Sample COMP1680 Cloud Computing Coursework Report

COMP1680 Cloud Computing Coursework Report Sample

Detailed Specification

This Coursework is to be completed individually Parallel processing using cloud computing

The company you work for is looking at investing in a new system for running parallel code, both MPI and Open MP. They are considering either using a cloud computing platform or purchasing their own HPC equipment. You are required to write a short report analyzing the different platforms and detailing your recommendations. The report will go to both the Head of IT and the Head of Finance and so the report should be aimed at a non-specialist audience. Assume the company is a medium sized consultancy with around 50 consultants, who will likely utilize an average of 800 CPU hours each a month. Your report should include:

1) A definition of cloud computing and how it can be beneficial.
2) An analysis of the advantages and disadvantages of the different commercial platforms over a traditional HPC.
3) A cost analysis, assume any on site HPC will likely need dedicated IT support
4) Your recommendations to the company.
5) References

Solution

Introduction

The report is focusing to help the company to invest in a new system so that the company can run parallel code for OpenMP and MPI. The company is looking forward to use a cloud computing platform or to purchase separate HPC equipment for the company. At the very beginning of the report, the meaning of cloud computing is mentioned. In the same part, the definition of cloud computing is also mentioned. For Assignment Help All of the possible ways how cloud computing is useful and can it benefit a company are also written in the report. The report shows how the High-performance Computing act and how other commercial platform works. A comparison of both platforms has been done in the report. In the next section of the report, an analysis has been done about HPC and different commercial platforms. It has stated the different advantages and disadvantages of using different commercial platforms over the traditional HPC. All the points that show why the company should go with the particular platform have been highlighted in the given report. In the next section of the report, cost analysis has been mentioned. The cost analysis is done on the basis of an assumption made. A site has been assumed that the HPC will likely need dedicated IT support. After analyzing all these points, some of the recommendations to the company have been given. The recommendation that is given to the company will help the company to choose whether it should go with High-Performance Computing or it should choose a commercial platform. It will give the company an idea to invest in the new system for running parallel code. At the end of the report, a short conclusion has also been mentioned that summarizes each point presented in the report in short form.

Definition of Cloud Computing

Cloud computing is referred to as the delivery of service on demand like storage of data, computing power, and computer system resources. The delivery of services using internet, databases, networking, data storage, software, and servers. is termed cloud computing. The name of this on-demand availability is cloud computing because the information and data accessed by it are found remotely in the virtual space or in the cloud. All the heavy lifting or activities involved in the processing or crunching of the data in the computer device are removed by cloud computing (Al-Hujran et al. 2018). Cloud computing moves away from all the work to huge computer clusters very far away in cyberspace. The definition of cloud computing is, it is a general term for anything that involved delivering hosted service with the help of the internet. Both the software and hardware components are involved in the cloud infrastructure and these are required for implementing the proper cloud computing model. Cloud computing can be thought of in a different way such as on-demand computing or utility computing. Hence, in short, it can be said that cloud computing is the delivery of resources of information technology with the use of the internet (Arunarani et al. 2019).

Following are the points that show how cloud computing can be beneficial:

Agility- Easy access to a large number of technologies is given by the cloud so that the user can develop anything that the user has ever imagined. All the available resources can be quickly spined up to deliver the desired result from infrastructure services such as storage, database, computer, IoT, data lakes, analytics, machine learning, and many more. All the technologies services can be deployed in a matter of seconds and the ways to implement various magnitude. With the help of cloud computing, a company can test several new ideas and experiments to make differentiation of customer experiences, and it can transform the business too (Kollolu 2020).

Elasticity- With the presence of cloud computing in any business enterprise, the system become capable to adapt all the changes related to workload by deprovisioning and provisioning in autonomic manner. With cloud computing, these resources can diminished and can be maximize it instantly or to shrink the capacity of the business.

Security- Security of the data is one thing that almost every organization is concerned about. But, with the help of cloud computing, one can keep all the data and information very private and safe (Younge et al. 2017). A cloud host carefully monitors the security that is more important than a conventional in-house system. In cloud computing, a user can set different security according to the need.

Quality control- In a cloud-based system, a user can keep all the documents in a single format and in a single place. Data consistency, avoidance of human error, and risk of data attack can be avoided in cloud computing. If the data and information are recorded in the cloud-based system then it will show a clear record of the updates or any revisions made in the data. On the other hand, if the data are being kept in silos then there will be a chance of saving the documents in different versions that can lead to diluted data and confusion.

Analysis of Different Platforms vs HPC

HPC stands for High-Performance Computing is the ability to perform complex calculations and process data at a very high speed. The supercomputer is one of the best-known solutions for high-performance computing. It consists of a large number of computer nodes that work together at the same time to complete one or more tasks. This processing of multiple computers at a time is called parallel processing. Compute, network and storage are the three components of HPC solution. In general terms, to aggregate all the computing powers in such a way that it could deliver high performance than one could get out of a typical desktop is determined as HPC. Besides, having so many advantages, it has some disadvantages also. The following points show the analysis of the advantages and disadvantages of the different commercial platforms over a traditional HPC.

The advantage of different platforms over a traditional HPC is as follows:-

- With the perspective of the cost of equipment of high-performance computing, it seems to have been very high. The cost of using the high-performance computing cluster is not fixed and varies according to the type of cloud instances that the user uses to create the HPC cluster. If the cluster is needed for a very short time, the user uses on-demand instances for creating the cluster after the creation of the cluster, the instances are needed to be deleted (Tabrizchi and Rafsanjani 2020). This cost is almost five times higher than that of the local cluster. But when moving to the other platform of cloud computing, reduces the cost of managing and maintaining of IT system. in the different platforms, there is no need to buy any equipment for the business. The user can reduce the cost of using the cloud computing service provider resources. This is one of the benefits of other platforms over traditional HPC.

- In the different platforms of cloud computing rather than traditional HPC, cloud computing permits the user to be more flexible in their performance and work practices. For instance, the user is able to access the data from home, commute to and from the work, or on holiday. If the user needs access to data in case the user is off-site, the user can connect to the data anytime very easily, and quickly. In contrast, there is no such case with traditional HPC. In the traditional HPC, the user has to reach the system to access the data. In this HPC cloud, it is difficult to move the data that are used in HPC (Varghese and Buyya 2018).

- Having separate clusters in the cloud in the data centers poses some security challenges. But in the other platform of cloud computing, there is no such risk associated with data. One can keep and store the data in the system safely and privately in the case of other platforms. The user does not need to worry in any situation such as natural disaster, power failure or in any difficulties, the data are stored very safely in the system. This is another advantage of other platforms over traditional HPC (Mamun et al. 2021).

Disadvantages of other platforms over traditional HPC are as follows:

- The cloud, in any other setup, has the chance to experience some of the technical problems like network outages, downtimes, and reboots. These activities created troubles for the business by incapacitating the business operations and processes and it leads to damaging of business.

- In the several other platforms of cloud computing rather than HPC, the cloud service provider monitors own and manages the cloud infrastructure. The customer doesn't get complete access to it and has the least of the control over the cloud infrastructure. There is no option to get access to some of the administrative tasks such as managing firmware, updating firmware, or accessing the server shell.

- Not all the features are found in the platforms of cloud computing (Bidgoli 2018). Every cloud services are different from each other. There are some cloud providers that tend to provide limited versions and there are some cloud providers that offer only the popular features so the user does not get every customization or feature according to the need.

- Another disadvantage of cloud computing platforms over high-performance computing is that all the information and data are being handed over by the user while moving the services to the cloud. For all the companies that have an in-house IT staff, they are not able to handle the issues on their own (Namasudra 2021).

- Downtime is one of the disadvantages of cloud-based service. There are some experts that have considered downtime as the biggest cons of cloud computing. It is very well known that cloud computing is based on the internet and so there is always the chance of service outrage due to unfortunate reasons (Kumar 2018).

- All the components that remain online in cloud computing that exposes potential vulnerabilities. There are so many companies that have suffered from severe data attacks and security breaches.

The above sections deal with the analyses of other commercial factors over traditional higher performance computing.

Cost Analysis

It is very important to know how the cloud provider sets the prices of the services. A team of cost analytics has been referred by the company so that the team can calculate the total cost that the company has to endure to set the cloud-based platform. The team will decide which cost is to be taken into consideration and which is to be eliminated while calculating the cost. In the cost analysis site, HPC will need dedicated IT support. Network, compute, and storage are the three different cost centers of the cloud environment. Below are the points that show the cost analysis of the cloud service and an idea has been provided about how the cloud providers decide how much to charge from the user.

Network- While setting the price of the service, the expenses to maintain the network are determined by the cloud providers. The expenses of maintaining the network include calculation of the costs for maintaining network infrastructure, the cost of network hardware, and the labor cost. All these costs are summed up by the provider and the result is divided by the rack units needed by the business for the Infrastructure as a Service cloud (Netto et al. 2018).

- Maintainance of network infrastructure- the cost of security tools, like firewalls, patch panels, LAN switching, load balancers, uplinks, and routings are included in this. It covers all the infrastructure that helps the network to run smoothly.

- Cost of network hardware- every service provider needs to make its investment in some type of network hardware. The providers buy hardware and charge the depreciation cost over the device lifecycle

- Labor cost- labor cost includes the cost of maintaining, monitoring, and managing the troubleshoot cloud computing infrastructure (Zhou et al. 2017).
Compute- every business enterprises have their own requirement that includes CPU. Cost of CPU is calculated by the service providers by making an determination of the cost of per GB RAM endured by the company.

- Hardware acquisition- hardware acquisition computation stated the cost of acquiring hardware for every GB of RAM that the user will use. Depreciation of the cost is also made here over the lifecycle of the hardware.

- Hardware operation- total cost of the RAM is considered by the provider in the public cloud and then the cost per rack unit of the hardware divide it. This cost includes subscription costs based on usage and licensing.

Storage- storage costs are as same as the compute cost. In this, the service provider finds out what is the cost of operating the storage hardware and to get the new hardware as per the storage need of the users (Thoman et al. 2018).

Recommendations

As the company is looking forward to investing in the new system for running parallel codes, some of the recommendations to the company are mentioned in the below section. These recommendations will help the company to decide in which of the system, it needs to make its investment so that the company can run parallel codes smoothly. On the basis of analyses done in the above part of the report, it can be said that the company should go with the different platforms available in cloud computing. In the traditional HPC, there are lots of barriers to the users like high cost, the problem of moving and storing the data, and many more. It is not like cloud computing is all set for the company to use and so the following are the recommendations for the company that can enhance all the other platforms of cloud computing:

Recommendations to minimize planned downtime in the cloud environment:

- The company should design the new system's device services with disaster recovery and high availability. A disaster implemented plan should be defined and implemented in line with the objectives of the company that provides the recovery points objectives and lowest possible recovery time.

- The company should leverage the different zones that are provided by the cloud vendors. To get the service of high tolerance, it is recommended to the company consider different region deployments attached to automated failover in order to ensure the continuation of the business.

- Dedicated connectivity should be implemented like AWS Direct Connect, Partner Interconnect, Google Cloud’s Dedicated Interconnect as they provide the dedicated connection of network between the user and the point of cloud service. This will help the business to exposure the risk of business interruption from the public internet.

Recommendations to the company with respect to the security of data:

- The company is recommended to understand the shared responsibility model of cloud providers. Security should be implemented in every step of the deployment. The company should know who holds access to each resource and data of the company and is required to limit the access to those who are least privileged.

- It is recommended to the company implement multiple authentications for all the accounts that provide the access to systems and sensitive data. The company should turn on every possible encryption. A risk-based strategy should be adopted by the company that secured all the assets assisted in the cloud and extended the security to the devices.

Recommendations to the company to reduce the cost of the new cloud-based system:

- The company should get ensured about the presence of options of UP and DOWN.

- If the usage of the company is minimum, then it should take the advantage of pre-pay and reserved instances. The company can also automate the process to stop and start the instances so that it can save money when the system is not in use. To track cloud spending, the company can also create an alert.
Recommendations to the company to maintain flexibility in the system:

- The company should consider a cloud provider so that it can take help for implementing, supporting, and running the cloud services. It is also necessary to understand the responsibilities of the vendor in the shared responsibility model in order to reduce the chance of error and omission.

- The company must understand the SLA as it concerns the services and infrastructure that the company is going to use and before developing the new system the company should understand all the impacts that will fall on the existing customers.

Following all the above-mentioned recommendations, the company can decide how to and where to invest in the development of the new system.

Conclusion

The report was prepared with the goal to help the company to decide on which system should it make the investment in. The company has two options, either it can use a cloud computing platform or it can purchase all the equipment of HPC. In this report, the meaning of cloud computing has been explained very well. Not only the meaning of cloud computing, but it has also focused on the benefits of cloud computing in any business organization. In simple term, the delivery of a product or service with the use of the internet is called cloud computing. In the report, an analysis has been made regarding both of the systems. The advantages and disadvantages of the other platforms over the platform of high-performance computing are included in the analysis so that it can be easy to decide where the company should make its investment. For the comparison, different bases are taken as the cost of the system, security of data, control over the access, and many more. A structure of cost analysis has also been presented in the report so that company can imagine how much cost it has to invest in the new system. some of the recommendations to the company are also given. The company is recommended to choose the cloud computing platform as it is very secure and the cost of setting up the system is lower in comparison to others.

References