The Experts below are selected from a list of 6 Experts worldwide ranked by ideXlab platform

Olivier Mehani - One of the best experts on this subject based on the ideXlab platform.

  • Disaster-Tolerant Storage with SDN?
    2016
    Co-Authors: Vincent Gramoli, Guillaume Jourjon, Olivier Mehani
    Abstract:

    Abstract. Cloud services are becoming centralized at several geo-replicated datacentres. These services replicate data within a single dat-acentre to tolerate isolated failures. Unfortunately, the effects of a dis-aster cannot be avoided, as existing approaches migrate a copy of data to backup datacentres only after data have been stored at a primary datacentre. Upon disaster, all data not yet migrated can be lost. In this paper, we propose and implement SDN-KVS, a disaster-tolerant key-value store, which provides strong disaster resilience by replicating data before storing. To this end, SDN-KVS features a novel commu-nication primitive, SDN-cast, that leverages Software Defined Network (SDN) in two ways: it offers an SDN-multicast primitive to replicate critical update Request flows and an SDN-anycast primitive to Redirect Request flows to the closest available datacentre. Our performance evalu-ation indicates that SDN-KVS ensures no data loss and that traffic gets Redirected across long distance key-value store replicas within 30 seconds after a datacentre outage.

Vincent Gramoli - One of the best experts on this subject based on the ideXlab platform.

  • Disaster-Tolerant Storage with SDN?
    2016
    Co-Authors: Vincent Gramoli, Guillaume Jourjon, Olivier Mehani
    Abstract:

    Abstract. Cloud services are becoming centralized at several geo-replicated datacentres. These services replicate data within a single dat-acentre to tolerate isolated failures. Unfortunately, the effects of a dis-aster cannot be avoided, as existing approaches migrate a copy of data to backup datacentres only after data have been stored at a primary datacentre. Upon disaster, all data not yet migrated can be lost. In this paper, we propose and implement SDN-KVS, a disaster-tolerant key-value store, which provides strong disaster resilience by replicating data before storing. To this end, SDN-KVS features a novel commu-nication primitive, SDN-cast, that leverages Software Defined Network (SDN) in two ways: it offers an SDN-multicast primitive to replicate critical update Request flows and an SDN-anycast primitive to Redirect Request flows to the closest available datacentre. Our performance evalu-ation indicates that SDN-KVS ensures no data loss and that traffic gets Redirected across long distance key-value store replicas within 30 seconds after a datacentre outage.

Guillaume Jourjon - One of the best experts on this subject based on the ideXlab platform.

  • Disaster-Tolerant Storage with SDN?
    2016
    Co-Authors: Vincent Gramoli, Guillaume Jourjon, Olivier Mehani
    Abstract:

    Abstract. Cloud services are becoming centralized at several geo-replicated datacentres. These services replicate data within a single dat-acentre to tolerate isolated failures. Unfortunately, the effects of a dis-aster cannot be avoided, as existing approaches migrate a copy of data to backup datacentres only after data have been stored at a primary datacentre. Upon disaster, all data not yet migrated can be lost. In this paper, we propose and implement SDN-KVS, a disaster-tolerant key-value store, which provides strong disaster resilience by replicating data before storing. To this end, SDN-KVS features a novel commu-nication primitive, SDN-cast, that leverages Software Defined Network (SDN) in two ways: it offers an SDN-multicast primitive to replicate critical update Request flows and an SDN-anycast primitive to Redirect Request flows to the closest available datacentre. Our performance evalu-ation indicates that SDN-KVS ensures no data loss and that traffic gets Redirected across long distance key-value store replicas within 30 seconds after a datacentre outage.

Halimi Bahrul - One of the best experts on this subject based on the ideXlab platform.

  • Rancang Bangun Sistem Penyeimbang Beban Pada Klaster Server Dengan Prioritas Berbasis Konten Dan Kontrol Ketersediaan Layanan
    2015
    Co-Authors: Halimi Bahrul
    Abstract:

    Aplikasi berbasis web semakin diminati untuk berbagai proses bisnis yang ada di lingkungan kita. Salah satu yang menjadi sorotan adalah penerimaan peserta didik baru yang dilaksanakan secara online. Dengan adanya aplikasi ini pengguna akan dihadapkan pada dua jenis halaman yaitu halaman pengisian informasi untuk daftar dan halaman untuk menampilkan informasi. Tentu dengan pengaturan yang biasa, server akan melayani dua jenis halaman ini secara bersamaan. Kondisi ini akan mengakibatkan bottle neck atau penumpukan permintaan. Muncul gagasan untuk membagi beban kerja ke komputer lain agar setiap permintaan yang masuk dapat dilayani. Gagasan ini sudah biasa dilakukan dengan menggunakan Nginx sebagai balancer. Namun dengan beragamnya tipe permintaan pengguna, waktu yang dibutuhkan untuk melayani menjadi tidak stabil. Gagasan lain muncul dengan adanya pembagian beban kerja berdasarkan konten yang diakses pengguna. Konten ini dapat diartikan sebagai dua jenis halaman sebelumnya. NodeJS akan bekerja sebagai balancer dan membaca setiap permintaan pengguna dan mengarahkan permintaan kepada worker yang sesuai untuk setiap URL yang diakses pengguna. Dibantu dengan MongoDB sebagai basis data, NodeJS akan bekerja lebih konsisten terhadap data masuk ke sistem. Hasil menunjukkan dengan NodeJS dan prioritas berbasis konten, waktu respon meningkat seiring dengan bertambahnya permintaan yang masuk. Namun dengan terfokusnya kerja worker untuk melayani satu jenis halaman, membuat setiap permintaan dapat terlayani hingga permintaan selesai. =============================================================================================== Web-based application become viral in our living society nowadays. One of those application is an on line application for student enrollment. In this kind of application user will access two type of page, the first one is a page with information from database or not and the second one is a page with an insert action to database. With a basic installation of server, server will serve this two type of page and make a bottle neck condition for high access. There is an idea to load balance every Request to another computer, so every Request can be handled. Usually, system administrator will use Nginx as a load balancer. But with multiple kind of Request, response time to make every Request handled, become unstable. Another idea appear, every Request will be load balance with content based priority. Content is represented as two type of page before. Node JS will be used as load balancer system and read every header of Request and Redirect Request to specific worker. With MongoDB as a database of system, NodeJS will work consistently for every data that come to the system. The result show that NodeJS and content-based priority will increase response time for increasing number of access from user. But the mechanism to make a worker focus on one type of page, make every Request will be server well