Home Blog Page 1923

TACC’s Hadoop Cluster Makes Big Data Research More Accessible

Over at the Texas Advanced Computing Center, Aaron Dubrow writes that researchers are using a specialized cluster at TACC to do experimental Hadoop-style studies on a current production system.

This system offers researchers a total of 48, eight-processor nodes on TACC’s Longhorn cluster to run Hadoop in a coordinated way with accompanying large-memory processors. A user on the system can request all 48 nodes for a maximum of 96 terabytes (TB) of distributed storage. What’s special about the Longhorn cluster at TACC isn’t simply the beefed-up hardware for running Hadoop; rather it’s the ability for researchers to leverage the vast compute capabilities of the center, including powerful visualization and data analysis systems, to further their investigations. The end-to-end research workflow enabled by TACC could not be done anywhere else, and as a bonus, researchers get access to the full suite of tools available at the center to do computational research.

According to TACC Research Associate Weijia Xu, the best part is that Hadoop is easy to use without requiring users to be experts. It handles a lot of the low-level computing behavior, so people don’t need to have a lot of knowledge about I/O or memory structures to get started.

Read the Full Story.

Related posts:

 

The post TACC’s Hadoop Cluster Makes Big Data Research More Accessible appeared first on insideHPC.

 
Read more at insideHPC

Firefox OS: Go Away Fanbois, Fandroids – You Wouldn’t Understand

Repeat after me: developers, developers, developers, developers…

Hands-on  The Western world’s smartphone market has devolved into a duopoly of Apple’s iOS and Google’s Android. In the rest of the world, however, the mobile story has yet to be written… and this is where Mozilla hopes users will embrace its mobile operating system, Firefox OS.…

Read more at The Register

Chakra 2013.05 Gets Graphical Package Manager

The latest edition of Linux distribution Chakra includes a new default graphical package management interface. The Arch-Linux-based distribution now relies entirely on KDE 4.10 and associated applications

Read more at The H

Samsung Unveils 8-inch and 10.1-inch Galaxy Tab 3 Android Tablets, Coming in June

Galaxy_tab_3_10

Samsung has added two new Android tablets to its Galaxy Tab 3 range, introducing new 8-inch and 10.1-inch models that complement its existing 7-inch device. The 8-inch Galaxy Tab 3 features a 1.5 GHz dual core processor, WXGA TFT 1280 x 800 (189 PPI) display, 1.5GB of RAM, a 5-megapixel rear-facing (1.3-megapixel front-facing) camera, a 4,450 mAh battery, and 16/32GB of internal storage.

It’s bigger counterpart has a 1.6 GHz dual core processor, a 1280 x 800 (149 PPI) WXGA TFT display, 1GB of RAM, and 6,800 mAh battery. However, the 10.1-inch tablet only features a 3-megapixel rear-facing camera, but does include the same 1.3-megapixel camera as the 8-inch Tab. Both variants run Android 4.2.2 and feature Samsung’s suite of TouchWiz…

Continue reading…

Read more at The Verge

Asus Announces Transformer Book Trio, Runs Windows 8 and Android With Two Intel CPUs

At Computex 2013, Asus announced the Transformer Book Trio, a device it calls the “world’s first three-in-one notebook, tablet, and desktop PC.” The Transformer Book Trio is designed to instantly switch between Windows 8 and Android (Jelly Bean), offering a notebook dock “for work,” and basic tablet functionality “for play.” It’s powered by dual Intel processors, including a core i7 Haswell CPU and “the highest-performance, yet power-efficient” Atom. Asus says the Transformer Book Trio will offer up to 15 hours battery life, 1TB storage, and 64GB SSD storage.

Continue reading…

Read more at The Verge

Haswell CPUs to Deliver 7-15W TDP, Says Intel

Intel has released new information on its more power-efficient next generation “Haswell” family of Core processors. Quad-core Core i7 Haswell CPUs will offer 15W TDP power consumption, down from 20W on similar Ivy Bridge processors, resulting in up to 9.1 hours of HD playback, while future tablet-ready dual-core parts could lower power consumption by up […]

Read more at LinuxGizmos

ARM’s New Cortex-A12 is Ready to Power 2014’s Midrange $200 Smartphones

Xv02-27_12-45-0420_large

We already know what ARM has planned for 2014’s high-end smartphones, but what about cheaper handsets? The company who designs many of the processor cores inside today’s mobile devices is preparing new mid-range silicon that it believes can offer increased performance in phones that could cost as little as $200 off-contract. The new Cortex-A12 core will offer 40 percent more performance as the existing Cortex-A9 which appears in chips like today’s Tegra 3 and Snapdragon S4, though it won’t be quite up to the standard set by the Cortex-A15 you’ll find in devices that have Samsung’s Exynos 5250 or Nvidia’s Tegra 4, to say nothing of next year’s Cortex-A57 based silicon.

 

Continue reading…

Read more at The Verge

Windows 8 Continues to Fail

Microsoft can only hope that it’s re-invention of Windows 8, Windows 8.1 aka Blue, works because Windows 8 continues to fall behind even Vista’s dismal desktop operating system market acceptance numbers. As for the mobile operating system space, Microsoft needs a miracle.

Dell Goes Evergreen

Investing in R&D remains a top priority and it seems to be paying off

Intro to OpenStack Part Two: How to Install and Configure OpenStack on a Server

This is the second article in our Introducton to OpenStack series. If you’re interested in more in-depth professional training, The Linux Foundation offers OpenStack training courses.

Last week we learned what OpenStack is and what it does. Today we’ll install it on a single machine and make it do stuff. This is not how you would set up a production server, but it’s a wonderful fast way to get a testing and learning server running.

There are three fairly easy ways to get your hands on OpenStack and try it out: one is to use a commercial public cloud like Rackspace or Cloudwatt, or the free Trystack. If you’re in a hurry go for one of the paid services, because it can take days to weeks to get approved for a Trystack account. Using a public cloud is a good way to dive right into developing and testing applications.

A successful OpenStack installation.If you’re more interested in spelunking into the guts of OpenStack and learning how to administer it then you can build your own server to play with, and that is what we’re going to do with the DevStack installer. DevStack is an amazing shell script that installs the OpenStack components, a LAMP stack and CirrOS, which is a tiny Linux distro built for running cloud guests. (Cirrus? Get it? Finally a good geek pun.) I am going to cover installation in detail, because even though it’s easier than it’s ever been it’s still a bit tricksy.

Getting Started

With most Linux applications it’s safe to install and remove and play with whatever you want to test on your main Linux PC, because Linux is a grown-up operating system that does not keel over when you ask it to do work. Unlike certain overpriced proprietary operating systems that are delicate and full of excuses. But I digress.

Don’t put OpenStack on your main PC because it needs a dedicated system, so for this article I’m running it in VirtualBox on Lubuntu 12.04 on my Linux Mint 13 system. Sure, I know, real server nerds don’t run a graphical environment on their servers, but for testing it’s a nice convenience, and Lubuntu is lightweight. If you elect to run OpenStack server in a virtual machine give it a minimum of 1.5GB RAM and 6GB storage. If you have a multicore system and can spare more than one core, do so because OpenStack, even in a simple testing setup, gets hungry.

First create a user named stack to use for installing DevStack:

$ sudo useradd stack
$ sudo passwd stack
Enter new UNIX password: 
Retype new UNIX password: 

Then give stack full sudo privileges:

$ sudo visudo
stack ALL=(ALL:ALL) NOPASSWD: ALL

Now logout, and then log back in as stack. If you don’t have git then install it:

$ sudo apt-get install git -y

Then pull OpenStack from Github. This copies it into the current directory, so I cd to /var and then run git:

$ git clone git://github.com/openstack-dev/devstack.git

This puts everything in /var/devstackcd to devstack, and take a few minutes to look in the various scripts and files. For whatever reason, which I have not figured out, I ran into permissions problems on my first run, so I changed ownership of /var/devstack and /opt/stack to stack:

$ sudo chown -R stack:stack /opt/stack
$ sudo chown -R stack:stack /var/devstack

I also changed /var/www to www-data:www-data; Ubuntu’s default is root, which is not a good practice.

It is good to have logging, so create /var/stacklog, and make it owned by stack.

Configuration

There is one more prerequisite, and that is to create /var/devstack/localrclocalrc always goes in your DevStack root, and it configures networking, passwords, logging, and several other items we’re going to ignore for the time being. This is what mine looks like, just a minimal configuration:

HOST_IP=10.0.1.15
FLAT_INTERFACE=eth0
FLOATING_RANGE=10.0.1.224/28
ADMIN_PASSWORD=supersecret
MYSQL_PASSWORD=supersecret
RABBIT_PASSWORD=supersecret
SERVICE_PASSWORD=supersecret
SERVICE_TOKEN=supersecret

OpenStack uses a lot of passwords, so for testing I make it easy on myself by recycling the same one. The HOST_IP is the ethX inet addr of your OpenStack server, whether it’s virtualized or not, like this example:

$ ifconfig
eth0  Link encap:Ethernet  HWaddr 90:ee:aa:a2:50:aa  
      inet addr:10.0.1.15  Bcast:10.0.1.255  Mask:255.255.255.0

Do create a static IP address for your DevStack server, or you will suffer. Networking is rather involved for OpenStack, and we’ll get into that more in the future; for now we’ll keep it as simple as possible.

FLAT_INTERFACE is the server’s Ethernet interface; if you have just one it’s not necessary to include this line. You could have an internal and a public-facing interface, just like on non-cloud servers, and the FLAT_INTERFACE corresponds to the internal interface.

FLOATING_RANGE is a pool of addresses for any OpenStack servers that need to be available to the network. This must not overlap with the server’s IP address, which is why my example is way out at the end of the address range.

The Horizon dashboard, after OpenStack installation.Alrighty then, it’s time to finish the installation. Change to /var/devstack and run:

$ ./stack.sh

This will run for a while and fill your screen with all kinds of output. Go take a nice break and think about pleasant things. When it completes a successful run you’ll see something like figure 1, above.

Now fire up a Web browser on your OpenStack server and point it to the IP address it told you, which in my example is http://10.0.1.15. If you see the login page you may congratulate yourself for a successful installation, and for accessing the Horizon dashboard (figure 2.) Go ahead and login as admin with whatever password you set in localrc. You can poke around and explore the different screens without hurting anything. There isn’t much to see yet, but you’ll find a few images and report pages.

If you make a mess, the good DevStack people included a do-over script, clean.sh. This reverses stack.sh and leaves your git clone files in place, so run clean.sh and then stack.sh to re-do your installation.

That’s all for today, so come back next time to learn how to access Horizon from a remote PC, and how to do some basic setup.