=Paper=
{{Paper
|id=Vol-1482/676
|storemode=property
|title=Distributed adaptive VoIP load balancing in hybrid clouds
|pdfUrl=https://ceur-ws.org/Vol-1482/676.pdf
|volume=Vol-1482
}}
==Distributed adaptive VoIP load balancing in hybrid clouds==
https://ceur-ws.org/Vol-1482/676.pdf
Суперкомпьютерные дни в России 2015 // Russian Supercomputing Days 2015 // RussianSCDays.org
Distributed Adaptive VoIP Load Balancing in Hybrid Clouds*
Jorge M. Cortés-Mendoza 1, Andrei Tchernykh 1†, Alexander Yu. Drozdov 2, Pascal Bouvry3,
Ana-Maria Simionovici3, Arutyun Avetisyan 4
CICESE Research Center1, Moscow Institute of Physics and Technology2, University of Lux-
embourg3, ISP RAS4
Cloud computing as a powerful economic stimulus widely being adopted by many compa-
nies. However, the management of cloud infrastructure is a challenging task. Reliability,
security, quality of service, and cost-efficiency are important issues in these systems. They
require resource optimization at multiple layers of the infrastructure and applications. The
complexity of cloud computing systems makes infeasible the optimal resource allocation,
especially in presence of uncertainty of very dynamic and unpredictable environment.
Hence, load balancing algorithms are a fundamental part of the research in cloud compu-
ting. We formulate the problem of load balancing in distributed computer environments and
review several algorithms. The goal is to understand the main characteristics of dynamic
load balancing algorithms and how they can be adapted for the domain of VoIP computa-
tions on hybrid clouds. We conclude by showing how none of these works directly address-
es the problem space of the considered problem, but do provide a valuable basis for our
work.
1. Introduction
Cloud computing is a pay-per-use model for enabling on-demand computing resources. It is de-
fined as a model for enabling ubiquitous, convenient, on-demand network access to a shared pool of
configurable computing resources (e.g., networks, servers, storage, applications, and services) that can
be rapidly provisioned and released with minimal management effort or service provider interaction
[1].
In on-demand computing, resources are made available to the user as needed and they provide
easy way to keep and retrieve data and files. A generic structure and components of clouds is com-
monly described as a three-layered concept, each layer treated as a service.
Infrastructure as a Service (IaaS) involves offering computational resources that include pro-
cessing, disk storage, network and other computational resources, and users have control over the
software, storage and processing capacity. Platform as a Service (PaaS) involves offering a develop-
ment platform on the cloud, the user deploys applications and possible configuration settings for the
application. Software as a Service (SaaS) includes the software offered by the cloud. The user rents
software applications running on a demand infrastructure and can only change the application configu-
ration settings.
Cloud computing considers four deployment models based on locations of the clouds: Public, Pri-
vate, Community, and Hybrid. Public Clouds are managed by their providers. The infrastructure is
shared between organizations, users grant access to the resources through subscriptions. Private
Clouds can be accessed only by the provider of the resources; the cloud is fully owned by a single
company who has total control over the applications running on the infrastructure. Hybrid Cloud com-
bines private and public cloud, user applications run on a private or public infrastructure (applications
with relative importance are scheduled in private clouds). Community Clouds allow sharing infrastruc-
ture between organizations with common concerns or similar polices.
We address five important characteristics of cloud computing: cost, performance, scalability, mo-
bility, and virtualization. (Cost) In cloud computing, the resources do not belong to users, and the us-
ers do not have to buy or maintain them, the initial investment is not needed. (Performance) The goal
is to improve the processing power or maximize storage capacity by consolidation of CPUs, memory,
*
This work is partially supported by the Ministry of Education and Science of Russian Federation under contract
No02.G25.31.0061 12/02/2013 (Government Regulation No 218 from 09/04/2010).
†
Corresponding author
676
Суперкомпьютерные дни в России 2015 // Russian Supercomputing Days 2015 // RussianSCDays.org
and storage to deploy services. (Scalability) The user can increase or decrease resources (storage,
CPUs, memory, etc.) at any time, he only pays for what he is really used. (Mobility) The data can be
accessed anytime anywhere, the user only needs a device with internet connection (laptop,
smartphone, etc.). (Virtualization) It is a technology used to share resources. It makes a single physical
resource appear as many individually separate virtual resources. It allows the use of the server capacity
effectively, reducing unused CPU cycles, and minimizing wasted energy.
The management of the large-scale cloud infrastructure is a challenge. Resource management for
clouds has been subject to research and development for many years. Here, we discuss different as-
pects related with our study.
Load balancing is a mechanism to improve the overall performance of the system by distribution
of the workload between the nodes. By reducing idle times, providers can improve the profit and
achieve a higher user satisfaction. Proper load balancing can help on utilizing the available resources,
thereby, minimizing the resource consumption [2]. It also helps in enabling scalability, avoiding bot-
tlenecks and over-provisioning, reducing response time, and energy consumption.
Quality of service (QoS). In cloud computing, providers need to ensure that sufficient amount of
resources are provisioned to ensure that QoS requirements of cloud service consumers such as dead-
line, response time, and budget constraints are met [3, 19]. Service Level Agreements (SLAs) are
binding contracts between a service provider and the user.
These SLAs contain the list of services, metrics, responsibilities of the provider and auditing
mechanisms. Any violation will lead to a penalty. Several ways exist to provide QoS: scheduling, ad-
mission control, traffic control, dynamic resource provisioning, etc.
Energy management. Energy consumption is determined by hardware efficiency, resource man-
agement system deployed on the infrastructure, and the efficiency of applications running on the sys-
tem. The efficiency is very important due to its impacts to users in terms of resource usage costs,
which are typically determined by the total cost of ownership incurred by a resource provider [4]. The
goal is to avoid utilization of more resources than is required by the application. One solution to this
problem is to migrate virtual machines for one node to another and shut down the idle nodes. Such a
Dynamic Component Deactivation (DCD) policy [5, 18] switches off parts of the computer system
that are not utilized.
The paper is structured as follow. The next section presents several important issues of dynamic
load balancing algorithms. Section 3 provides details of the Internet telephony and voice over IP te-
lephony (VoIP). Section 4 describes VoIP provider model and quality of service (QoS). Section 5 pre-
sents more formal definition (jobs, cloud infrastructure and criteria). Section 5 gives a brief overview
of the load balancing algorithms in cloud computer environments. Section 6 presents our load balanc-
ing algorithm. Section 7 concludes the paper.
2. Load balancing
Load balancing is a job distribution decision-making process used in many production systems
and computing. It is widely known as a technique for the efficient utilization of resources, and it can
be implemented with hardware and software support. Jobs arrival rate, communication delay, the vari-
ability of the job parameters, and other factors affect the performance of the systems, to deal with such
complex factors it is essential to design efficient and scalable load balancing algorithms. It helps in
implementing fail-over, scalability, avoiding bottlenecks, over-provisioning, reducing response time,
reducing energy consumption, etc. [6]
Load balancing of services, computational jobs, virtual machines, virtual storages, database re-
quests, and VoIP traffic on the network are identified as a major concern for the efficient use of cloud
computing.
The development of an effective dynamic load balancing algorithm involves many important is-
sues: load estimation, load levels comparison, performance indices, system stability, amount of infor-
mation exchanged among nodes, job resource requirements estimation, job selection for transfer, re-
mote nodes selection, etc. [7]. Important aspects of the problem are distribution of the nodes, storage
replications, and virtual machine migrations.
677
Суперкомпьютерные дни в России 2015 // Russian Supercomputing Days 2015 // RussianSCDays.org
Distribution of nodes. Some algorithms are efficient only if the nodes are closely located and the
communication delays are negligible. However, it is necessary to consider communication delay in the
cloud infrastructure.
Storage replications. Full replication of data increases the storage and communication overhead.
Partial replication saves information in different nodes (with overlap), so that, the utilization, fault tol-
erance and availability of data are increased.
Virtual machine migrations. A heavily loaded node can migrate its virtual machines (VM) to re-
duce the overload, but determinate which VMs have to be moved, the destination node, and the profit
of the migration are questions difficult to answer.
3. Internet telephony
The Internet telephony (VoIP–Voice over Internet Protocol) refers to the provisioning of commu-
nication services over the Internet, rather than via the traditional telephone network. VoIP services
significantly reduce calling rates, leading VoIP vendors continue offering extraordinary service using
modern cloud technology. Selection of a cloud based VoIP further reduces resource costs, adds new
features and capabilities, provides easier implementations, uniform deployments, and integrates ser-
vices that are dynamically scalable.
To deploy and manage effective telephony tools via clouds a variety of factors need to be im-
proved. The most important one is the utilization of the infrastructure.
Fig. 1. VoIP architecture.
Fig. 1 shows the general VoIP architecture. The elements of communication infrastructure con-
nect phones remotely through the Internet. The servers use software to emulate a telephone exchange.
A drawback of this architecture arises when the hardware reaches its maximum amount of connec-
tions.
Traditional VoIP solutions are not very scalable. It is necessary to duplicate infrastructure or re-
place existing physical hardware.
Fig. 2 shows a cloud based solution of the problem. The voice nodes are virtual machines that ex-
ecute a variety of services (call transfer, voice mail, music on hold, etc.). The advantage of this archi-
tecture is the increasing scalability.
Fig. 2. Single cloud VoiP architecture.
Fig. 3 shows next step in distributed cloud based VoIP architecture, where voice nodes are
grouped in data centers geographically distributed.
678
Суперкомпьютерные дни в России 2015 // Russian Supercomputing Days 2015 // RussianSCDays.org
(a) (b)
Fig. 3. Data centers in clouds (a) and the cloud federation (b).
However, it has several unsolved problems. To optimize the overall system performance, a pro-
cessor load of the voice signal processing over IP (jobs) should be balanced. The overload of a proces-
sor reduces the quality of the call. A similar problem could happen with the network capability. Fur-
ther, the processor idle time increases the useless expenses of the provider.
Load-balancing maximizes VoIP performance by keeping processor idle time and interprocessor
communication overhead as low as possible. To minimize the overall computation time, all processors
should contain the same amount of computational work.
It is necessary to design a multi-level distributed VoIP load balancer to improve the local load
imbalance in data centers, and new techniques to scale on federation of data centers (Figure 4).
The most important cause of load imbalance in VoIP is the dynamic nature of the problem over
time (in both computational and communication costs). Other causes may include the interference
from other users in time-sharing mode, the migration process, the time arrival, variability on the utili-
zation process, etc.
Most load balancing algorithms focus on deterministic environments assuming knowledge of the
user jobs and system parameters. In general, it is impossible to get exact knowledge about the system.
Parameters like processor speed, number of available processors, and actual bandwidth are changing
over the time. However, load balancing algorithms should search how to improve resources and ensure
Quality of Service (QoS).
4. Cloud provider model
An important factor of VoIP on cloud computing is the rental time of the infrastructure. The VoIP
providers rent the resources during a time interval, this factor affects the incoming of the providers and
even can increase the cost of the service, for this reason is important to consider when a new resources
are rented and the rented time interval.
Fig. 4 shows an example of VMs that provide VoIP services. This example consider the load bal-
ancing of the utilization on the VMs. In Figure 5a, the VoIP provider needs three VMs to deal with the
load of time during the day. starts its execution when cannot process all calls in the system.
During the rental time of , its utilization is low, in this case the provider has to paid for the rent of
three VMs. Figure 5b shows the same example, but in this case the load balancer distributes the arrival
calls between and . When rental time is almost completed, a consolidation technique is
used to reduce utilization (number of call running on ). This approach helps providers to re-
duce the number of VMs for calls processing.
679
Суперкомпьютерные дни в России 2015 // Russian Supercomputing Days 2015 // RussianSCDays.org
(a) (b)
Rent of 3 virtual machines. Rent of 2 virtual machines.
VM 2 VM 2
Amount of VM
VM 1 VM 1
Time-interval Overflow of call on the VM Calls consolidation (Migration)
Fig. 4. VoIP load balancing.
VoIP providers also have to guarantee QoS to the users. Quality of service requirements for VoIP
are very important but several factor can affect the quality of calls. The quality degradation is deter-
mined by the transit of the packets across the Internet, queuing delays at the routers, packet travel time
from source to destination, jitter (deviations of the packet inter-arrivals), packet loss, call set-up time,
and call tear-down time, etc. Processor utilization of 100% provides the best expected performance.
However, with increasing number of call, hence utilization, CPU cannot be able to handle the stress
anymore and jitters and broken audio symptoms will appear (Fig. 5).
Fig. 5. Voice quality versus processor load (utilization).
Fig. 6 shows an example of VM rental to provide VoIP services with QoS. In Figure 6a, the VoIP
provider needs three VMs to deal with the load during a day. In this case, the QoS is not considered as
independent optimization criteria. The utilization of is above the utilization threshold. Figure 6b
shows the same example. However, load balancing guarantees the QoS by maintaining the VMs utili-
zation under the utilization threshold. Moreover, it reduces the amount of VMs to process calls.
This model allows providers to deploy services in different countries by renting infrastructure in
public or/and private clouds providers.
(a) (b)
Rent of 3 virtual machines without QoS. Rent of 3 virtual machines with QoS.
VM VM2
2
Amount of VM
VM1 VM
1
Time-interval Overflow of utilization to ensure Utilization upper threshold.
QoS
Fig. 6. VoIP with Quality of service.
680
Суперкомпьютерные дни в России 2015 // Russian Supercomputing Days 2015 // RussianSCDays.org
5. Formal definition
We address load balancing problem in the hierarchical federated cloud environment, where clouds
of different providers collaborate to be able to fulfill requests during peak demands. We assume heter-
ogeneous clouds and data centers with different number of cores, execution speed, energy efficiency,
amount of memory, bandwidth, etc.
5.1 Infrastructure model
Let us consider cloud that consists of nodes (data centers, sites) . Each node
, for all , consists of servers (blades, boards) and processors per board. We assume
that processors in the data center are identical and have the same number of cores. Let be the num-
ber of identical cores of one processor in .
We denote the total number of cores belonging to the data center by , and be-
longing to all data centers of the cloud by .
The processor of data center is described by a tuple { , , , , }, where is a
measure of instruction execution speed (MIPS), is the amount of memory (MB), is the
available bandwidth (Mbps), and is energy efficiency (MIPS per watt).We assume that data cen-
ters have enough resources to execute any job but their resources are limited.
A data center contains a set of routers and switches that transport traffic between the servers and
to the outside world. They are characterized by the amount of traffic flowing through it (Mbps). A
switch connects a redistribution point or computational nodes. The connections of the processors are
static but their utilization change, overload can occur due to a large amount of I/O being pushed
through it. The interconnection network architecture is Three-tier data center architectures that in-
clude: access, aggregation, and core layers. The interconnection between clouds will be done through
public Internet.
In addition, to satisfy requests during the peak demands that exceed the capacity of the cloud , it
collaborates with external independent clouds (sites) .
5.2 Job model
We consider n independent jobs that must be scheduled on federation of clouds. The
job (call) is described by a tuple { } that consist of: its release time , the duration
of the job and the contribution to the processor utilization of the job j. The release time of a job is not
available before the job is submitted, the duration of the job is unknown until the job has completed its
execution, and contribution is a constant for a given job. Due to the virtualization technique and re-
source sharing, the resources are constantly changing. A job can be allocated to one cloud only. Jobs
submitted to the one cloud can be migrated to another one.
The admissible set of data centers for a job is defined as a set of indexes { } of data
centers that can be used for migration of the job.
5.3 Optimization criteria
In this paper, two criteria are considered for the model: the billing hours for VMs to provide a
service, and their utilization to increase quality of service.
In order to optimize VoIP cloud solutions, we use metrics that are useful for systems with VMs.
They have to allow the provider to measure the cost of the system in terms of number of demanded
VMs and time of their using. These metrics enable us to compare different load balancing policies:
total VM rented ( ) (number of billing hours) and utilization of VM ( ).
allows providers to measure the cost of the system in terms of parameters that helps
him to establish utility margins.
In general, QoS standards for VoIP traffic are set for voice. One of the possible generalizations of
the voice quality is processor utilization. Each codec provides a certain quality of speech only if pro-
cessor utilization is low enough in order to ensure QoS. If the number of calls is increased, and utiliza-
681
Суперкомпьютерные дни в России 2015 // Russian Supercomputing Days 2015 // RussianSCDays.org
tion is close to 100%, CPU cannot be able to handle the stress anymore and reduced audio quality will
appear. Hence, we use as a way of measure the user satisfaction for the service.
6. Related work
Previously, we briefly overview the load balancing algorithms in different computer environments
[8]. Now, we consider the advances in the field of cloud computing. Table I presents the main charac-
teristics of the algorithms, and metrics used to study their quality.
HBB -LB - Honeybee Foraging Behavior [9]. It is a dynamic load balancing algorithm for
scheduling of tasks in cloud computing environment. The proposed algorithm balances the priority of
tasks on the machines to minimize the waiting time of the tasks in the queue. The tasks (honey bees)
are removed from the over loaded VM, they update the number of priority tasks and load of VMs, this
will help tasks to choose a VM based on load and priorities. Whenever a high priority task has to be
submitted to other VMs, it should consider the VM that has minimum number of high priority tasks so
that the particular task will be executed at the earliest.
DT - PALB - Power Aware Load Balancing [10]. A new version of the algorithm Power Aware
Load Balancing named Double Threshold PALB uses the migration of VMs to minimize the energy
consumption in the system. When the utilization of a node is under 25% (lower threshold), the load
balancer migrates workload (VMs) of the node to reduce its utilization to zero and shut down the node.
TBSLB-PSO - Particle Swarm Optimization [11]. It is a load balancing mechanism for cloud
computing. This algorithm uses Central Task Scheduler (CTS) to transfer extra tasks from an over-
loaded VM to a new similar VM by applying the information on a blackboard; the blackboard contains
all cloud schedulers about VM features, executing tasks and Quality of Service (QoS). The migration
process considers the amount of data, memory, bandwidth and numbers of CPU of VM. Idle Physical
Machines (PM) will not be chosen as the new PM hosts, this allows to decrease energy consumption.
LBS-BF - Load Balance Scheduling Based on firefly [12]. It is a load balancing mechanism for
cloud computing. The load balancer computes a load index for resources shared and it will be initiated
to effectively use the resources dynamically. The fireflies attraction is linked to objective function and
monotonic decay of the attractiveness with distance, it (in the firefly algorithm) helps to generate
scheduling index and the distance calculation serves to find the closely associated nodes in the cloud
network. The technic proposed three parameters for the algorithm: attraction between the nodes and
the request, the scheduling index, and the distance between nodes. This parameters consider the CPU
rate, memory rate, processing time and the loads to the nodes.
A2LB - Autonomous Agent Based Load Balancing Algorithm [13]. It is a dynamic load bal-
ancing algorithm for cloud environment. The load balancer mechanism comprises of three agents:
Load agent (LA), Channel agent (CA) and Migration agent (MA). LA is a static agent and it controls
information policy and maintains all detail of a datacenter, it most important job is to calculate the
load on every available VM. CA is a static agent, it controls the transfer policy, selection policy and
location policy. Finally, MA is an ant (special category of mobile agents), it will move to other data-
centers and communicate to enquire the status of VMs.
GA – Genetic Algorithm based on load balancing [14]. It is a load balancing algorithm for
cloud computing. It uses a binary representation for the chromosomes, a random single point crosso-
ver, and a mutation with probability of 0.05. This algorithm considers an estimate of penalty (delay
cost), the amount of money that cloud service provider needs to pay to customer when job finishing
time being more than the deadline advertised by the service provider
Table 1. Load balancing algorithms.
Characteristics Metrics
Energy consump-
Non-clairvoyant
Migration cost
Response time
Decentralized
Fault tolerant
Hierarchical
Performance
Centralized
Replication
Throughput
Utilization
Scalability
Migration
Dynamic
Overhead
Off line
On line
Static
QoS
Cost
tion
682
Суперкомпьютерные дни в России 2015 // Russian Supercomputing Days 2015 // RussianSCDays.org
HBB-LB Task
DT-PALB VM
TBSLB-PSO Task
LBS-BF
A2LB
GA
FFA-DLB Task
FFA-DLB - Fuzzy-based Firefly Algorithm for Dynamic Load Balancing in Cloud Compu-
ting [15]. It is a dynamic load balancing in Cloud computing environment and it is a combination of
Firefly algorithm with the fuzzy logic, this algorithm separate the cloud based on the frequent node
allocation to balance the load across the variety of partitions. The goal is to separate the hotspots and
least loaded nodes, then classify the nodes into groups (like lightly loaded, normal, and heavily load-
ed). The set of tasks enter into the load balancer after the partition of the cloud. This algorithm consid-
er a balancing factor based on the parameters of the VM and the files to be processed from input. The
fuzzy inference engine determine to assign the tasks, with a condition that already assigned tasks are
migrated only when a high necessity arises.
7. Adaptive load balancing algorithm
Campos and Scherson proposed a dynamic distributed load balancing algorithm named Rate of
Change (RoC-LB) [16]. The balancer (Bal) makes job distribution decisions at runtime, locally and
asynchronously. Each Bal considers its own load; migration does not depend on the load of other Bals.
The migration decision depends on current load, load changes in the time interval (Rate of Change),
and current load balancing parameters.
To define WHEN load balancing should be started the algorithm considers three thresholds:
upper bound, lower bound, and critical bound. If the load is larger than then the Bal is con-
sidered as a source of the load and can satisfy job requests. If the predicted load is less than the Bal
is considered as a sink. When the current load is between these two bounds, the Bal is in the neutral
state. However, if the load or the predicted load fall below , the Bal immediately initiates a request
for a load.
We extend this algorithm and design a VM-Aware Adaptive Rate of Change (VMA-AdRoC) al-
gorithm that is based on an adaptive decision policy and virtual machine utilization optimization.
Adaptability is essential for the efficient use of cloud infrastructure. Clouds differ from previous com-
puting environments in the way that they introduce a continuous uncertainty into the computational
process. The uncertainty becomes the main feature of the cloud computing and the principal difficulty
of the efficient resource management [17].
There are several major sources of uncertainty: dynamic elasticity, dynamic performance chang-
ing, virtualization, loosely coupling application to the infrastructure, among many others. A workload
in such an environment is not predictable and can be changed dramatically. It is impossible to get ex-
act knowledge about the system. Parameters such as an effective processor speed, number of available
processors, and actual bandwidth are changing over the time.
VMA-AdRoC takes into account these uncertainties. The accuracy of each balancing decision de-
pends on the actual cloud characteristics at the moment of balancing. Cloud parameters are changing
over time and balancing parameters should be adapted to these changes. This dynamic and adaptive
approach can cope with different workloads, and cloud properties. To adjust , , and , the past
information within a given time interval can be analyzed to determine an appropriate parameters. This
interval should be set according to the dynamics of the system.
Let be the utilization of ith Bal at time . Let be the utiliza-
tion change during sample interval = . We named it utilization change speed or utilization
consumption speed. The sampling time interval is an adaptive parameter; finer sampling allows de-
tecting the need to balance the system faster, but it generates a larger communication overhead.
Bal uses as a predictor of the future utilization. It can be also used to estimate the number of
sampling intervals to reach an idle state. Let be the response delay of at time . It is an adaptive
683
Суперкомпьютерные дни в России 2015 // Russian Supercomputing Days 2015 // RussianSCDays.org
parameter, and it is defined as the time it takes between the initiation of a load request and the recep-
tion of load. If the time to reach idle state is less than , then Bal must initiate a migration re-
quest. Let us note that and are independent from others Bals.
Fig. 7 shows possible load balancing scenarios. Solid line shows real utilization, dashed lines are
predicted utilization.
Fig. 7. Dynamic load balancing scenario.
At time the Bal immediately initiates a request for load regardless of the predicted future utili-
zation based upon the estimation value or it initializes migration of the load to other VM. Esti-
mation on is under but Bal cannot initiates a request because the request at is not arrived, a
new request only can be generate at . In VMA-AdRoC, unlike RoC-LB, if the utilization is above
than then the Bal sends jobs to the sinks.
To define WHERE a load is requested from or send to, each Bal keeps two lists. The sink list rec-
ords Bals that previously needed jobs, and source list enrolls Bals that previously offered jobs. Bal that
initiates a request is considered to be a sink. A sink selects a Bal from its source list for a load request,
and sends a requesting message. The source can accept the request or broadcast the request to other
Bals from its own source list. Bal does not send several load requests at the same time. It has to wait
an answer for the first request until it sends another message. The result of this message is the load
coming from other Bal or the request comes back as unfulfilled.
In our algorithm, each element of the source and sink lists includes not only IP address like in
RoC-LB, but utilization , of the corresponding Bal. This information is not accurate and updated
dynamically. In our case, choosing the sink/source node is a two-parameter problem. In the future
work, we consider also requested load , , , answer time of the request , admis-
sibility of Bals, etc. The goal is to choose the most adequate compromise solution.
8. Conclusions
In this paper, we formulate and discuss load balancing problem addressing VoIP in cloud compu-
ting federation. We define models for VoIP load balancing. We overview the last advances of load
balancing in distributed computer environments to understand the main characteristics and require-
ments of load balancing algorithms. We show that none of these works directly addresses the problem
space of the considered problem, but do provide a valuable basis for our study.
In real clouds, the load balancing bounds can be dynamically adjusted to cope with the dynamic
workload situation. To this end, the past workload must be analyzed for a certain time interval to de-
termine an appropriate lower and upper bounds. The time interval should be set according to workload
characteristics, communication delays, and cloud configurations.
684
Суперкомпьютерные дни в России 2015 // Russian Supercomputing Days 2015 // RussianSCDays.org
While the scope of this work is the introduction of the adaptive dynamic distributed load balanc-
ing algorithm, in future work, we also intend to evaluate the practical performance of the proposed
strategy and their derivatives to assess its actual efficiency and effectiveness. To this end, we plan
simulations using real VoIP traces and corresponding VoIP cloud configurations. Further, we will
compare our approach with other existing strategies which are typically based on round robin ap-
proach.
References
1 Mell, P., and Grance, T. (2011). The NIST definition of cloud computing (draft).NIST special
publication, 800 (145), 7.
2 García, J. L. G., Yahyapour, R., and Tchernykh, A. (2013). “Load Balancing for Parallel Compu-
tations with the Finite Element Method”. Computación Y Sistemas, 17(3), 299–316.
3 Buyya, R., Calheiros, R. N., and Li, X. (2012). “Autonomic Cloud computing: Open challenges
and architectural elements”. In Emerging Applications of Information Technology (EAIT), 2012
Third International Conference on, 3-10. IEEE.
4 Beloglazov, A., Abawajy, J., and Buyya, R. (2012). “Energy-aware resource allocation heuristics
for efficient management of data centers for cloud computing”. Future Generation Computer Sys-
tems, 28(5), 755-768.
5 Tchernykh, A., Pecero, J. E., Barrondo, A., and Schaeffer, E. (2014). “Adaptive energy efficient
scheduling in Peer-to-Peer desktop grids”. Future Generation Computer Systems. Elsevier Sci-
ence, Volume 36, July 2014, Pages 209–220. DOI: 10.1016/j.future.2013.07.011,
6 Kansal, N. J., and Chana, I. (2012). “Cloud Load Balancing Techniques: A Step Towards Green
Computing”. IJCSI International Journal of Computer Science Issues, 9(1), 238-246.
7 Alakeel, A. M. (2010). “A guide to dynamic load balancing in distributed computer systems”. In-
ternational Journal of Computer Science and Information Security, 10(6), 153-160.
8 Tchernykh, A., Cortés-Mendoza, J. M., Pecero, J. E., Bouvry, P., and Kliazovich, D. (2014).
Adaptive energy efficient distributed VoIP load balancing in federated cloud infrastructure. In
Cloud Networking (CloudNet), 2014 IEEE 3rd International Conference on (pp. 27-32). IEEE.
9 Krishna, P. Venkata (2013). "Honey bee behavior inspired load balancing of tasks in cloud com-
puting environments." Applied Soft Computing 13.5: 2292-2303.
10 Adhikari, J., and Patil, S. (2013, July). Double threshold energy aware load balancing in cloud
computing. In Computing, Communications and Networking Technologies (ICCCNT), 2013
Fourth International Conference on (pp. 1-6). IEEE.
11 Ramezani, F., Lu, J., and Hussain, F. K. (2014). Task-based system load balancing in cloud com-
puting using particle swarm optimization. International Journal of Parallel Programming, 42(5),
739-754.
12 Florence P., and Shanthi V., (2014). A load balancing model using firefly algorithm in cloud com-
puting. Journal of Computer Science, 10(7), 1156-1165.
13 Singh, A., Juneja, D., and Malhotra, M. (2015). Autonomous Agent Based Load Balancing Algo-
rithm in Cloud Computing. Procedia Computer Science, 45, 832-841.
14 Dasgupta, K., Mandal, B., Dutta, P., Mandal, J. K., and Dam, S. (2013). A genetic algorithm (ga)
based load balancing strategy for cloud computing. Procedia Technology, 10, 340-347.
15 Susila, N., Chandramathi, S., and Kishore, R. (2014). A Fuzzy-based Firefly Algorithm for Dy-
namic Load Balancing in Cloud Computing Environment. Journal of Emerging Technologies in
Web Intelligence, 6(4), 435-440.
16 Campos, L. M., and Scherson, I. D. (2000). Rate of change load balancing in distributed and paral-
lel systems. Parallel Computing, 26(9), 1213-1230.
17 Tchernykh, A., Schwiegelsohn, U., Alexandrov, V., and Talbi, E. G. (2015). Towards Understand-
ing Uncertainty in Cloud Computing Resource Provisioning. SPU’2015 - Solving Problems with
Uncertainties (3rd Workshop). In conjunction with The 15th International Conference on Compu-
tational Science (ICCS 2015), Reykjavík, Iceland, June 1- 3, 2015. Procedia Computer Science,
Elsevier, Volume 51, Pages 1772–1781.
18 Tchernykh, A., Lozano, L., Schwiegelshohn, U., Bouvry, P., Pecero, J. E., and Nesmachnow, S.,
Alexander Yu. Drozdov (2015). Online Bi-Objective Scheduling for IaaS Clouds with Ensuring
685
Суперкомпьютерные дни в России 2015 // Russian Supercomputing Days 2015 // RussianSCDays.org
Quality of Service. Journal of Grid Computing, Springer-Verlag. DOI 10.1007/s10723-015-9340-
0
19 Schwiegelshohn, U., Tchernykh, A.: Online Scheduling for Cloud Computing and Different Ser-
vice Levels, 26th Int. Parallel and Distributed Processing Symposium, Los Alamitos, CA, pp.
1067–1074 (2012)
686