or has a security implication, then the fix for that is a real fix. But if it's something that has never worked, even if it "fixes" some behavior, then it's new development, and that should come in during the merge window. Just because you think it's a "fix" doesn't mean that it really is one, at least in the "during the rc series" sense. Anyway, with that small rant out of the way, the rest is mostly arch updates (x86, powerpc, arm64, mips), and core networking. Go forth and test. Things look fairly sane, it's not really all that scary. Shortlog appended for people who want to scan through what changed. Linus
  • Linux 4.18-rc2 Released With A Normal Week's Worth Of Changes
    Due to traveling in China, Linus Torvalds has released the Linux 4.18-rc2 kernel a half-day ahead of schedule, but overall things are looking good for Linux 4.18.

A GTK+ 3 update

  • A GTK+ 3 update
    When we started development towards GTK+ 4, we laid out a plan that said GTK+ 3.22 would be the final, stable branch of GTK+ 3. And we’ve stuck to this for a while. I has served us reasonably well — GTK+ 3 stopped changing in drastic ways, which was well-received, and we are finally seeing applications moving from GTK+ 2.
  • GTK+ 3.24 To Deliver Some New Features While Waiting For GTK4
    While the GNOME tool-kit developers have been hard at work on GTK4 roughly the past two years and have kept GTK3 frozen at GTK+ 3.22, a GTK+ 3.24 release is now being worked on to deliver some new features until GTK+ 4.0 is ready to be released. While GTK+ 4.0 is shaping up well and GTK+ 3.22 was planned to be the last GTK3 stable release, the developers have had second thoughts due to GTK+ 4 taking time to mature. Some limited new features are being offered up in the GTK+ 3.24 release to debut this September.

Finally: First stable release of KBibTeX for KDE Frameworks 5

After almost exactly two years of being work-in-progress, the first stable release of KBibTeX for KDE Frameworks 5 has been published! You can grab the sources at your local KDE mirror. Some distributions like ArchLinux already ship binary packages. After one beta and one release candidate, now comes the final release. You may wonder why this release gets version number 0.8.1 but not 0.8 as expected. This is simply due to the fact that I noticed a bug in CMakeLists.txt when computing version numbers which did not work if the version number just had two fields, i. e. no ‘patch’ version. As the code and the tag of 0.8 was already pushed, I had no alternative than to fix the problem and increase the version number. Otherwise, the ChangeLog (alternative view) is virtually unchanged compared to the last pre-release. Read more

Today in Techrights