Red Hat A Robust Platform for the Scientific User
Red Hat offers packages specifically for the scientific high-performance computing (HPC) user to deploy clusters of systems to tackle the most challenging missions. Subscriptions for Red Hat
Enterprise Linux in clusters are affordable and easy to deploy, and support scale-out systems. With Red Hat´s subscription pricing for your high-performance cluster´s nodes, plus
support costs tied to the cluster head node, Red Hat Enterprise Linux for Scientific Computing costs less. If you have an HPC cluster managed as a single scale-out system, we make it simple for you
to run Red Hat Enterprise Linux everywhere. With its excellent ability to detect and optimize hardware configurations, Red Hat Enterprise Linux quickly and easily gets the most from your hardware.
Red Hat´s deep relationships across the industry translate into a platform optimized for the latest capabilities.
Red Hat Enterprise MRG Grid provides high-throughput (HTC) and high-performance computing (HPC). And it lets you move to a cloud-based grid computing model to achieve higher peak computing capacity and higher IT usage by using your existing infrastructure to build high-performance grids. The MRG Grid scheduler (Scalable) is based on Condor, which powers many of the largest grids in the world and easily scales beyond tens of thousands of nodes. A mechanism known as flocking lets independent pools (Federated Grids/Clusters) use each others´ resources, controllable by customizable policies. A web service interface (Multiple standards-based APIs) provides job submission and management functionality. A command-line interface (CLI) provides a highly scriptable-with consistent output-interface to all functionality. Ace Computers is a Red Hat Ready Partner, and can implement your Red Hat solution across our LogiCAD workstations, Powerworks servers, and HPC systems.
SUSE Mission-Critical Workloads
Ace Computers is an OEM Partner with SUSE, and provides complete HPC Solutions built around SUSE Linus Enterprise Server. With support for high-speed interconnects, Open Fabrics Enterprise Distribution (OFED) support enables affordable, high-speed remote direct memory access (RDMA)
capable fabrics and unified interconnects based on InfiniBand and 10 Gigabit Ethernet. Support for Fiber Channel over Ethernet (FCoE) allows existing fiber channel storage to be accessed using an Ethernet adapter.
Data Center Bridging (DCB) enhancements allow network traffic with differing requirements to operate and co-exist on Ethernet networks.
Advanced power-saving features maximize performance per watt, while tickless idle reduces system power consumption and lowers energy costs. When the CPU is idle, tickless idle eliminates the periodic timer in the Linux kernel, allowing the CPU to remain in power-saving states longer. More granular, file-based power profiles simplify enterprise-wide power management, and support for the Processor Clock Control (PCC) interface driver ensures complete processor state information is shared between the operating system and hardware BIOS. With more efficient algorithms, you gain greater control over CPU speeds and hardware power management, maximizing performance per watt.
Cluster-ready means SUSE Linux Enterprise Server is "Intel Cluster Ready." This OS supports advanced memory management; native POSIX thread libraries; highly scalable, multi-threaded file systems like XFS; advanced multi-pathing; and high-speed interconnects. The rich feature set makes SUSE Linux Enterprise Server the overwhelming favorite for high-performance computing clusters worldwide.
SP2 kernel improvements increase the performance of memory and compute intensive workloads through support for transparent huge pages. Support for transparent per-CPU load balancing on multi-queue devices and faster packet filtering improve network performance. Enhanced control groups help you get the most out of your systems. Contact your Ace Account Team to talk about Expanded Support Options with SUSE and Ace.
CentOS An Enterprise-Class Linux Distribution
CentOS is an Enterprise-class Linux Distribution derived from sources freely provided to the public by a prominent North American Enterprise Linux vendor. CentOS conforms fully with the upstream vendor´s redistribution
policy and aims to be 100% binary compatible. (CentOS mainly changes packages to remove upstream vendor branding and artwork.) CentOS is free. CentOS is developed by a small but growing team of core developers.
In turn the core developers are supported by an active user community including system administrators, network administrators, enterprise users, managers, core Linux contributors and Linux enthusiasts from around
the world. CentOS has numerous advantages over some of the other clone projects including: an active and growing user community, quickly rebuilt, tested, and QA´ed errata packages, an extensive mirror
network, developers who are contactable and responsive, multiple free support avenues including IRC Chat, Mailing Lists, Forums, and a dynamic FAQ. Our purpose is to provide stable Linux solutions
for organizations and individuals who do not need strong commercial support to achieve successful operation. Ace Computers has been supplying CentOS based solutions to Federal Government Agencies and Universities for many years, and our
engineering team is very comfortable installing and configuring the latest CentOS releases, including initial driver support and integration with commercial cluster management toolsets from StackIQ or Bright Computing.
CentOS goals include easy maintenance, self hosting distribution (one that can build on itself), suitability for long term use in production environments, friendly environment for users and package maintainers, long-term support of the core, active development, community infrastructure, open management, open business model and commercial support - offered by partner vendors (including Ace Computers).
Ubuntu Performance and Versatility
Fast, secure, deploy-anywhere technology for fast-moving companies from Ubuntu. Secure, fast and powerful, Ubuntu helps you make the most of your infrastructure. Whether you want to deploy a web farm
or deploy a cloud, Ubuntu Server supports the most popular hardware and software. Our regular release cycle helps you keep pace with the latest tech developments. And our lean initial install, with thousands
of easy-to-find packages, makes it a great solution for simple deployment and management at scale.
What´s new in Ubuntu Server 12.04 LTS includes significant changes, designed to help you deliver new services faster and with more flexibility than ever. Do note that the packages in the main repository for every LTS release are supported for five years from launch, so there is no pressure to upgrade from Ubuntu Server 10.04 LTS. To help you quickly understand the updates included, we have categorised the improvements as follows: cloud and virtualisation; hardware and architecture support; file systems and storage; and new ISV support. Cloud and virtualisation (Standards-based agility, flexibility and compatibility) means Ubuntu Cloud is central to Ubuntu Server 12.04 LTS - and it will remain at the heart of platform technology decisions in future releases. Ubuntu Cloud Infrastructure is now powered by OpenStack, the default choice. OpenStack has been integrated as the default cloud infrastructure technology for Ubuntu Cloud, for the following reasons: A growing open source project in terms of partners, in terms of contributions (120 companies involved in the Essex release), in terms of time to market (five major public clouds in beta by end of 2011), and it is built to support the needs of public clouds (which supersede private). Built to prohibit any single point of failure, massively extensible in terms of functionality (SAN, network, hypervisors, etc.), excellent dissociation of components, allowing for a more scalable and maintainable architecture, so the current release of OpenStack in Ubuntu Server 12.04 LTS is Essex.
Scientific Linux Open Source Expertise
SL is a Linux release put together by Fermilab, CERN, and various other labs and universities around the world. Its primary purpose is to reduce duplicated effort of the labs,
and to have a common install base for the various experimenters. The base SL distribution is basically Enterprise Linux, recompiled from source.
Our main goal for the base distribution is to have everything compatible with Enterprise, with only a few minor additions or changes. Examples of items that were added are Alpine, and OpenAFS.
Our secondary goal is to allow easy customization for a site, without disturbing the Scientific Linux base. The various labs are able to add their own modifications to their own site areas.
By the magic of scripts, and the anaconda installer, each site is to be able to create their own distributions with minimal effort. Or, if a user wishes, they can simply install the base SL release.
Scientific Linux Customizations made to the standard Enterprise release primarily include the following: Because the main goal of Scientific Linux is to have everything compatible with Enterprise, we have kept our modifications to a minimum. But we have made changes. The majority of the changes are not seen, or merely cosmetic. Sites / Spins are our biggest customization, the ability to make ´sites´. Nowadays they are called ´spins´. These spins allow a facility to customize Scientific Linux to meet their own needs. They can customize the install program, rpm´s that are installed, configurations, almost anything they can think of, without ever disturbing the base Scientific Linux distribution. A naming convention was agreed upon when creating spins. If you create your own spins, you should name it Scientific Linux (spins name), such as Scientific Linux Edge. Some of the most popular spins are Scientific Linux Fermi and Scientific Linux CERN. If you do not want to share your spin with the rest of the world, you do not have to. Ace Computers has a long history of supplying HPC systems to Fermi Lab and other Federal Government and University installations, many of which are utilizing Scientific Linux.