19-04-2013, 04:15 PM
COMPARATIVE ANALYSIS OF LOW-LATENCY ON DIFFERENT BANDWIDTH AND GEOGRAPHICAL LOCATIONS WHILE USING CLOUD BASED APPLICATIONS
COMPARATIVE ANALYSIS.pdf (Size: 222.33 KB / Downloads: 26)
ABSTRACT
Cloud computing, a approach of computing where scalable and elastic IT-related capabilities are provided as
shared assorted services (IaaS, PaaS,SaaS, DaaS), metered by use, to customers using internet technologies
built on top of diverse technologies like virtualisation, distributed computing, utility computing, and more
recently networking, web infrastructure and software services. It represents a paradigm shift in how we think
about our data, the role of our computing devices and on managing computing resources. Cloud computing
comes with Zero latency so the application hosted on cloud computing will act just like desktop application. The
speed of internet is main factor to avoid latency but internet speed is not same in all the geographical location.
From the survey it’s found out that some latency occurs in different geographical location. Even different
browser gives different latency. The problem of latency or low bandwidth may be automatically solve in
coming years as the speed of internet is changing day to day. If users are going to adopt cloud computing then
user must think of internet speed which is available in mean time. The internet connection should be good
enough to support cloud computing minimum requirement would be 1MPBS.
INTRODUCTION
Outsourcing of data center functionality and availability of desktop application online in web mode
via a network connection is what we tem cloud computing [50][52][51]. Companies and SME are all
under the cloud computing[49][48] many businesses has been move to cloud, moving of cloud cut
down the IT cost having the security with less IT staff, having the power of scalability and elasticity
of cloud computing has changes who we use the computer. The expenses have been cut down and
traffic of network has increase to double times [47]. Having a big question in mind the future of cloud
service will replace desktop application and we all will depend on internet [15][19]. Will the future of
internet support us? A concern for many organizations is that cloud computing relies on the
availability, quality and performance of their internet connection. Poor application performance
causes companies to lose customers, reduce employee productivity, and reduce bottom line revenue
[40][42]. Because application performance can vary significantly based on delivery environment,
businesses must make certain that application performance is optimized when written for deployment
on the cloud or moved from a data center to a cloud computing infrastructure [41].
CLOUD PROVIDER
Infrastructure as a Service (IaaS) provisions hardware, software, and equipments to deliver
software application environments with a resource usage-based pricing model. Infrastructure can scale
up and down dynamically based on application resource needs [3].
B. Platform as a Service (PaaS) offers a high-level integrated environment to build, test, and deploy
custom applications. Generally, developers will need to accept some restrictions on the type of
software they can write in exchange for built-in application scalability. An example is Google’s App
Engine, which enables users to build Web applications on the same scalable systems that power
Google applications, Web application frameworks, Python Django (Google App Engine), Ruby on
Rails (Heroku), Web hosting (Mosso), Proprietary (Azure, Force.com) [3].
RELATED WORK
Cloud computing promise to deliver desktop like application in web which is support by any internet
enabled device. Web applications are now hosted in elastic cloud environments where the unit of
resource allocation is a virtual machine (VM) instance. A variety of techniques can reduce the latency
of communication between VMs co-located on the same server in, say, a private cloud[17] on elastic
cloud systems, there will be co-located VMs of the same web application on the same physical server
that can take advantage of shared-memory IPC between the VMs. It is shown that Nahanni
memcached, used with VDE networking, can improve the total read-related latency for a workload by
up to 45% (i.e., read latest workload) compared to standard memcached, resulting from reductions in
cache read latency of up to 86%. When combined with state-of-the-art paravirtualized network
mechanisms, such as vhost, Nahanni memcached can still reduce the total read-related latency by up
to 29%.[5][6][8][9][11]
PROBLEM DEFINITION AND TECHNICAL STUDY
Latency seems to be major problem. Latency is also depends on geography and distance from server
to host. Great latency happens when the end user and internal networks system is far from the cloud
environment [20][21]. End user is platform dependent and all depend on net speed, if the end user is
not having high speed connection then end user has to cost more money to get the cloud speed [5]
[30]. A minimum requirement to use cloud is 1mpbs. All cloud is going to cost us in high bandwidth
even if they provide us the all *.aas service [31][32]. We can see the difference of cloud when we
access in cyber café or GPRS connection, when I check in cyber café and try to open Google docs it
took 20 sec while when it is try to open at the campus of university which provide 5.4 mbps it open in
2 sec.
CONCLUSION
In the context of different geographic location and different web browser web application has to be
design to avoid latency. Web application is design to give us like desktop application any delay in
latency would be a great lose too many clients. Cloud computing expect to give us no latency but
from the survey it is found that latency occurs on different geographical location. Above diagram and
chart shows that difference in response time in different location with different web browser. A better
algorithm is needed to prevent the latency while using the cloud computing. Cloud computing
represents a powerful and proven solution for IT departments to increase elasticity and cut costs
typically associated with the deployment of new platforms. However, the consideration of cloud
computing should not be limited to this view and should take into consideration the entire system
solution – from Cloud to end-user. Use of redundant cloud architecture, redundant data center
infrastructure and an innovative route optimization technology can provide the required performance
and availability of applications. Additionally, use of technologies beyond the cloud edge such as TCP
acceleration and content delivery networks can further improve content delivery to end-users.