IO optimizations and data access: Difference between revisions

From Gridkaschool
Jump to navigationJump to search
(Created page with "== Technical specification/requirements == {| class="wikitable" width="85%" style="border-collapse: separate; border-spacing: 0; border-width: 1px; border-style: solid; border-c…")
 
 
(13 intermediate revisions by 3 users not shown)
Line 1: Line 1:
== Technical specification/requirements ==
===[[Internals:IO|Technical specification/requirements]]===


Every participant will have a private node to run the exercises. It is '''not possible''' nor needed '''to run the exercises on a personal laptop''' because results are not comparable.
{| class="wikitable" width="85%" style="border-collapse: separate; border-spacing: 0; border-width: 1px; border-style: solid; border-color: #000; padding: 0"

! colspan="6" style="background:#ABE" | IO optimizations and data access
The exercises will be provided on a dynamic web page with hints and solutions. The link will be published on Tuesday.
|-

| colspan="6" style="background:#ffffcc" | Andreas-Joachim Peters (CERN) (in progress)

|-
== Exercise Overview ==
! style="width: 5%; border-style: solid; border-width: 1px 1px 0 0" | # of nodes

! style="width: 20%; border-style: solid; border-width: 1px 1px 0 0" | Hardware

! style="width: 25%; border-style: solid; border-width: 1px 1px 0 0" | Software/Utilities
The tutorial will start with an introduction to tools used in Linux environments to measure and analyze IO performance. Based on few examples participants will be able to measure individual IO bottlenecks present in hard- and software. The second part will demonstrate optimizations in various parts of the local storage architecture like write-caching, read-ahead, compound reads etc. The third part will be an exercise illustrating the power and potential of Cloud storage where students will apply the technology of distributed hash tables to build a private cloud storage system.
! style="width: 30%; border-style: solid; border-width: 1px 1px 0 0" | Miscellaneous

! style="width: 10%; border-style: solid; border-width: 1px 1px 0 0" | Status (ToDo/DONE)
[[ File:IO-Exercises.pdf]]
! style="width: 10%; border-style: solid; border-width: 1px 0 0 0" | Tested (YES/NO)
[[ File:IO-Exercises+Hint.pdf ‎ ]]
|-
[[ File:IO-Exercises+Hint+Solution.pdf]]
| style="border-style: solid; border-width: 1px 1px 0 0" align="center" | 16
[[ File:DataTechnologies-Exercises-3.pdf‎]]
| style="border-style: solid; border-width: 1px 1px 0 0" |
[[ File:cloud.sh]]
* HDD: ??GB

* RAM: ??GB
== Practical Information==
* ...
==== Assigned Exercise Machines ====
| style="border-style: solid; border-width: 1px 1px 0 0" |
<pre>
* vmstat, dstat, iostat, wget, curl, ROOT, gcc compiler
gks-126 Markus Alex
| style="border-style: solid; border-width: 1px 1px 0 0" |
gks-127 Alexander Ditter
* open at least one port on each machine where other nodes can connect
gks-128 Bogdan Dumitru
* port number does not matter, could be 8080
gks-129 Alexander Voigt
* if 16 nodes not available, two people sit together i.e. 8/9 nodes
gks-130 Alexander Yasnogor
| style="border-style: solid; border-color: #000; border-width: 1px 1px 0 0; color: red;" align="center" | ToDo
gks-131 Michael Bontenackels
| style="border-style: solid; border-color: #000; border-width: 1px 0 0 0; color: red;" align="center" | NO
gks-132 Herbert Cornelius
|}
gks-133 Daniel Funke
gks-134 Christian Gumpert
gks-135 Dominik Haitz
gks-136 Silke Halstenberg
gks-137 Christopher Jung
gks-138 Barbara Krasovec
gks-139 Roland Laifer
gks-140 Patrick Martin
gks-141 Maritn Spoo
gks-142 Sang Oh Park
gks-143 Stephan Pfister
gks-145 Angela Poschlad
gks-147 Christian Pulvermacher
gks-037 Doris Ressmann
gks-038 Martin Spoo
gks-039 Alexander Voigts
gks-042 Andreas Wirooks
gks-043 Manuel Zeise
</pre>
If possible create an SSH tunnel from your laptop to your assigned exercise machine like

<pre>
ssh -p 24 -L 8000:gks-017:80 gks115@gks-144 # => username@exercisemachine
</pre>
Open the follwing address in your browser: [http://localhost:8000/dm http://localhost:8000/dm]
If you are not able to do that, open this page to see the exercise to read the exercises online: [http://gks-017/dm http://gks-017/dm]
<br>
<br>

Open atleast three shell windows
<pre>
ssh -p 24 gks115@gks-144
</pre>

Latest revision as of 16:26, 28 August 2012

Technical specification/requirements

Every participant will have a private node to run the exercises. It is not possible nor needed to run the exercises on a personal laptop because results are not comparable.

The exercises will be provided on a dynamic web page with hints and solutions. The link will be published on Tuesday.


Exercise Overview

The tutorial will start with an introduction to tools used in Linux environments to measure and analyze IO performance. Based on few examples participants will be able to measure individual IO bottlenecks present in hard- and software. The second part will demonstrate optimizations in various parts of the local storage architecture like write-caching, read-ahead, compound reads etc. The third part will be an exercise illustrating the power and potential of Cloud storage where students will apply the technology of distributed hash tables to build a private cloud storage system.

File:IO-Exercises.pdf File:IO-Exercises+Hint.pdf File:IO-Exercises+Hint+Solution.pdf File:DataTechnologies-Exercises-3.pdf File:Cloud.sh

Practical Information

Assigned Exercise Machines

gks-126 Markus Alex
gks-127 Alexander Ditter
gks-128 Bogdan Dumitru
gks-129 Alexander Voigt
gks-130 Alexander Yasnogor
gks-131 Michael Bontenackels
gks-132 Herbert Cornelius
gks-133 Daniel Funke
gks-134 Christian Gumpert
gks-135 Dominik Haitz
gks-136 Silke Halstenberg
gks-137 Christopher Jung
gks-138 Barbara Krasovec
gks-139 Roland Laifer
gks-140 Patrick Martin
gks-141 Maritn Spoo
gks-142 Sang Oh Park
gks-143 Stephan Pfister
gks-145 Angela Poschlad
gks-147 Christian Pulvermacher
gks-037 Doris Ressmann
gks-038 Martin Spoo
gks-039 Alexander Voigts
gks-042 Andreas Wirooks
gks-043 Manuel Zeise

If possible create an SSH tunnel from your laptop to your assigned exercise machine like

ssh -p 24 -L 8000:gks-017:80 gks115@gks-144   # => username@exercisemachine 

Open the follwing address in your browser: http://localhost:8000/dm If you are not able to do that, open this page to see the exercise to read the exercises online: http://gks-017/dm

Open atleast three shell windows

ssh -p 24 gks115@gks-144