Home Blog Page 951

Seagate and Newisys Demonstrate 1 TB/s Flash Architecture

Today Seagate and Newisys announced a new flash storage architecture capable or 1 Terabyte/sec performance. Designed for HPC applications, the “industry’s fastest flash storage design” comprises 21 Newisys NSS-2601 with dual NSS-HWxEA Storage Server Modules deployed with Seagate’s newest SAS 1200.2 SSD drives. These devices can be combined in a single 42U rack to achieve block I/O performance of 1TB/s with 5PB of storage. Each Newisys 2U server with 60 Seagate SSDs is capable of achieving bandwidth of 49GB/s.

The post Seagate and Newisys Demonstrate 1 TB/s Flash Architecture appeared first on insideHPC.

 
Read more at insideHPC

Android Studio 2.0 Aims to Rev App Development

Google last week released a preview of Android Studio 2.0 at the inaugural Android Dev Summit in Mountain View, California. The preview offers several updates, including Instant Run and a new GPU Profiler. It’s available in the Canary release channel for the Linux, Mac and Windows platforms. Instant Run lets developers see changes to code running on their devices or emulators, while the GPU Profiler lets users profile their OpenGL ES Android code.

Read more at LinuxInsider

Altair to Open Source PBS Professional HPC Technology in 2016

altair2

Altair has announced that it will provide an open source licensing option of PBS Professional® (PBS Pro). PBS Pro will become available under two different licensing options for commercial installations and as an Open Source Initiative compliant version. Altair will work closely with Intel and the Linux Foundation’s OpenHPC Collaborative Project to integrate the open source version of PBS Pro.

The post Altair to Open Source PBS Professional HPC Technology in 2016 appeared first on insideHPC.

 
Read more at insideHPC

OpenMW 0.37 Released, Switches To OpenSceneGraph

The open-source community working on OpenMW as an engine re-implementation of Elderscrolls III: Morrowind have announced the release of OpenMW v0.37…

Read more at Phoronix

Intel Lands First Round Of Graphics Work For Linux 4.5, Includes Kaby Lake

The first batch of Intel DRM graphics driver changes targeting the Linux 4.5 kernel have landed into DRM-Next…

Read more at Phoronix

More Hands-on With the Raspberry Pi Zero: Loading, Booting and Configuring

Today I share more information and first-hand experiences with the Raspberry Pi Zero, including loading, booting, configuring and using the PiHub for both USB expansion and power.

Read more at ZDNet News

How to Install Unity 8 on Ubuntu 16.04 LTS and Ubuntu 15.10

We’ve created the following tutorial at the request of our readers who asked as to post some easy-to-follow instructions on how to install the next-generation Unity 8 user interface on the Ubuntu Linux operating system.

The fact of the matter is that the latest Unity 8 packages, including the next-gen Mir display server, are available in the default software repositories of Ubuntu 16.04 LTS (Xenial Xerus), which is currently under development, and Ubuntu 15.10 (Wily Werewol… (read more)

Getting Started With Docker by Dockerizing This Blog

Docker is an interesting technology that over the past 2 years has gone from an idea, to being used by organizations all over the world to deploy applications. In today’s article I am going to cover how to get started with Docker by “Dockerizing” an existing application. The application in question is actually this very blog!

What is Docker

Before we dive into learning the basics of Docker let’s first understand what Docker is and why it is so popular. Docker, is an operating system container management tool that allows you to easily manage and deploy applications by making it easy to package them within operating system containers.

Posted by Benjamin Cane

 
Read more at bc-log

Linux Cli: Background and Foreground Process

Linux Background and Foreground Process

So we know there is a first process named ‘init’ with pid. This is parent 
of all process in the system. And a process named ‘bash’ interacts with
Kernel on behalf of user requests or commands.

Now when I log in and type ps – I get below output:

$ ps
  PID TTY          TIME CMD
 5254 pts/1    00:00:00 bash
 5336 pts/1    00:00:00 ps

See this is different from our previous output. Everytime you login,new
parent bash is created.In case our Manager Bash job id is 5254.
You know each command is a process, right? Lets create few process
for this session.

Type

sleep 5

Did it hang for 5 seconds and then provide you the bash prompt again?

The way, our bash creates child process is blocking call. It means,
run the child process and wait for it to complete and then return
to me again.

Background process

So when we ran our child process (sleep), ‘Bash’ shell waited for
it to finish. User request below commands:

sleep 5
sleep 2

Shell will run sleep 5 first and wait for it to end. Then it
runs sleep 2. But what if job-2 in this case ‘sleep 2’ is 
more critical than first job. There is unnecessary delay of
5 seconds right? Shell has an option of running child process
on background – that means it wont wait for child to finish
before accepting inputs from user.

We can put any child process in background by just appending
& character to it!

Lets try that:

$ sleep 5 &
[1] 5781
$ sleep 2

Look, we can ran sleep 2 without hanging for 5 seconds! and the
number you see 5781 is the background child process id.
Its for our reference. How to verify its indeed the pid of child?
Simple just start child process in background and execute ‘ps’ command.

$ sleep 5 &
[1] 6095
$ ps
  PID TTY          TIME CMD
 5254 pts/1    00:00:00 bash
 6095 pts/1    00:00:00 sleep
 6099 pts/1    00:00:00 ps

Our sleep 5 pid matches with ps output. Do we have better way to visualize
this? yes do. We have a command named ‘pstree’ which will tell you the
mapping between child and parent process!!

Lets try again, note down your Bash shell pid (here its 5254)
$ pstree 5254
bash───pstree

It tells, we have Bash and one child process named ‘pstree’.run our child,

$ sleep 5 &
[1] 6208
$ pstree 5254
bash─┬─pstree
     └─sleep

Now it says, ‘Bash’ has two child named, pstree and our background process ‘sleep’
pstree has an option to mention -p which displays pid next to process name.

$sleep 5 &
[1] 6272
$ pstree -p 5254
bash(5254)─┬─pstree(6276)
    	   └─sleep(6272)

the pid we got while starting background process is same as what we got
from pstree output.

Lets say we have started 4 long running background jobs each runs for 145 seconds and
1 very long process for 3000 seconds.

$sleep 45 &
[1] 6393
$sleep 45 &
[2] 6397
$sleep 45 &
[3] 6401
$sleep 45 &
[4] 6406
$ sleep 3000 &
[5] 6557
$pstree -p 5254
bash(5254)─┬─pstree(6410)
       ├─sleep(6393)
       ├─sleep(6397)
       ├─sleep(6401)
       └─sleep(6406)
   ├─sleep(6557)

List background jobs

pstree gives information about all jobs. We don’t need pstree(6410), because we are
interested in only background jobs. How to view only those jobs. For this purpose, we have ‘jobs’ command will give output like:

 jobs
[1]   Running                 sleep 145 &
[2]   Running                 sleep 145 &
[3]-  Running                 sleep 145 &
[4]+  Running                 sleep 145 &
[5]+  Running                 sleep 3000 &

It lists only our background processes and its status. So far good right?

Foreground process

Our process-[5] runs for 3000 seconds, it takes long time to complete. 
Background will take less CPU time compared to non-background process
.ie foreground process. So lets bring to foreground process.
type, fg in our case, we need to bring background job-5.

fg 5
sleep 3000

Can you see shell hanging now? wait..hanging is a wrong word to use. shell executing
sleep command now?  Are you going to wait for 3000 seconds? aka 50 minutes?

Lets admit it, by mistake without thinking, you (yes, its you !)brought this 
background process to foreground, now you desperately want to put in on background again!!

Switch between foreground to background

Dont worry, Linux is so flexible we can do that too. Just press ‘ctrl+z’.You get
output which says job is stopped.

^Z
[5]+  Stopped                 sleep 3000

verify the status by checking output of jobs command.

$jobs
[5]+  Stopped                 sleep 3000

Its stopped, we can restart the process again in background with ‘bg

$bg 5
[5]+ sleep 3000 &
$jobs
[5]+  Running                 sleep 3000 &

Great, Today we learned about background and foreground process and how to move
between them. We will discuss in upcoming lessons.

 

Read more: http://community.webminal.org/c/commandline/bash

IO Visor: Bringing the Network to the Linux Kernel

Valentina Alaria PLUMgridAs enterprise customers are moving into software-defined everything, including compute, storage and networking, new challenges are emerging. And since open source plays a very critical role in today’s stack, it becomes increasingly important for players to deal with such challenges in an open source manner.

We are witnessing a trend in which more and more proprietary products are being released as open source. However, companies are not just throwing the code at the community; they are actually creating collaborative projects so that the project gets the best brains to work on it, and in most cases those best brains come from the very competitor of the company open sourcing it. We have seen projects like OpenStack, Cloud Foundry, and the Open Compute Project where arch rivals are working together to improve the stack.

We haven’t before seen such an unprecedented collaboration between commercial players. The credit goes to the Linux Foundation, which has made big players comfortable with open source and collaboration.

IO Visor is one such project which brings different players together to address some serious issues around networking. IO Visor was initially developed by PLUMgrid, an SDN (software defined network) startup founded by Cisco engineers. The company then worked with The Linux Foundation to open source the project. Being a Linux Foundation Collaborative Project it now enjoys contributions from companies such as Barefoot Networks, Broadcom, Canonical, Cavium, Cisco, Huawei, Intel, and SUSE.

According to the project it brings universal IO extensibility to the Linux kernel and enables infrastructure/IO developers to create, publish, and deploy applications in live systems without having to recompile or reboot kernel code.

I recently interviewed Valentina Alaria, chair of the IO Visor project’s marketing steering committee and senior director of product solution marketing at PLUMgrid. Here she discusses the project, why it became a Collaborative Project, and what that means for project governance and development.

Swapnil Bhartiya: Can you tell us about the project?
Valentina Alaria: The IO Visor project is a Linux Foundation Collaborative Project chartered to create an open source, technical community where industry participants easily contribute to and adopt the IO Visor project’s technology for an open programmable data plane for modern IO and networking applications.

The IO Visor Project provides universal IO extensibility for the Linux kernel.

With a programmable data plane and development tools to simplify the creation and sharing of dynamic IO modules, IO Visor Project enables a new way to innovate, develop and share IO and networking functions within the community.

Why did you choose to become a Linux Foundation Collaborative Project? What benefits do projects like IO Visor and participating companies receive by being a collaborative project?
Alaria: Becoming a Linux Foundation Collaborative Project allows the IO Visor to continue to work with the large developer community that has been actively involved with the project for several years as well as to reach out to new community members, which we are seeing growing on a regular basis. IO Visor Project provides a broad set of benefits to any company leveraging Linux to build out data centers and clouds to run agile applications.

Can you tell us about the governance model of the project?
Alaria: IO Visor Project adheres to the governance model of Linux Foundation Collaborative Projects with a board, technical steering committee, marketing, legal, and finance operations. IO Visor Project board members are platinum sponsors with one silver member for every five silver sponsors. You can find more details here: https://www.iovisor.org/about/join

How does the project receive and manage funding to run the project?
Alaria: IO Visor Project receives funds through two levels of sponsorship: platinum and silver. The board is responsible for allocating use of these funds.

How do you engage with the open source community and Linux in general? What is your collaboration with different projects?
Alaria: There are a number of open source projects that are focused on new network functions, management/orchestration, improving ease of use or performance in different areas of the networking stack. IO Visor provides the in-kernel enhancements that these projects will leverage and expedites developer access to IO capabilities that may not be available today. By providing programmable data plane abstraction, in-kernel capabilities, and tools, the IO Visor Project will help developers in the networking space achieve faster innovation.

IO Visor Project is still in its early days but it had a presence at both the most recent OpenStack Summit in Tokyo as well as the first OPNFV Summit in San Francisco, showing good synergy between those projects.

How do the member companies share their resources — mainly engineering with fellow members who also happen to be competitors in the market?
Alaria: IO Visor Project member companies are contributing equally the resources needed to support all areas (mainly technical, thought leadership and marketing). The TSC provides a key function of funneling short and long term goals and aligning members around those. We are also lucky to have a community of developers that is already actively contributing code to the project.

Open source has kind of become a norm, a standard for development. More and more companies are collaborating with partners and competitors. What’s driving this adoption?
Alaria: The power of the community is what makes open source initiatives extremely powerful and transformational. The ecosystem of solutions and vendors that then blossoms around each community enables users to adopt such technology across a variety of use cases and environments.