Changes between Version 7 and Version 8 of GENIFireCollaborationWorkshopSeptember2015/Session6


Ignore:
Timestamp:
10/06/15 23:51:10 (7 years ago)
Author:
griff@netlab.uky.edu
Comment:

--

Legend:

Unmodified
Added
Removed
Modified
  • GENIFireCollaborationWorkshopSeptember2015/Session6

    v7 v8  
    1616
    1717
     18The first session (which occurred last) began by looking at the problem of
     19monitoring power usage information for testbeds networks and clouds.  The
     20session began with a description of the Cloudlab power monitoring system
     21which uses the Intelligent Platform Management Interface (IPMI) to collect
     22power measurements over time and make them available to experimenters via a
     23web page.  A related talk described the power monitoring service supported by
     24the Kwapi system which was designed to offer high frequency data collection,
     25allowing users to correlate and associate power measurements with other
     26network behaviors (e.g., bandwidth usage).
     27The federation of testbeds presents additional monitoring challenges
     28ranging from the complexities of dealing with heterogeneous infrastructure,
     29software systems, and data formats, to scalable data collection and
     30aggregation, to trust/reliability issues.  Example systems such as
     31MODACLOUDS, SPECS, and DICE were described that use monitoring architectures
     32consisting of many distinct components, each with a specific purpose, to
     33address the complexity of these systems.  The FELIX monitoring service
     34handles the scalability challenge by arranging measurement services
     35hierarchically at the highest level, but then using a distributed approach
     36within testbed islands. The Federated Trust and User Experience framework
     37address the problem of assessing a testbed's reliability and trustworthiness
     38by combining conventional monitoring information (objective information) with
     39reputation-based trust  management systems (subjective information) to
     40define a reputation value that can be used to compare testbeds.
     41While much work has focused on monitoring testbed infrastructure,
     42relatively few tools are available to users to measure their
     43specific experiment or slice.  Looking at experimenter mailing lists and
     44discussion groups shows that the majority of problems that experimenters
     45face would not be detected by monitoring the infrastructure, but rather
     46require slice/experiment-specific measurement and validation tools designed to
     47test and evaluate the performance of a user's slice or experiment.
     48
    1849
    1950
    2051== Presentations ==
    2152
    22 === Monitoring Wireless Infrastructure ===
     53=== Session 2: Monitoring Wireless Infrastructure ===
    2354Discussion leader: Ivan Seskar
    2455
     
    2758|| James Miller || Measuring Broadband to the Sidewalk || [attachment:"MBA Fixed and Mobile Speed Test App - GENI.pptx" slides] ||
    2859
    29 === Monitoring the Internet as a Whole ===
     60=== Session 3: Monitoring the Internet as a Whole ===
    3061Discussion leader: Scott Kirkpatrick
    3162
     
    3768
    3869
    39 === Monitoring Testbed Networks, Clouds, Clusters, and Slices ===
     70=== Session 1: Monitoring Testbed Networks, Clouds, Clusters, and Slices ===
    4071Discussion leader: Jim Griffioen
    4172