Grid'5000 user report for Matthieu Imbert

Jump to: navigation, search

User information

Matthieu Imbert (users, user, account-manager, network-staff, ct, lyon, lyon-staff, ml-users, digitalis user)
More user information in the user management interface.


  • A metrology platform for grid environments (Networking) [in progress]
    Description: We want to develop a metrology platform for grid environment.
    This platform will acquire information of different type, from multiple sources (with different locations) and with differents means, such as (but not limited to) :
    - packet capture
    - netflow
    - sflow
    Then it will computes some results to :
    - characterize network traffic in grid environment
    - study the behavior of grid flows
    This second objective will be usefull for users to see the impact of other flows on their experiment ans/or to debug their experiment.

    The first step is to develop a prototype which performs packet capture. And to test this prototype we need to generate traffic between sites on grid5000.
  • Metrology on Grid5000 (Networking) [in progress]
    Description: We want to analyse network traffic on the grid in terms of statistical properties like Long Range Dependance, self-similarity, etc. We especially want to understand the impact of statistic characteristics of the traffic on the QoS. The experiments consist in creating traffic with some imposed characteristics (flow size distribution, etc.) and looking at the resulting arrival and bandwidth processes and the QoS.
    Results: not yet
  • Investigating self-similarity and heavy tailed distributions on a large scale experimental facility (Networking) [achieved]
    Description: After seminal work by Taqqu et al. relating self-similarity to heavy tail distributions, a number of research articles verified that aggregated Internet traffic time series show self-similarity and that Internet attributes, like WEB file sizes and flow lengths, were heavy tailed. However, the validation of the theoretical prediction relating self-similarity and heavy tails remains unsatisfactorily addressed, being investigated either using numerical or network simulations, or from uncontrolled web traffic data. Notably, this prediction has never been conclusively verified on real networks using controlled and stationary scenarii, prescribing specific heavy-tail distributions, and estimating confidence intervals. In the present work, we use the potential and facilities offered by the large-scale, deeply reconfigurable and fully controllable experimental Grid5000 instrument, to investigate the prediction observability on real networks. To this end we organize a large number of controlled traffic circulation sessions on a nation-wide real network involving two hundred independent hosts. We use a FPGA-based measurement system, to collect the corresponding traffic at packet level. We then estimate both the self-similarity exponent of the aggregated time series and the heavy-tail index of flow size distributions, independently. Comparison of these two estimated parameters, enables us to discuss the practical applicability conditions of the theoretical prediction.
  • TCP traffic self-similarity under loss (Networking) [in progress]
    Description: Over the last decade, many research efforts have been devoted to the study of aggregated traffic time series collected at the core of networks. The pioneering works by Paxson and Leland showed that the Poisson hypothesis, which is relevantly used in phone networks, was not suitable to describe computer networks. Instead, self-similarity was proved a much more appropriate paradigm. Then, the theoretical work from Taqqu and collaborators identified the heavy-tailed nature of the file size distribution as a possible origin for the observed self-similarity. In addition, it gave the exact relation between the self-similarity index and the tail index that should be observed when the sources behavior is modeled with the ON/OFF model. Despite a controversial debate on the question, it has then been more recently stated that the TCP congestion control mechanism cannot be responsible for the self-similarity observed in the large time scales. On the opposite side, we show in this work that when the file size is heavy-tailed, the TCP congestion control mechanism under sufficiently high loss can annihilate the self-similarity that would be observed without any loss. For this work, we use large scale controled experiments performed on Grid5000. Independant TCP sources send files in an ON/OFF scenario with a heavy-tailed ON periods; and a constant loss rate is created via UDP cross traffic.
  • Pilgrim metrology framework (Middleware) [in progress]
    Description: Instrumentation of Grid5000 with the Pilgrim metrology framework.

    We aim to provide:
    - several kinds of metrics at the system / network / application levels.
    - a platform model that can be queried
    - an aggregation of several topology / metrics data sources
    - prediction tools (eg. given a list of data transfers, be able to accurately predict their durations)

    Pilgrim is an open application/system/network metrology and topology framework for any kind of computer platform, especially cluster/grid/cloud platforms. It offers an object oriented platform model and a full client API, because it is designed to be usable not only by humans but also programmatically.
    More information here

  • Engineering of high speed network capture solutions (Networking) [in progress]
    Description: We integrate and evaluate several tools for high speed network trafic capture:
    - AIST GtrcNET-1 (4x 1 Gbit/s ports, FPGA. traffic capture, traffic shaping, latency emulation, etc.).
    - AIST GtrcNET-10 (3x 10 Gbit/s ports, FPGA, traffic capture, traffic shaping, latency emulation, etc.).
    - Myri10GE Ethernet cards with sniffer 10G.
    illustrating chart picture not found
    More information here
  • Development of tools for conducting experiments on distributed platforms (Middleware) [in progress]
    Description: Development of various tools for conducting experiments on distributed platforms, especially the execo tool
    Scripting experiments allows them to be reproducible, and also acts as a kind of self-documentation.
    More information here


  • \emphMetroflux: a high performance system for very fine-grain flow analysis [2009] (national)
    EntryType: inproceedings
    Resotos: web
    Author: Loiseau, Patrick and Gonçalves, Paulo and Guillier, Romaric and Imbert, Matthieu and Goga, Oana and Kodama, Yuetsu and Vicat-Blanc Primet, Pascale
    Booktitle: Grid'5000 Spring School
    Address: Nancy, France
    Month: April
    X-editorial-board: yes
    X-proceedings: no
    X-international-audience: no
  • Automated Traffic measurements and analysis in Grid5000 [2009] (international)
    EntryType: misc
    Resotos: web
    Author: Loiseau, Patrick and Guillier, Romaric and Goga, Oana and Imbert, Matthieu and Gonçalves, Paulo and Vicat-Blanc Primet, Pascale
    Note: ACM Sigmetrics/Performance demonstration contest (\textbf{Best Student Demonstration Award})
    Month: June
    X-editorial-board: yes
    X-proceedings: no
    X-international-audience: yes
  • Metroflux: A high performance system for analyzing flow at very fine-grain [2009] (international)
    EntryType: inproceedings
    Resotos: web
    Author: Loiseau, Patrick and Gonçalves, Paulo and Guillier, Romaric and Imbert, Matthieu and Kodama, Yuetsu and Vicat-Blanc Primet, Pascale
    Booktitle: TridentCom
    Address: Washington DC, USA
    X-editorial-board: yes
    X-proceedings: yes
    X-international-audience: yes
    Abstract: Researches in network traffic analysis embrace a large diversity of goals and are based on a variety of methodologies and tools. To have a better insight on the real nature and on the evolution of network traffic we argue that fine-grain analysis of real traffic traces have to complement simulations studies as well as coarse grain measurement performed by classical flow measurement systems. In particular, packet level measurements and analysis are needed. However, such methodologies are resource consuming and require very high performance devices to be operational in real high speed networks. In this paper we present the \emph{Metroflux} system which aims at providing researchers and network operators with a very flexible and accurate packet-level traffic analysis toolkit configured for 1~Gbps and 10~Gbps speed links. This system is based on the GtrcNet FPGA-based device technology and on specific statistical analysis tools. We show the potential and the facilities offered by the \emph{Metroflux} system coupled with the \emph{Grid5000} large scale experimental platform and the Network eXperiment Engine (\emph{NXE}) we have developed. We illustrate the application of the \emph{Metroflux} system with the practical validation of the theoretical prediction relating self-similarity and heavy tails given by Taqqu theorem. We also illustrate several usages of this toolset, such as the investigation of conditions under which several traffic theories apply, as well as studies on traffic, protocols and systems interactions.


    Success stories and benefits from Grid'5000

    last update: 2010-08-06 11:37:16

    Personal tools

    Public Portal
    Users Portal
    Admin portal
    Wiki special pages