Pedro Juan ROİG, Salvador ALCARAZ, Katja GILLY, Carlos JUİZ

Veri Merkezi Topolojilerindeki Anahtarlama Aygıtlarında Enerji Tasarrufuna Yönelik Aritmetik Çalışma

Paralel bilgi işlem şemaları giderek büyüdükçe, veri merkezi tesisleri hızla artıyor. Bu tür tesisler, farklı topolojilere göre tasarlanmıştır, ancak bunların çoğu yeterince kullanılmamaktadır ve bu durum, son kullanıcı bilgisayarları için ağ bağlantısı sağlayan anahtarlama aygıtlarının çoğunda enerjinin kötüye kullanılmasına neden olmaktadır. Bu çalışmada, enerji tasarrufu amacıyla performanslarını öğrenmek için daha tutarlı bir karşılaştırma yapabilmek için her tasarımda aynı sayıda kullanıcı bilgisayarı ile İri Ağaç (Fat Tree), Omurga ve Kanat (Spine and Leaf), N-Hiperküp ve Katlanmış N-Hiperküp gibi bazı yaygın veri merkezi topolojileri incelenmiştir. Her topolojinin kendi avantaj ve dezavantajları olduğu söylenebilir, bu da her birinin bir uygulamanın başlangıç koşullarına bağlı olarak en iyi seçim olabileceği anlamına gelir. Bununla birlikte, sabit gecikme ve seğirme için belirlenen koşullarda, İri Ağaç topolojilerinin veri merkezinin boyutu büyüdükçe daha iyi sonuçlar elde ettiği, diğer tarafta Omurga ve Kanat topolojilerinin bunun aksini yaptığı görülüyor.

Anahtar Kelimeler:

Veri Merkezi, iri ağaç, katlanmış N-Hiperküp, omurga ve kanat

Arithmetic Study about Energy Save in Switches for some Data Centre Topologies

Data Centre facilities are rapidly increasing as parallel computing schemes are ever growing. Those kinds of facilities are designed according to different topologies, but many of them are underutilized, leading to energy misuse in many of the switches providing network interconnection for the end hosts. In this paper, some common Data Centre topologies are studied, such as Fat Tree, Leaf and Spine, N-Hypercube and Folded N-Hypercube, in order to find out their performances in terms of energy saving purposes for the same number of hosts hanging on each design, hence obtaining a coherent comparison among them all. It is to be said that each topology has its own benefits and drawbacks, meaning that each one may be the best option depending on the initial conditions of an implementation. However, in the conditions established for steady latency and jitter, it seems that Fat Tree topologies get better results as the size of the Data Centre grows up, whereas Leaf and Spine does it otherwise.

Keywords:

Data Centre, fat tree, folded N-Hypercube, leaf and spine,

PDF

___

[1] Mao Y, You C, Zhang J, Huang K, Ketaief K B. A Survey on Mobile Edge Computing: The Communication Perspective. IEEE Communications Surveys & Tutorials, 19, 4, 2322-2358, (2017).
[2] Ai Y, Peng M, Zhang K . Edge computing technologies for Internet of Things: a primer. Digital Communications and Networks, 4, 2, 77-86, (2018).
[3] Okwuibe J, Ahmad I, Yianttila M, Liyanage M. Cloud and MEC security. A Comprehensive Guide to 5G Security, 373-397, (2018).
[4] Prabhu C S R . Overview - Fog Computing and Internet-of-Things (IoT). EAI Endorsed Transactions on Cloud Systems, 10, 5, 1-24. China Communications Magazine Co. Ltd, Beijing, (2017).
[5] Habibi P et al . Fog Computing: A Comprehensive Architectural Survey. IEEE Access, 8, 69105-69133, (2020).
[6] Waqas M et al . Mobility-Aware Fog Computing in Dynamic Environments: Understandings and Implementation. IEEE Access, 7, 38867-38879, (2018).
[7] Mukherjee M, Gorlatova M, Gross J, Aazam M. Sustainable Infrastructures, Protocols and Research Challenges for FOG Computing. IEEE Access, 8, 110943-110946, (2020).
[8] S. Vikram . Green Computing.International Conference on Green Computing and Internet of Things, 1, 767-772, (2015).
[9] B. Wadhwa, A. Verma . Carbon efficient VM placement and migration technique for green federated cloud datacenters. International Conference on Advances in Computing, Communications and Informatics,17041923, 2297-2302, (2014).
[10] R. Bala and J. Mann . A Research Paper on Green Computing Using Energy Efficient Task Allocation Strategy in Cloud Environment. International Journals of Advanced Research in Computer Science and Software Engineering, 31907523, 186-191, (2017).
[11] T. Wang, Z. Su, Y. Xia, M. Hamdi . Rethinking the Data Center Networking: Architecture, Network Protocols, and Resource Sharing. IEEE Access, 2, 1481-1496, (2014).
[12] C. Koronen, M. Åhman, L.J. Nilsson . Data centres in future European energy systems—energy efficiency, integration and policy. Energy Efficiency Journal, 13, 129–144, (2020).
[13] Mahmud R, Ramamohanarao K, Buyya R . Latency-Aware Application Module Management for Fog Computing Environments. ACM Transactions on Internet Technology, 19, 1, 9, 1-21, (2019).
[14] Kamath R . Fog Computing – a practical overview. International Journal of Innovative Research in Computer and Communication Engineering, 5, 4, 7819-7822, (2017).
[15] Filiposka S, Mishev A, Juiz C . Community-base VM placement framework.The Journal of Supercomputing, 71, 12, 4504-4528. Springer-Heidelberg, (2015).
[16] Roman R, Lopez J, Mambo M . Mobile edge computing, Fog et al.: A survey and analysis of security threats and challenges. Future Generation Computer Systems, 78, 2, 680-698, (2018).
[17] Sabella D, Vaillant A, Kuure P, Rauschenbach U . Mobile-Edge Computing Architecture: The role of MEC in the Internet of Things. IEEE Consumer Electronics Magazine, 5, 4, 84-91, (2016).
[18] Sethi P, Sarangi S R . Internet of Things: Architectures, Protocols, and Applications. Journal of Electrical and Computing Engineering, 1, 1-25, (2017).
[19] Porambage P. et al . Survey on Multi-Access Edge Computing for Internet of Things Realization. IEEE Communications Survey & Tutorials, 20, 4, 2961-2991, (2018).
[20] Kaur P, Rani A . Virtual machine migration in Cloud computing. International Journal of Grid Distribution Computing, 8, 5, 337-342, (2015).
[21] Roig P J, Alcaraz S, Gilly K, Juiz C . Algebraic modelling of a generic Fog scenario for moving IoT devices. Lecture Notes in Networks and Systems (Springer), in press, (2021).
[22] Osanaiye O, Chen S, Yan Z, Lu R, Choo K R, Dlodlo M. From Cloud to Fog Computing: A Review and a Conceptual Live VM Migration Framework. IEEE Access, 5, 8284-8300 (2017).
[23] Roig P J, Alcaraz S, Gilly K . Arithmetic Study on Energy Saving for some common Data Centre Topologies. Proceedings of the 8th European Conference ECRES 2020, Istanbul (Turkey), 744-751. Published without a DOI, (2020).
[24] Al-Fares M, Loukissas A, Vahdat A . A scalable, commodity data center network architecture.ACM SIGCOMM Computer Communication Review, 38, 4, 63-74 (2008).
[25] Roig P J, Alcaraz S, Gilly K, Juiz C . Modelling VM Migration in a Fog Computing Environment.Journal Elektronika ir Elektrotechnika, 25, 5, 75-81 (2019).
[26] Okafor K C . Leveraging Fog Computing for scalable IoT datacenter using Spine-Leaf network topology. Journal of Electrical and Computer Engineering, 1, 1-11 (2017).
[27] Roig P J, Alcaraz S, Gilly K, Filiposka S, Aknin N . Formal Algebraic Specification of an IoT/Fog Data Centre for Fat Tree or Leaf and Spine Architectures. Proceedings of International Conference on Electrical, Communication and Computer Engineering, 1-6 (2020).
[28] Roig P J, Alcaraz S, Gilly K, Juiz C . Modelling a Leaf and Spine Topology for VM Migration in a Fog Computing Environment. Proceedings of 24th International Conference ELECTRONICS 2020, 1-5, (2020).
[29] Harary F, Hayes J P, Wu J-H . A survey of the theory of hypercube graphs. Computer and Mathematics with Applications, 15, 4, 277-289 (1988).
[30] Roig P J, Alcaraz S, Gilly K, Juiz C . Modelling a Plain N-Hypercube Topology for migration in Fog Computing. Lecture Notes in Electrical Engineering (Springer), in press, (2021).
[31] Zhu Q, Xu J-M, Hou X, Xu M . On reliability of the folded hypercubes. Journal of Information Sciences, 177, 8, 1782-1788 (2007).
[32] Roig P J, Alcaraz S, Gilly K, Juiz C . Modelling a Folded N-Hypercube Topology for migration in Fog Computing. Lecture Notes in Electrical Engineering (Springer), in press, (2021).
[33] Agarwal, U., Jain, N. Distributed Energy Resources and Supportive Methodologies for their Optimal Planning under Modern Distribution Network: a Review. Technology and Economics of Smart Grids and Sustainable Energy, 4, 3, 1-21 (2019).