Changes between Version 9 and Version 10 of Internal/VirtualPL/ParallelImaging


Ignore:
Timestamp:
Sep 26, 2006, 2:57:07 PM (18 years ago)
Author:
shruti
Comment:

—

Legend:

Unmodified
Added
Removed
Modified

  • Internal/VirtualPL/ParallelImaging

    v9 v10  
     1'''
     2Steps in Imaging a Node:'''
     3
    14The task of parallelization of imaging basically involves running multiple instances of nodeHandler which would launch different images on different nodes. The basic sequence of imaging of nodes as of now is as follows:
    25
    … …  
    4245
    4346The pxe service also provides a mean to set boot image for all the nodes at once with allSetBootImage? which is not working right now. Possibly its just a dummy query.
    44 
    45 For launching the frisbee server:
     47''''''
     48'''For launching the frisbee server:'''
    4649
    4750http://frisbee:5012/frisbee/getAddress?img=<image name to be launched>
    … …  
    6366Thus the revised version of imageNodes completly by passes nodeHandler.
    6467
    65 Design Notes:
     68'''Design Notes:'''
    6669
    67701. One problem being faced in this process is when to terminate the pxe links and delaunch the frisbee server. One simple but dumb solution would be to wait a comfortable period of time (say 5 mins) and break the pxe links. This is a matter of furthur consideration.
    … …  
    7477Kindly refer to imageNodes2 as the most recent copy of imageNodes.
    7578
     79Date: 09/26/06
     80
     81After the imagenodes is run at the console, the frisbee client needs to be launched at the individual nodes. The frisbee client can be launched after the nodes boot into the basic image served by the pxe link. To launch the client, one needs to telnet into the node and then launch the application as follows:
     82
     83./frisbee -p <port number> -m <multicast address> -i <interface ip address> <the memory location for the image>
     84
     85frisbee is located in the /usr/sbin directory.
     86
     87
     88'''Experiments to observe the performance of orbit while nodes are being imaged in parallel.'''
     89
     90In order to evaluate any performance issues, if occuring, while the nodes are being imaged in parallel a basic ping experiment was performed to calculate the RTT from the console to the individual nodes while the nodes were being imaged in parallel. Following are the results:
     91
     92'''Expermient 1''': On sb1 both the nodes were imaged at the same time using two different images, The RTT obtained was:
     93
     94round-trip min/avg/max/ = 0.12/0.2/1.8/ ms
     95
     96while the RTT values obtained while a single node was being imaged were:
     97
     98round-trip min/avg/max = 0.1/0.2/1.2 ms
     99
     100'''Experiment 2''': On grid
     101
     102RTT values when 5 nodes were being imaged with same image:
     103
     104rtt min/avg/max/mdev = 0.142/0.287/6.648/0.691 ms
     105
     106RTT values when 10 nodes were being imaged with 2 different images:
     107
     108rtt min/avg/max/mdev = 0.126/0.288/7.413/0.734 ms
     109
     110RTT values when 15 nodes were being imaged with 3 different images:
     111
     112RTT min/avg/max/mdev = 0.138/0.315/7.215/0.792 ms
     113
     114RTT values once all the images are loaded and nodes have been reset:
     115
     116RTT min/avg/max/mdev = 0.087/0.148/0.221/0.039 ms