Cloud System Failure Case Study: Analysis of the PayPal Outage

This case study examines the global PayPal outage caused by a cloud system failure, which resulted in significant transactional disruptions and financial losses. The outage, lasting approximately one hour with a subsequent partial outage, stemmed from the cloud service provider's inability to handle the high transaction volume. The report identifies potential causes, including network issues, technical glitches, software bugs, and inadequate infrastructure to manage peak loads. The analysis highlights the importance of robust cloud architecture, load balancing across geographically diverse data centers, and reliable partnerships with telecommunication providers to ensure service continuity. The study concludes that the PayPal incident underscores the critical need for cloud service providers to maintain adequate processing capabilities and implement resilient systems to prevent service disruptions.

Running head: CLOUD SYSTEM FAILURE
CLOUD SYSTEM FAILURE
Name of student
Name of university
Author’s note:

1
CLOUD SYSTEM FAILURE
Table of Contents
Introduction....................................................................................................................2
Discussion......................................................................................................................2
Reason for the outage.................................................................................................2
Conclusion......................................................................................................................4
References......................................................................................................................5

2
CLOUD SYSTEM FAILURE
Introduction
This report aims to discuss the case study of the cloud system failure and discuss the
reasons for the failure of the system. The considered situation of the cloud system failure is
the global power outage of the PayPal Company that led to the loss of significant amount of
capital. The reason of the cloud system failure has been briefly discussed in this report along
with the discussion of the effects has been provided in this report. Lastly, this report
concludes with the appropriate conclusion for the report.
Discussion
The PayPal outage led to the disruption of over $32 Billion in the transactions of the
company and this setback the company from the significant development of the company.
The PayPal is the online payment platform that offers the service of online payment that
could be executed using the internet and the company manages the transaction details of all
the transactions (Lim et al., 2014). On August 3, 2009, this platform suffered the global
outage for almost an hour and afterwards this situation, this service suffered the partial outage
for almost three and the half hours. During this period of time, the company claimed that the
cloud service provider was discovered to be processing almost an average of $2000 in the
payments per second and this load could not be handled properly by the cloud and the outage
took place (Malik & Shakshuki, 2014). The cloud outage could be refereed as the period of
time during which the services of the cloud are temporarily down and are not available for the
clients.
Reason for the outage
This unavailability of the cloud services could also be referred to the performance
inadequacy of services of the cloud according to the metrics of the SLA (Fu, Kim & Prior,
2015). As an instance, it could be considered that the incident while the outage might have

3
CLOUD SYSTEM FAILURE
the partial impact on the data center of the vendor for the execution of the required
maintenance and the measures of the restoration. The cloud downtime could be caused by the
number of the various factors, that could be categorised as the loss of the power, the issues in
network connectivity, the offline mode of the data centre for the maintenance, the failed
backups, the natural disaster, the bugs within software, attack by any hacker and the human
error. The situation of the PayPal is due to the significant load on the processing capabilities
of the cloud computing of the service provider (Malik & Shakshuki, 2016). The cloud
outages might result from the range of the caused beyond and within the control of any cloud
vendor. The situation of the PayPal where the cloud service provider could not provide the
service for several hours could be analysed as the reason for the networking issues and the
technical along with the software issues (Malik & Shakshuki, 2017).
The cloud infrastructure includes the complex system of the software and the
hardware technologies. The glitches and the bugs are highly probable causes for the
occurrences in the enterprise grade data centres that has the capability of providing the power
to the organisation regardless of the verticals and the size (Alam, Muqeem & Khan, 2018).
These particular issues associated with the technical sector might be overlooked or might
remain under the radar during the time when the analysis is being done and then it translates
into the actual incident in the service that impact majorly the end users. As the solution to the
issues is significantly not apparent or it might not be applicable for the instant resolution of
the issue, the service might remain in the state of the outage. The cloud vendors might form
the partnership with the service providers in the telecommunication sector and the
organisations that are managed by the government and it operates within the networks of the
communication across the long distances. The issues are linked with the networks beyond
organisatuion and specifically across the borders might be beyond control of any service
provider and specifically in the terms of the resolution of the issues in the connectivity (Tran

4
CLOUD SYSTEM FAILURE
& Feuerlicht, 2016). In this situation, the customers and the cloud vendors depends
significantly on the telecommunication partners for the ensuring that the service has been
restored. For the effective addressing of restriction, the majority of the large scale cloud
vendors solely operates globally in the multiple countries and they are have the ease of the
balancing of the workloads dynamically all across the geographically disparate datacentres. It
allows the company to constantly provide the service to the end users even during the
situation when the issues in the networking are significantly beyond the control of the internal
aspects (Shon et al., 2014). In the situation of the PayPal, it could be analysed that the cloud
service provider did not have the proper architecture for the proper handling of the additional
load that could be developing on the network of the organisation.
Conclusion
Therefore, it could be concluded that the PayPal organisation faced the significant
outage of the services from the cloud service provider due to the fact that the service provider
could not process the huge amount of the transactions. This unavailability of the cloud
services could also be referred to the performance inadequacy of services of the cloud
according to the metrics of the SLA. The cloud infrastructure includes the complex system of
the software and the hardware technologies. The glitches and the bugs are highly probable
causes for the occurrences in the enterprise grade data centres that has the capability of
providing the power to the organisation regardless of the verticals and the size. These
particular issues associated with the technical sector might be overlooked or might remain
under the radar during the time when the analysis is being done and then it translates into the
actual incident in the service that impact majorly the end users. The cloud vendors might
form the partnership with the service providers in the telecommunication sector and the
organisations that are managed by the government and it operates within the networks of the
communication across the long distances.

5
CLOUD SYSTEM FAILURE
References
Alam, S., Muqeem, M., & Khan, S. A. (2018). Review on security aspects for cloud
architecture. International Journal of Electrical and Computer Engineering, 8(5),
3129.
Fu, S., Kim, H., & Prior, R. (2015, December). FlowBox: Anomaly Detection Using Flow
Analysis in Cloud Applications. In 2015 IEEE Global Communications Conference
(GLOBECOM) (pp. 1-6). IEEE.
Lim, M. H., Lou, J. G., Zhang, H., Fu, Q., Teoh, A. B. J., Lin, Q., ... & Zhang, D. (2014,
December). Identifying recurrent and unknown performance issues. In 2014 IEEE
International Conference on Data Mining (pp. 320-329). IEEE.
Malik, H., & Shakshuki, E. M. (2014). Classification of post-deployment performance
diagnostic techniques for large-scale software systems. Procedia Computer
Science, 37, 244-251.
Malik, H., & Shakshuki, E. M. (2016). Towards identifying performance
anomalies. Procedia Computer Science, 83, 621-627.
Malik, H., & Shakshuki, E. M. (2017). Detecting performance anomalies in large-scale
software systems using entropy. Personal and Ubiquitous Computing, 21(6), 1127-
1137.
Shon, T., Cho, J., Han, K., & Choi, H. (2014). Toward advanced mobile cloud computing for
the internet of things: Current issues and future direction. Mobile Networks and
Applications, 19(3), 404-413.

6
CLOUD SYSTEM FAILURE
Tran, H. T., & Feuerlicht, G. (2016, September). Improving reliability of cloud-based
applications. In European Conference on Service-Oriented and Cloud Computing (pp.
235-247). Springer, Cham.