Language Selection

English French German Italian Portuguese Spanish

Server

All Linux, all the time: Supercomputers Top 500

Filed under
GNU
Linux
Server

Starting at the top, two IBM-built supercomputers, Summit and Sierra, at the Department of Energy's Oak Ridge National Laboratory (ORNL) in Tennessee and Lawrence Livermore National Laboratory in California, respectively to the bottom -- a Lenovo Xeon-powered box in China -- all of them run Linux.

Linux supports more hardware architectures than any other operating system. In supercomputers, it supports both clusters, such as Summit and Sierra, the most common architecture, and Massively Parallel Processing (MPP), which is used by the number three computer Sunway TaihuLight.

When it comes to high-performance computing (HPC), Intel dominates the TOP500 by providing processing power to 95.6% of all systems included on the list. That said, IBM's POWER powers the fastest supercomputers. One supercomputer works its high-speed magic with Arm processors: Sandia Labs' Astra, an HPE design, which uses over 130-thousand Cavium ThunderX2 cores.

And, what do all these processors run? Linux, of course.
.
133 systems of the Top 500 supercomputers are using either accelerator or co-processor setups. Of these most are using Nvidia GPUs. And, once more, it's Linux conducting the hardware in a symphony of speed.

Read more

Red Hat welcomes Oracle to the oVirt community

Filed under
Red Hat
Server

On behalf of the oVirt community, its contributors and Red Hat, we welcome Oracle to the oVirt community. oVirt is the open source component that enables management of the Linux Kernel Virtual Machine (KVM), the hypervisor for virtualized environments running on the Linux kernel.

At Red Hat, we believe that upstream collaboration drives innovation, even among competitors. To this end, Red Hat has a 10+ year tenure of thought leadership, contributions and collaboration in the oVirt and KVM communities. Our development and release processes are designed to ensure that Red Hat contributions to these communities are pushed upstream so the benefits gained from our efforts are available to the community at large and available for any and all to draw from.

Read more

Also: IBM-Powered Supercomputers Lead Semi-Annual Rankings

Server: Red Hat, CentOS 8, Linux On ARM Servers and IBM

Filed under
Server
  • Why Chefs Collaborate in the Kitchen

    In a large commercial kitchen, for example hotels or cafeterias, chefs collaborate to create the recipes and meals. Sure, there is more than enough work for one person, and tasks are divided into chopping, mixing, cleaning, garnishing; but the recipe is collaboratively created.

    Suppose one chef broke away and created his or her own recipe? How would the kitchen maintain standards, tastes and reputation? Developing software using open source principles follows a similar theory.

    [...]

    Red Hat is the second largest corporate contributor to the Linux kernel. This means Red Hat engineers and support staff are well versed and able to resolve customer issues involving the Linux kernel. Every application container includes part of the Linux distribution and relies on the Linux kernel, which is the center of the Linux Operating System.

  • CentOS 8 Status 17-June-2019

    Since the release of Red Hat Enterprise Linux 8 (on 07-May) we've been looking into the tools that we use to build CentOS Linux. We've chosen to use the Koji buildsystem for RPMs, paired with the Module Build Service for modules, delivered through a distribution called Mbox.

    Mbox allows us to run the Koji Hub (the central job orchestrator), and the Module Build Service in an instance of OKD that we maintain specifically for our buildsystem work. We have 2 instances of mbox; one for the primary architectures (x86_64, ppc64le, and aarch64), and one for the secondary architecture (armhfp). OKD lets us run those instances on the same hardware but in separate namespaces. The builder machines are separate from the OKD cluster, and connect back to the individual buildsystems that they're assigned to.

  • CentOS 8.0 Is Looking Like It's Still Some Weeks Out

    For those eager to see CentOS 8.0 as the community open-source rebuild of Red Hat Enterprise Linux 8.0, progress is being made but it looks like the release is still some weeks out.

    There's been the Wiki page detailing the state of affairs for CentOS 8.0. New today is a blog post summing up the current status. Progress is being made both on building the traditional RHEL8 RPM packages as well as the newer modules/streams. Koji is being used to build the RPMs while the Module Build Service with Mbox is handling the modules.

  • NVIDIA Brings CUDA to Arm, Enabling New Path to Exascale Supercomputing

    International Supercomputing Conference -- NVIDIA today announced its support for Arm CPUs, providing the high performance computing industry a new path to build extremely energy-efficient, AI-enabled exascale supercomputers.

  • NVIDIA Delivering CUDA To Linux On Arm For HPC/Servers

    NVIDIA announced this morning for ISC 2019 that they are bringing CUDA to Arm beyond their work already for supporting GPU computing with lower-power Tegra SoCs.

  • Nvidia pushes ARM supercomputing

    Graphics chip maker Nvidia is best known for consumer computing, vying with AMD's Radeon line for framerates and eye candy. But the venerable giant hasn't ignored the rise of GPU-powered applications that have little or nothing to do with gaming. In the early 2000s, UNC researcher Mark Harris began work popularizing the term "GPGPU," referencing the use of Graphics Processing Units for non-graphics-related tasks. But most of us didn't really become aware of the non-graphics-related possibilities until GPU-powered bitcoin-mining code was released in 2010, and shortly thereafter, strange boxes packed nearly solid with high-end gaming cards started popping up everywhere.

  • At ISC: DDN Launches EXA5 for AI, Big Data, HPC Workloads
  • IBM Makes Takes Another Big Step To Hybrid Computing

    Today, IBM announced the ability to leverage its unique turnkey operating environment, IBM i, and its AIX UNIX operating systems on IBM Cloud. Both OSs debuted in the 1980s and have a long history with many IBM customers. In addition, IBM i remains one of the most automated, fully integrated, and low-maintenance operating environments. Extending both OSs to IBM Cloud will allow customers to expand their resources on-demand, to migrate to the cloud, to leverage the latest Power9 servers, and to leverage IBM’s extensive resources. IBM is rolling out the service first in North America for customers using IBM i or AIX on Power servers. In conjunction with the extension of the hybrid cloud platform, IBM also announced a program to validate business partners with Power Systems expertise.

CERN Is Working To Move Further Away From Microsoft Due To License Costs Going Up By 10x

Filed under
GNU
Linux
Server
Microsoft

CERN, The European Organization for Nuclear Research that is home to the Large Hadron Collider and a lot of other experiments, is experimenting with moving further away from Microsoft products. Due to Microsoft license fee increases affecting their work in the research laboratory and its budget, they established the Microsoft Alternatives "MAlt" project.

CERN had already long been involved with developing Scientific Linux (now shifting to CentOS) but they have still been reliant upon Microsoft products in other areas, on some Windows systems as well as using the likes of Skype for Business.

Read more

Also today: Ubuntu preinstalled by Lenovo.

Ubuntu Server development summary – 11 June 2019

Filed under
Server
BSD
Ubuntu

The purpose of this communication is to provide a status update and highlights for any interesting subjects from the Ubuntu Server Team. If you would like to reach the server team, you can find us at the #ubuntu-server channel on Freenode. Alternatively, you can sign up and use the Ubuntu Server Team mailing list or visit the Ubuntu Server discourse hub for more discussion.

Read more

Also: DragonFlyBSD 5.6 RC1 Released With VM Optimizations, HAMMER2 By Default

Data: NGD, Sisense and More

Filed under
Server
  • NGD ships 8TB M.2 SSD with in-situ data processing

    In this case the processor is an ARM Cortex-A53 core running a 64-bit operating system, a version of Ubuntu Linux. This should enable the development of applications that run on the embedded core with minimal changes from code running on an X86 Linux system.

  • Sisense Brings Power to the Builders With New Cloud-Native Linux Platform, Insights to Everyone With Predictive AI Technology

    Sisense, the world's leading modern platform for analytics builders, kicked off its second annual customer conference, Sisense Eureka!, with a series of product innovations designed to help developers, data scientists and business analysts simplify complex data and provide insights to everyone across a business.

  • Future Kubernetes Will Mimic What Facebook Already Does

     

    And just to be clear, Chunqiang Tang, an engineering manager at Facebook who works on Tupperware and who was previously in charge of cloud automation research at IBM’s TJ Watson Research Center, tells The Next Platform that Facebook has no plans to take its learnings from Tupperware and then apply them and converge onto Kubernetes, as Google might someday do if it can. (Already, there are lots of Google services that run atop Kubernetes on Google Cloud Platform instead of on Borg/Omega on bare metal.)
     

    While Facebook has no current plans to open source the Delos low-latency, pluggable API data store that is being used with Tupperware, Jason Flinn, a professor of computer science and engineering at the University of Michigan who worked on the Delos project with Facebook, hinted that this project started only a year ago and has only been used in production for about four months, so it is early in the cycle to be opening it up, even if it is a possibility in the long term.

Servers Closing Down (MariaDB and IBM)

Filed under
Server
  • MariaDB Enterprise Server 10.4 Now Available, Pulumi Announces Pulumi Crosswalk for AWS, KDE Launches Plasma 5.16, IBM Announces Its List of Women Pioneers for AI in Business and Microway Provides Clemson University with an NVIDIA DGX-2 Supercomputer

    MariaDB today announces the release of MariaDB Enterprise Server 10.4, "code-named 'Restful Nights' for the peace of mind it brings enterprise customers". The press release notes that this version "is a new, hardened and secured Server (different from MariaDB Community Server aka MariaDB Server) and has never been available before. MariaDB Enterprise Server 10.4 includes features not available in the community version that are focused on solving enterprise customer needs, providing them with greater reliability, stability and long-term support in production environments."

  • MariaDB opens up on locked down Enterprise Server

    MariaDB finally took the wraps off its Enterprise Server product today, which it said was aimed at massive deployments where scale and security are more important than rushing out new features for developers.

    The product was first flagged up back in February, and scheduled to appear in “spring”. We’ll leave it to you to check your calendar and decide whether it hit its deadline.

    “It’s one thing to experiment with open source, to evaluate it,” said MariaDB’s senior director of product marketing Shane Johsnon. “But once you move to the stage where you’re migrating mission critical applications to it, or you’re deploying it at massive scale you encounter new types of challenges, not necessarily functional challenges.”

    So, security in MariaDB Enterprise Server 10.4 has been boosted, with end-to-end encryption for multi-master clusters. At the same time, the company has clamped down on plugins, with only those deemed tested and production-ready permitted.

    [...]

    MariaDB ES will be GPL, said Johnson, “It’s the community version that gets GA’d with additional QA and additional plugins. The result of that is Enterprise Server.”

  • Server Buying Cools, But It's Cool – Don't Panic

    When a market is comprised of hundreds of thousands of customers, things tend to level out and are a lot more predictable than when there are relatively few customers. Before the public clouds took off a decade ago and before the hyperscalers created such large infrastructures to support billions of users running their applications, server buying was a lot smaller and it was also more predictable. Things tended to grow slowly, methodically and they also took time to slow down because not everyone felt an economic decline or a transition to a new system architecture at the same time.

    That is no longer so with the modern server business. Enterprises are offloading some of their compute needs to the public clouds, and in other cases they are employing services provided by the hyperscalers – email, collaboration, and so on – instead of hosting them in their own datacenters. The hyperscalers and cloud builders are at the front of the line ahead of any new server chip generation, and they tend to buy aggressively ahead of the formal launches by Intel and AMD, and if the most recent quarter is any test, they are slowing down server purchases as they await the right time to invest in future chips from those two companies.

  • IBM i Roadmap Promises A Long Ride, Few Bumps

    It would be hard to find a group of enterprise IT shops that are more conservative – meaning averse to risk – than the IBM midrange. Arguably, IBM System z mainframe shops are even more risk averse, but perhaps it is a matter more of scale than degree. In the average IBM i shop, one person – or maybe a handful of people – is keeping risk at bay, while in a mainframe shop there could be dozens or hundreds that are trying to steer the ship without rocking the boat.

    Every now and then, Big Blue publishes an IBM i Strategy And Roadmap document to trying to calm the fears of the IBM i faithful while at the same time trying to fire them up a little. It is a delicate balance, and such documents are generally not full of information. But there are always some things to consider and that can be used to make the ongoing case that the IBM i platform deserves to be preserved in the enterprise and to have continuing investment. So it is with the 2019 edition of the IBM i Strategy And Roadmap, which you can get at this link.

Why containers and Kubernetes have the potential to run almost anything

Filed under
Server

The future of Kubernetes is bright, and like virtualization before it, workload expansion is inevitable. Learning how to drive Kubernetes is probably the biggest investment that a developer or sysadmin can make in their own career growth. As the workloads expand, so will the career opportunities. So, here's to driving an amazing dump truck that's very elegant at moving dirt...

Read more

Servers/Back End: Kubernetes, Apache Tomcat, SUSE and More

Filed under
Server
  • 4 tools to help you drive Kubernetes

    In the third article in this series, Kubernetes basics: Learn how to drive first, I emphasized that you should learn to drive Kubernetes, not build it. I also explained that there is a minimum set of primitives that you have to learn to model an application in Kubernetes. I want to emphasize this point: the set of primitives that you need to learn are the easiest set of primitives that you can learn to achieve production-quality application deployments (i.e., high-availability [HA], multiple containers, multiple applications). Stated another way, learning the set of primitives built into Kubernetes is easier than learning clustering software, clustered file systems, load balancers, crazy Apache configs, crazy Nginx configs, routers, switches, firewalls, and storage backends—all the things you would need to model a simple HA application in a traditional IT environment (for virtual machines or bare metal).

    In this fourth article, I'll share some tools that will help you learn to drive Kubernetes quickly.

  • Apache Tomcat 9.0.20 And 8.5.41 is Released, which Fixes many Bugs

    The Apache Tomcat Project is proudly announced the new release of version 9.0.20 of Apache Tomcat 9 branch and version 8.5.41 of Apache Tomcat 8 branch.

    Apache Tomcat 8.5.x replaces 8.0.x and includes new features, which back ported from Tomcat 9.0.x.

    The minimum Java version and implemented specification versions remain unchanged.

    The notable changes from both the versions are listed below.

  • Digital Transformation: An Interview with Rania Mohamed (SUSE Global Services Consultant)

    We’ve heard the mantra loud and clear.  Business must transform to become digital businesses to survive.  And the numbers are stark, with 80 % of businesses striving to make the transition by 2021.

    We sat down with SUSE Solutions Architect, Rania Mohamed, to get her perspective on what ditgital transformation is and how you can start your own transformation.

  • A New Era In Servers Is Starting Now

    You know the world is a different place when shipping 2.58 million servers in a quarter feels like a slowdown, a disappointment, and perhaps a leading indicator of an overall economic slowdown in the world.

    But the slight 5.1 percent downtick in server shipments in the first quarter of 2019 is probably more a function of the hyperscalers and cloud builders having spent a fortune on infrastructure in late 2017 and early 2018 and then waiting to see what the future holds for future processors from Intel and AMD.

  • Kubernetes Users Need Training: Mark Brandon, SuperGiant CEO

    With the explosive growth of Kubernetes adoption, there is a growing demand for training. “People want to use new tool kits that make their jobs easier, but they don’t know how to use them,” said Mark Brandon, founder and CEO of SuperGiant. “We have partnered with the Linux Foundation to create courses that help these users.”

    Brandon’s team created Supergiant toolkit created to “scratch our their itch,” as they say. Their cloud bills were too high because the hardware utilization was too low. They found that the so-called autoscaling tools from most vendors only meant it was easy to scale up.

  • Top 20 Best Big Data Tools and Software That You Can Use in 2019

    In our old days, we traveled from one city to another using a horse cart. However, nowadays is it possible to go using a horse cart? Obviously no, it is quite impossible right now. Why? Because of the growing population and the length of time. In the same way, Big Data emerges from such an idea. In this current technology-driven decade, data is growing too fast with the rapid growth of social media, blogs, online portals, website, and so forth. It is impossible to store these massive amounts of data traditionally. As a consequence, thousands of Big Data tools and software are proliferating in the data science world gradually. These tools perform various data analysis tasks, and all of them provide time and cost efficiency. Also, these tools explore business insights that enhance the effectiveness of business.

5 reasons to use Kubernetes

Filed under
Server

Kubernetes is the de facto open source container orchestration tool for enterprises. It provides application deployment, scaling, container management, and other capabilities, and it enables enterprises to optimize hardware resource utilization and increase production uptime through fault-tolerant functionality at speed. The project was initially developed by Google, which donated the project to the Cloud-Native Computing Foundation. In 2018, it became the first CNCF project to graduate.

This is all well and good, but it doesn't explain why development and operations should invest their valuable time and effort in Kubernetes. The reason Kubernetes is so useful is that it helps dev and ops quickly solve the problems they struggle with every day.

Following are five ways Kubernetes' capabilities help dev and ops professionals address their most common problems.

Read more

Syndicate content

More in Tux Machines

today's howtos

Leftovers: IBM, Mozilla and SUSE

  • What Is Razee, and Why IBM Open Sourced It
    The continuous delivery software that's been doing the heavy lifting on IBM's global Kubernetes platform is now open source.
  • View Source 5 comes to Amsterdam
    Mozilla’s View Source Conference is back for a fifth year, this time in Amsterdam, September 30 – October 1, 2019. Tickets are available now.
  • SUSE & SAP “A 20 years of Partnership”
  • SUSE on the IO500 List for HPC Storage
    If you haven’t been hanging around the Ceph world for a bit, you may not realize that Ceph was originally intended to provide a distributed file-system to service HPC clusters.  While this was the original intent, Ceph has taken a round-a-bout path to relevance in this space, especially given that we are only supporting multiple active MDS servers since the Luminous release.  The result is that we are, only now, really starting to see adoption in the HPC space, and mostly for the second tier storage needs. Enter, the science project.  Given an all-flash environment on SATA SSDS with a fast storage pool on Intel Optane for the metadata, would it be possible to provide a reasonable storage environment for HPC clusters?

GAFAM and 'Cloud': Google, Microsoft, Amazon and GitHub

  • Daniel Stenberg: Google to reimplement curl in libcrurl
    By throwing a lot of man power on it. As the primary author and developer of the libcurl API and the libcurl code, I assume that Cronet works quite differently than libcurl so there’s going to be quite a lot of wrestling of data and code flow to make this API work on that code. The libcurl API is also very versatile and is an API that has developed over a period of almost 20 years so there’s a lot of functionality, a lot of options and a lot of subtle behavior that may or may not be easy or straight forward to mimic. The initial commit imported the headers and examples from the curl 7.65.1 release.
  • Microsoft, you should look away now: Google's cloud second only to AWS in dev survey [Ed: Longtime Microsoft booster Tim Anderson  on Azure being a failure after so many entryism attempts and underhanded tactics]
    Coders use Google Cloud Platform (GCP) more than Microsoft Azure, though Amazon Web Services (AWS) has a comfortable lead, according to a Developer Ecosystem survey conducted by tools vendor JetBrains. Developer usage is 67 per cent AWS versus 28 per cent GCP and 21 per cent Azure, according to the new survey. Unfortunately, the question was posed in a different way in the 2018 survey, adding on-premises into the mix, but last year Azure and GCP had equal share after AWS. The survey had 19,000 participants invited via "Twitter ads, Facebook ads, Google Adwords and JetBrains' own communication channels," the tools vendor said, though "only the responses of 6,993 respondents were included in the report." Responses were removed to reduce bias, yet it warned "some bias may be present as JetBrains users may have been more willing on average to compete the survey".
  • Get your coat, you've pulled a Pull Panda: GitHub goes home with code collab specialists [Ed: Notice how Microsoft only takes GitHub in more of a proprietary software direction. That says a lot – they have plans and they’re really detrimental to FOSS]

Kernel: Linux Changes, Certifications, Graphics, PCI Express 6.0 and Bug

  • PowerCap/RAPL Code To Support Icelake Desktop / X / Xeon D With Linux 5.3
    While as of Linux 5.2 the support for Intel's Icelake CPUs appear production ready with all of the bits in place from new IDs to the much enhanced "Gen 11" graphics, there are a few stragglers of items to land with the upcoming Linux 5.3 merge window though could be back-ported to current series. Fortunately, we haven't found anything major to be missing. One of the latest bits of Icelake Linux support is handling of these next-generation processors within the PowerCap / RAPL (Running Average Power Limit) driver code. In particular, the desktop/workstation Icelake parts. This is the code for reading the estimated CPU package power consumption based on hardware performance counters and the ability to artificially limit the power draw of the processor via software.
  • Six Niche Linux Certifications
  • AMD Navi GPU stack bares all in Linux graphics driver update
    Eight Navi GPU variants have been spotted in Linux driver code. AMD’s next-gen RDNA graphics chips are set for launch on July 7, 2019 within the RX 5700 XT and RX 5700, but the red team has plenty of silicon in store for a range of applications. Including console, laptops, desktop, and mobile phones. The GPU codenames were spotted within Linux display drivers after the additional code was submitted and signed off by two AMD employees. The code adds support for Display Core Next, or DCN2, which “is the display block for Navi10.” Each entry following adds the necessary ASIC IDs for each Navi chip in the stack, starting with Navi 10 and down to Navi 21 LITE.
  • Nouveau Driver Picking Up NVIDIA TU116 GPU Support For Linux 5.3
    Building off the initial Turing mode-setting bits that were in place since Linux 5.0 and have continued stepping along to support newer variants on successive kernel releases, the Linux 5.3 kernel is slated to add support for the TU116 graphics processor.
  • PCI-SIG® Announces Upcoming PCI Express® 6.0 Specification to Reach 64 GT/s
  • PCI Express 6.0 Announced With 4-Times The Bandwidth Of PCIe 4.0
    With the increasing demand for bandwidth across a wide range of devices used in consumer and enterprise domains, PCI Express, the high-speed serial computer expansion bus standard has also evolved over the years. PCI Special Interest Group, a body that sets standards for PCIe, has announced PCI Express 6 that promises four times the bandwidth offered by PCIe 4.0 and twice of PCIe 5.0.
  • PCI Express 6.0 Announced For Release In 2021 With 64 GT/s Transfer Rates
    While PCI Express 4.0 up to this point has only been found in a few systems like Talos' POWER9 platforms and coming soon with the new AMD graphics cards and chipsets, the PCI SIG today announced PCI Express 6.0. PCI Express 5.0 was only announced last month with 32GT/s transfer rates while already the PCI SIG announced PCI Express 6.0.
  • Netflix researcher spots TCP SACK flaws in Linux and FreeBSD
  • TCP SACK Panic Flaw Could Compromise Production Linux Machines