Dynamic Resource Allocation in Cloud Environments

Abstract

The provisioning and use of computing resources has been completely transformed by cloud computing, which provides previously unheard-of levels of scalability, flexibility, and cost-effectiveness. In order to maximize resource utilizatio and guarantee peak performance in cloud settings, dynamic resource allocation is essential. In order to provide a thorough grasp of the most recent developments in this subject, this survey paper offers an in-depth review of dynamic resource allocation methods in cloud computing. The survey includes a wide range of dynamic resource allocation techniques, including as cost-optimization methods, resource provisioning and scheduling techniques, and load-balancing algorithms. Each method is examined in terms of its advantages, drawbacks, and applicability to particular cloud deployment scenarios. The research also investigates evaluation approaches and performance indicators for evaluating the efficacy of dynamic resource allocation systems. It emphasizes the need for continued research and innovation to address the evolving demands of cloud computing environments.

Introduction

I. INTRODUCTION

Cloud  computing  has revolutionized modern computing by providing  flexible  and scalable access to shared computing resources. Efficient resource allocation is crucial for optimal performance  and  cost-effectiveness  in  cloud environments. In  this  survey  paper,  we  aim to explore the state-of-the-art developments and drawbacks of dynamic    resource allocation (DRA) techniques in cloud computing[10]. Dynamic resource allocation involves the adaptive management    of    computing   resources   in   response   to changing workloads and user requirements. By intelligently distributing  resources,  DRA  techniques  aim  to  maximize resource   utilization,   enhance   system   performance, and improve user experience. Through  an  extensive review of literature and research, we seek to provide  a  comprehensive  understanding  of  the current  state-of-the-art  in  dynamic  resource  allocation[3]. We will examine various techniques, algorithms, and strategies employed for efficient resource allocation in cloud computing. Furthermore, we will identify the drawbacks and limitations of existing approaches[16], highlighting areas for improvement    and    future research. By   analyzing   the literature on dynamic resource allocation, this survey paper aims      to      inform      researchers,      practitioners, and decision-makers  about  the  current  landscape  and  inspire further  advancements  in  resource  allocation  techniques for enhanced   performance   and   cost-effectiveness   in   cloud environments.

II. BACKGROUND AND CONCEPTS

A. Basics of Cloud Computing

A quick, on-demand   access   to   a   shared   pool   of computing  resources  over  the  Internet  is made possible by the  revolutionary computing  architecture  known  as  cloud computing.   It   provides   scalability   to   organizations   by enabling them to quickly scale resources up or down to meet changing  demands.  The  flexibility  of  cloud  computing's deployment  options,  including  public,  private,  and  hybrid clouds,  allows companies to select the best setting for their applications.  With  this  model,  organizations  may  only pay for the resources they use, negating the need for significant up-front    expenditures   on   hardware   and   infrastructure upkeep. Businesses may achieve agility, scalability, and cost optimisation  in  their  computing  operations thanks to cloud computing.

B. Dynamic Resource Allocation

Dynamic  resource  allocation  is  a  critical  component  of cloud    environments    that    aims    to    optimize    resource utilization    and enhance system performance[1][6]. It involves  the  intelligent  distribution  and  management  of computing   resources   to   meet   the   varying   demands  of applications and users[24].

In cloud  computing,  where  resources  are  shared  among multiple users and applications, efficient resource allocation becomes  essential. Dynamic resource allocation techniques enable the allocation and reallocation of resources based on workload    variations, user requirements, and    system conditions.  By  dynamically  provisioning,  reallocating,  and deallocating  resources, cloud providers can ensure efficient utilization, scalability, and responsiveness[8].

To  better  understand  dynamic  resource  allocation  in  cloud environments,  it  is important to have a solid understanding of  the  prerequisite  concepts.  These  include  load  balancing algorithms[9],  which  distribute  workloads  across  available resources   to   optimize   resource   utilization;   auto-scaling mechanisms  that  automatically  adjust  resource  capacities based    on    workload    patterns;    and    provisioning    and scheduling  strategies  that  match  workload  demands  with suitable resource allocations[38].

C. Challenges

Challenges to dynamic   resource   allocation   in   cloud environments    are    multifaceted    and    demand    careful consideration. First and foremost, accurately predicting and effectively  handling  workload  variations  in  real-time  pose significant  challenges  due to the dynamic  nature  of  user demands. Balancing   resource   allocation   to   avoid   both overutilization  and  underutilization  is critical  to  achieve optimal performance    and    cost-effectiveness.    Ensuring fairness and preventing resource contention among multiple users  and  applications  is  also  a  crucial  challenge,  as  it directly   impacts   user   satisfaction and   system   stability. Efficient monitoring and     management    of    resource utilization,     scalability,     and     availability are further complexities that need to be addressed[20].

III. DESIGN

A. Technologies

Communication-aware and memory load balance(“CMLB”)  mapping  algorithm  technique[12]  is  an optimized  load  stabilizing  method  for  the  cloud  based audio-video  ecosystem..  It  wholly  accounts  the load of all hosts  or  machines  allocated under that ecosystem and the conditions of the network and thereby achieves economical and   efficient   scheduling   and   load   management.   Even though   this   method   is   considered   to   be   cheaper  and reasonable, it is not able to utilize and maximize the output which the system is getting in a very efficient manner. Load balancing  ant-colony  optimization  (“LBACO”)[13]  is  a technique  which  again  handles  the  load  of  the  complete ecosystem  while  reducing  the  make  span  of  the  list of all the   tasks.   But,   the tasks cannot   shelter   the   external processing.   In   [14]   the  paper  describes  how  the  load balancing  process  makes  use  of  the  concept  of  a  cluster switch,  cluster-based load balancing method [14] performs well  in  heterogeneous  nodes  by  considering  the  resource demands  of  each  task  and  reducing  the  overheads  by clustering  the  machines.  But  there  is  a  trade-  off between the  requirements  of  SLA  and  the  consumption  of  energy. Dynamic   allocation   of   resources   with   imbalance   [15] efficiently   allocates   the   resources   and   improves   the allocation and efficiency of total resources. Ant colony and network    theory    [16]    techniques    balanced    the    load efficiently  in  the  distributed  architecture  like  system  and considered the complex type of  networks. This method was utilized in open cloud computing.

B. Design Issues

In  the  modern  digital  world  it  is  hard  to  exist  without services  existing  on  cloud  based  infrastructure  to  achieve proper  Quality  of  Service  to  the  clients. Cloud computing provides  several  services  to  its  subscribers  or clients. The cloud  customer  could request a number of requirements to the  cloud  provider  which  can  account  for  a  number  of resources.   Then   this   cloud  provider  will  schedule  the resources to the server according to the client requirements and  feasibility  of  such  a  request  from  the  client.  Several design related concerns are arising network scalability [18], dynamic allocation of resources in a cloud infrastructure[19], cost structure and profit making [20],  cost  for  each  service  [21],  resource  scheduling [22], catering of resources [23], load stabilizing[24] etc.

To meet the  need  of  the  user  software  as  a  service  (SaaS)  lease resources  from  the infrastructure as a service (IaaS) which can impact the quality of the service because of its variable performance [25]. Load balancing is used to ensure that all the computing resources are allocated effectively and fairly, which make certain that no single node is overloaded. It has become a challenge for the provider of cloud when there is a  large  number  of  cloud  users  [26].  The  cloud  provider provides resources to the user in the form of pay per use.

To allocate the resources effectively and properly and to satisfy the expectation of the user, several algorithms are used for resource allocation [27].

IV. LITERATURE REVIEW

The papers surveyed related to dynamic resource allocation with such systems where some   of   them   are  dynamic     resource  allocation  [17], in cloud are mentioned below:

A. Dynamic Resource Allocation

Numerous   studies   have   explored   different   aspects   of dynamic   resource   allocation   techniques   to   address   the challenges and complexities of cloud environments.

The    prominent    focus    lay    on    algorithms    like    ant colony[22][25],  round-robin, game theory based and others with  the  aim  of  utilizing  resources  to  the  maximum  with minimum wastage and optimized costs.

By  S.  Wang  et  al.  [1], one of the significant developments was  described.  A  probabilistic  stacking  ensemble  learning approach  that  incorporates  Gaussian  process  regression, long  short-term  memory  networks,  linear  regression,  and random  forest  is proposed to anticipate the cloud resources required   by   CSS   applications   that   expand  to  complex system  simulation.  However,  as  the  article  only  examined one  ensemble  learning  algorithm,  it's  possible  that  other ensemble techniques would be more effective.

Moving  forward,  the  research  by  A.  Mampage  et  al.  [2] incorporated       Deadline-aware       Dynamic       Resource Management   in   order   to  efficiently  maximize  resource utilization   for   the   provider   and   dynamically   manage resources  to speed up function response times. Authors use the Container CloudSim toolkit to simulate their method and assess  it.  This  approach  may  require  frequent  re-tuning  to achieve optimal performance.

The research by A.Khanna and Sarishma [4] introduced the Resource   Allocation   System(RAS),   Resource   allocation system   implemented   in   CloudSim      cloud  environment simulator.  The  RAS  works  on  the  principles  of  Dynamic resource  allocation  whereas  J.Ghaderi  [5]  in  his  research employs   Markov   Chain   and  Knapsack  Problem  in  his approach.

B. Scheduling

Mahummad Ibrahim et al. [36]   investigate and analyze the working principles of eight recent state-of-the-art static task scheduling algorithms in the context of cloud environments. The   selected   approaches   were   chosen   for   comparative analysis  based  on  their  widespread  usage  in  performance comparisons, their use as basic approaches extended by new proposals,   and   their   consistent   behavior   on   multiple datasets.   The   paper   aims   to   provide   insights  into  the strengths and limitations of these heuristic-based algorithms.

Guojun  Sheng  et  al.[37]  propose  an  improved  scheduling algorithm  for  cloud  workflow  called  ISACW  (Improved Scheduling Algorithm for Cloud Workflow). The algorithm utilizes a Directed Acyclic Graph (DAG) model to represent cloud  workflows  and  employs  critical  path  retrieval  and depth-first-search methods to dynamically adjust the service time  of  workflow  tasks.  The  goal  is  to  reduce  the  overall execution time and cost of cloud workflow tasks.  The paper presents  the  architecture  of  cloud  workflow  scheduling, which  involves  multiple Cloud Workflow Engines (CWEs) interconnected  to  form  a  distributed  workflow  instance process  scheduling  network.  Each  CWE  has  a  Process Request    Scheduler    (PRS)    component   responsible   for receiving  and  scheduling  user  workflow  process instances. The Cloud Service Registration Center (CSRC) provides the list of candidate services for each workflow activity.

The  research  by  K.  M.  S.  U  Bandaranayake  et  al.[37] presents  a  new  algorithm  called  Total  Resource  Execution Time  Aware  (TRETA)  for  workflow  scheduling  in  cloud computing environments. The authors compare the proposed algorithm  with  existing  heuristics  such as First Come First Serve    (FCFS),    Maximum    Completion    Time    (MCT), Maximum  Execution  Time  (MET), MaxMin, MinMin, and Distributed  Heterogeneous  Earliest  Finish  Time  (DHEFT). The  algorithm starts by calculating the total execution time of each resource based on the tasks already scheduled. Then, it  selects  a  task  from  the unscheduled list and assigns it to the  resource  that  gives  the  minimum  total  execution  time with the new task. The algorithm updates the total execution time  for  the  selected  resource,  removes  the scheduled task from  the  unscheduled  list,  and  repeats the process until all tasks are scheduled.

R.Buyya   et  al.[23]  in  their  research  paper  discuss  the proposed  MPSO  (Modified  Particle  Swarm  Optimization), MCSO  (Modified Cat Swarm Optimization), and HYBRID (MPSO+MCSO)    techniques    for    task    scheduling    and resource    allocation    in    a    cloud    environment.    These techniques  aim  to  improve the efficiency and utilization of virtual  machines  (VMs)  and  cloud  resources.  The  MCSO algorithm  focuses  on  the  seeking  mode  of  the  Cat  Swarm Optimization   (CSO)   technique   for   resource   allocation management. 

It utilizes four memory pools:   Seeking Memory    Pool    (SMP),    Seeking    Range    of    Selected Dimension  (SRD),  Counts  to  Dimension  Change  (CDC), and Self Position Consideration (SPC). These memory pools play  a  role  in  assigning  cloud  resources  to  VMs.  The proposed   work   aims   to   schedule   tasks   on   VMs   and efficiently  allocate  CPU  and  memory  resources  using  the MPSO and MCSO algorithms. The algorithms consider the resource  demands  of  tasks  and  assign  resources  from  the VMs or the cloud resource pool based on specific conditions and  comparisons.  The  research  also  mentions  the  use  of a Branch-and-Bound  based  Exact  algorithm  as  a  benchmark for  performance  comparison  with the proposed algorithms. The  Exact  algorithm  provides  the global optimum solution for efficient resource utilization.

C. Load Balancing

Load balancing in cloud computing refers to the distribution of   incoming   network  traffic  across  multiple  servers  to optimize  resource  utilization,  improve  performance,  and ensure high availability of applications and services.

In  [25]  authors  Z  Zhang  and  X  Zhang  have  leveraged the Ant Colony Optimization technique for their load balancing in   Cloud   computing.  Their  model  of  cloud  computing supported heuristic systems by minimizing the make span.

Manickachezian,  R.  &  Eswaran  in  [35]  have  made  use  of Support  vector  machine(SVM)  and  dynamic  multiservice load  balancing  with  adaptive  genetic  algorithms(AGA)  It allows the provider to host multi service with minimum cost required for the resources.

Moving  forward  in  the  world of load balancing, authors in [26]   used   stochastic   models   for   load   balancing.  They incorporated  non-preemptive  policies  for  virtual  machines. They concluded that choosing long frame durations resulted in throughput optimal performance.

H.Wen  et  al.[21]  in  their  paper  aim  to  distribute  client requests  evenly  across  the  available  nodes.  It  operates  by measuring   network   latency,   calculating   costs,   selecting optimal  nodes,  and  maintaining  a  list  of optimal nodes for each  client.  The  algorithm  then  assigns  clients  to  their respective optimal nodes, giving priority to the first item in the  optimal  node  list  if  multiple options are available. The performance   of   the   proposed   algorithm   was  compared against three traditional algorithms: round-robin, number of connections-based,  and C/S distance-based algorithms. The evaluation considered two important metrics: response time (the time from user request to receiving a response) and task throughput  (the  number  of  completed  tasks  per  elapsed time)

Conclusion

In conclusion, dynamic resource allocation plays a crucial role in optimizing the efficiency and performance of cloud computing environments. By dynamically allocating resources such as computing power, storage, and network bandwidth based on real-time demands and workload patterns, cloud providers can effectively balance the utilization of their infrastructure while meeting service-level agreements (SLAs) and delivering a seamless user experience. Dynamic resource allocation enables the cloud to adapt to fluctuating workloads and varying resource requirements, ensuring optimal resource utilization and minimizing wastage. It allows for scaling resources up or down in response to changing demands, which leads to improved cost-effectiveness and scalability for both cloud providers and users. Furthermore, dynamic resource allocation techniques leverage intelligent algorithms and predictive analytics to optimize resource provisioning, workload distribution, and load balancing. By analyzing historical data and considering factors like workload characteristics, user preferences, and performance objectives, cloud providers can make informed decisions in real-time to allocate resources efficiently and optimize overall system performance. Through dynamic resource allocation, cloud computing environments can enhance their resilience, fault tolerance, and overall system reliability. The ability to detect and respond to failures or performance degradation promptly ensures that critical services remain available and minimizes downtime. However, it is essential to carefully design and implement dynamic resource allocation mechanisms to address challenges such as resource contention, network latency, and the complexity of distributed systems. Additionally, privacy and security considerations should be taken into account to protect sensitive data and ensure compliance with regulatory requirements. Overall, dynamic resource allocation empowers cloud computing to efficiently scale and adapt to changing demands, enabling businesses and users to leverage the benefits of the cloud while optimizing resource usage, enhancing performance, and driving cost savings. As cloud computing continues to evolve, dynamic resource allocation will remain a critical component for unlocking its full potential.

References

[1] S. Wang, Y. Yao, Y. Xiao and H. Chen, \"Dynamic Resource Prediction in Cloud Computing for Complex System Simulatiuon: A Probabilistic Approach Using Stacking Ensemble Learning,\" 2020 International Conference on Intelligent Computing and Human-Computer Interaction (ICHCI), Sanya, China, 2020, pp. 198-201, doi: 10.1109/ICHCI51889.2020.00050. [2] A. Mampage, S. Karunasekera and R. Buyya, \"Deadline-aware Dynamic Resource Management in Serverless Computing Environments,\" 2021 IEEE/ACM 21st International Symposium on Cluster, Cloud and Internet Computing (CCGrid), Melbourne, Australia, 2021, pp.483-492, doi: 10.1109/CCGrid51090.2021.00058. [3] Madni SH, Latiff MS, Coulibaly Y,“Recent advancements in resource allocation techniques for cloud computing environment: a systematic review”, Cluster Computing, vol.20, no.3, pp.2489-533, 2017. [4] A. Khanna and Sarishma, \"RAS: A novel approach for dynamic resource allocation,\" 2015 1st International Conference on Next Generation Computing Technologies (NGCT), Dehradun, India, 2015, pp. 25-29, doi:10.1109/NGCT.2015.7375076. [5] J. Ghaderi, \"Simple high-performance algorithms for scheduling jobs in the cloud,\" 2015 53rd Annual Allerton Conference on Communication, Control, and Computing (Allerton), Monticello, IL, USA, 2015, pp. 345-352, doi:10.1109/ALLERTON.2015.7447025. [6] K. Vasudewa and P. Gupta, \"A survey on elastic resource allocation algorithm for cloud infrastructure,\" 2016 International Conference on Innovation and Challenges in Cyber Security (ICICCS-INBUSH), Greater Noida, India, 2016, pp. 199-203, doi: 10.1109/ICICCS.2016.7542344. [7] E. I. Nehru, J. I. S. Shyni and R. Balakrishnan, \"Auction based dynamic resource allocation in cloud,\" 2016 International Conference on Circuit, Power and Computing Technologies (ICCPCT), Nagercoil, India, 2016, pp. 1-4, doi: 10.1109/ICCPCT.2016.7530347. [8] R. Madhumathi, R. Radhakrishnan and A. S. Balagopalan, \"Dynamic resource allocation in cloud using bin-packing technique,\" 2015 International Conference on Advanced Computing and Communication Systems, Coimbatore, India, 2015, pp. 1-4, doi: 10.1109/ICACCS.2015.7324065. [9] A. Jyoti, M. Shrimali and R. Mishra, \"Cloud Computing and Load Balancing in Cloud Computing -Survey,\" 2019 9th International Conference on Cloud Computing, Data Science & Engineering (Confluence), Noida, India, 2019, pp. 51-55, doi: 10.1109/CONFLUENCE.2019.8776948. [10] Aditya Narayan Singh and S. Prakash, \"Challenges and opportunities of resource allocation in cloud computing: A survey,\" 2015 2nd International Conference on Computing for Sustainable Global Development (INDIACom), New Delhi, India, 2015, pp. 2047-2051 [11] A. Alahmadi, A. Alnowiser, M. M. Zhu, D. Che and P. Ghodous, \"Enhanced First-Fit Decreasing Algorithm for Energy-Aware Job Scheduling in Cloud,\" 2014 International Conference on Computational Science and Computational Intelligence, Las Vegas, NV, USA, 2014, pp. 69-74, doi: 10.1109/CSCI.2014.97. [12] Xu M, Tian W, Buyya R,“A survey on load balancing algorithms for virtual machines placement in cloud computing, ” Concurrency and Computation: Practice and Experience. vol.29, no.12, 2017. [13] Muthulakshmi B, Somasundaram K,“A hybrid ABC-SA based optimized scheduling and resource allocation for cloud environment”. Cluster Computing.pp. 1-9, 2017. [14] Upadhyaya J, Ahuja NJ, “Quality of service in cloud computing in higher education: A critical survey and innovative model”, InI-SMAC (IoT in Social, Mobile, Analytics and Cloud) (I-SMAC), 2017 International Conference on IEEE, pp. 137-140, 2017. [15] M Katyal, A Mishra, \"A comparative study of load balancing algorithms in cloud computing environment\".arXiv preprint arXiv:1403.6918, 2014. [16] Q Zhang, L Cheng, R Boutaba, \"Cloud computing: state-of-the-art and research challenges\". Journal of internet services and applications, vol.1, no. 1, pp.7-18, 2010. [17] Kaneria, RK Banyal, \"Analysis and improvement of load balancing in Cloud Computing\",InICT in Business Industry & Government (ICTBIG), International Conference, IEEE. pp.1-5, 2016. [18] S Ray, A De Sarkar, \"Resource allocation scheme in cloud infrastructure\", In Cloud & Ubiquitous Computing & Emerging Technologies (CUBE), 2013 International Conference, IEEE. pp. 30-35, 2013. [19] Hameed A, Khoshkbarforoushha A, Ranjan R, Jayaraman PP, Kolodziej J, Balaji P, Zeadally S, Malluhi QM, Tziritas N, Vishnu A, Khan SU. “A survey and taxonomy on energy efficient resource allocation techniques for cloud computing systems”. Computing.vol.98, no.7, pp.751-74, 2016.. [20] FF Moghaddam, M Ahmadi, S Sarvari, M Eslami, A Golkar, \"Cloud computing challenges and opportunities: A survey\". In Telematics and Future Generation Networks (TAFGEN), 2015 1st International Conference, IEEE. pp. 34-38, 2015. [21] H Wen, Z Hai-ying, L Chuang, Y Yang, “Effective load balancing for cloud-based multimedia system\". In Electronic and mechanical engineering and information technology (EMEIT), 2011 international conference, IEEE.Vol. 1, pp.165-168, 2011. [22] Lal A, Krishna CR,“Critical Path-Based Ant Colony Optimization for Scientific Workflow Scheduling in Cloud Computing Under Deadline Constraint”. InAmbient Communications and Computer Systems, pp. 447-461, 2018. [23] Guddeti RM, Buyya R,“A Hybrid Bio-Inspired Algorithm for Scheduling and Resource Management in Cloud Environment”.IEEE Transactions on Services Computing.2017. [24] Z Xiao, W Song, Q Chen, \"Dynamic resource allocation using virtual machines for cloud computing environment\". IEEE transactions on parallel and distributed systems, vol.24, no.6, pp.1107-17, 2013. [25] Z Zhang, X Zhang, \"A load balancing mechanism based on ant colony and complex network theory in open cloud computing federation\". InIndustrial Mechatronics and Automation (ICIMA), 2010 2nd International Conference, IEEE, Vol. 2, pp. 240-243, 2010. [26] ST Maguluri, R Srikant, L Ying, \"Stochastic models of load balancing and scheduling in cloud computing clusters\", In INFOCOM, 2012 Proceedings IEEE, pp. 702-710, 2012. [27] X Meng, V Pappas, and L Zhang, \"Improving the scalability of data center networks with traffic-aware virtual machine placement,” InINFOCOM, 2010 Proceedings IEEE, pp. 1-9,2010. [28] YO Yazir, C Matthews, R Farahbod, S Neville, A Guitouni, S Ganti, Y Coady, \"Dynamic resource allocation in computing clouds using distributed multiple criteria decision analysis, \" In Cloud Computing (CLOUD), 2010 IEEE 3rd International Conference, IEEE, pp. 91-98, 2010. [29] K Tsakalozos, H Kllapi, E Sitaridi, M Roussopoulos, D Paparas, A Delis, \"Flexible use of cloud resources through profit maximization and price discrimination,\" In Data Engineering (ICDE), 2011 IEEE 27th International Conference, IEEE, pp. 75-86, 2011. [30] U Sharma, P Shenoy, S Sahu, A Shaikh, \"A cost-aware elasticity provisioning system for the cloud, \" In Distributed Computing Systems (ICDCS), 2011 31st International Conference, IEEE, pp. 559-570, 2011. [31] Yang J, Jiang B, Lv Z, Choo KK,“A task scheduling algorithm considering game theory designed for energy management in cloud computing, ” Future Generation Computer Systems. 2017. [32] S Khatua, MM Manna, N Mukherjee, \"Prediction-based instant resource provisioning for cloud applications, \" In Proceedings of the 2014 IEEE/ACM 7th International Conference on Utility and Cloud Computing, IEEE Computer Society, pp. 597-602, 2014 [33] J Grover, S Katiyar, \"Agent based dynamic load balancing in Cloud Computing,” In Human Computer Interactions (ICHCI), 2013 International Conference, IEEE, pp.1-6, 2013. [34] L Wu, SK Garg, R Buyya, \"SLA-based admission control for a Software- as-a-Service provider in Cloud computing environments,” Journal of Computer and System Sciences, vol.78 no.5, pp.1280-99, 2012 [35] Manickachezian, R. & Eswaran, Sivaraman. (2018). Unevenness measurement using the support vector machine and dynamic multiservice load balancing with modified genetic algorithm in cloud-based multimedia system. International Journal of Computer Aided Engineering and Technology. 10. 732. 10.1504/IJCAET.2018.10015510. [36] M. Ibrahim et al., \"A Comparative Analysis of Task Scheduling Approaches in Cloud Computing,\" 2020 20th IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGRID), Melbourne, VIC, Australia, 2020, pp. 681-684, doi: 10.1109/CCGrid49817.2020.00-23. [37] G. Sheng, Y. Zhang and Y. Li, \"Improved Scheduling Algorithm for Instance-Intensive Cloud Workflow,\" 2019 2nd International Conference on Safety Produce Informatization (IICSPI), Chongqing, China, 2019, pp. 636-638, doi: 10.1109/IICSPI48186.2019.9095905. [38] K. M. S. U Bandaranayake, K. P. N Jayasena and B. T. G. S. Kumara, \"TRETA-A Novel Heuristic Based Workflow Scheduling Algorithm in Cloud Environment,\" 2020 IEEE 15th International Conference on Industrial and Information Systems (ICIIS), RUPNAGAR, India, 2020, pp. 363-368, doi:10.1109/ICIIS51140.2020.9342666.

Copyright

Copyright © 2023 Rohith Sunilkumar Nair, Ameya Mahadev Gonal, Prof. Girish Rao Salanke N.S, Prof. B. K. Srinivas. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.