Cloud Computing For Data Management Services

Enterprise IT Solutions: to Purchase or Rent?

Discuss about the Cloud Computing for Data Management Services.

With the ever increasing demand for data management services, organizations strive for the best solutions to both manage their data seamlessly and securely.  To ensure the best service, organizations seek to have the latest equipment as IT equipment generally become redundant, or need replacement, especially for enterprise systems after every four years (‘Blue Data’, 2017).  In managing data, especially where many users and huge data amounts are involved, organizations have two choices; one is to use the traditional and widely used approach of purchasing the requisite equipment and managing them in-house or in their premises. Another more recent approach is the use of co-location services where an organization purchases the requisite equipment but then have them located in a different and shared physical location where it has complete control and access to all its data; however, the costs of power and cooling are shared, but the company pays for the use of the space (Puro, 2016). The third option, which is similar to co-location, is the use of cloud services. Cloud services are similar to co-location in how they work, except that in cloud services, the service provider offers all the hardware plus the necessary software and other applications and the customer pays per use. All data is managed by the cloud services provider on behalf of the customer (Mahmood, 2013).  All the three methods carry their respective merits and demerits; however, it is not easy to decide on the best without taking into consideration the requisite resource usage for each of the methods. This paper is going to make a comparison between an organizations’ purchasing all the equipment and managing its data and IT needs on site; the common and traditional method approach.

 This will be compared mainly with cloud computing because co-location is almost similar to cloud computing, but still entails purchasing equipment. However, a mention will be made on co-location during the discussions.  This paper begins by briefly defining and explaining the two main approaches to managing data, which is purchasing the IT equipment and infrastructure and cloud computing and discusses them at a theoretical level. The paper will then obtain various prices for the case scenario hardware and equipment, as well as the requisite software and other features; for instance air conditioning. The prices obtained will be based on standard or retail prices from vending sites such as New Egg, PC World, Radio Shark, Amazon, and E-Bay and then be used to compute the TCO (total ownership costs). The paper then selects a standard and widely used cloud services provider and compares the charges, based on similar data and computing needs as would cost if the equipment and IT facilities were bought and used on site (in house). The amounts are then compared over a four year period, which is when IT equipment for enterprise applications become obsolete and therefore require replacing. The paper then discusses the findings, compares them, while revisiting the respective merits and demerits of each, before making a decision and recommending the best solution that would suit the case scenario. The assumptions made are also discussed and justified, before a conclusion is drawn.

Costs

There are several benefits to running applications and storing data on an online platform; among them being flexibility, disaster recovery, automated software updates, reduced or no capital expenditure, the ability to work from anywhere, increased collaboration, security, and a low environmental foot print. Another commonly touted advantage of cloud computing is that it results in significant cost savings and enables an organization to focused on its main activity and leave the IT hassle to the specialists (Alexander, & more, 2017), (Froehlich, 2015). For on premise computing, there are also several benefits; the main one being total control over data and the It infrastructure; the business handles all its confidential information without the risk of this data being breached. Another benefit is that the organization depends less on the vendor such as having to be locked-in on various technology y decisions made by the vendor.  On premise solutions also allow for easy customization of data and IT infrastructure to suit the needs of the organization and can be a foundation for future expansi0on into the hybrid. Further, it limits on-going costs as compared to cloud computing where services are paid for regularly (Lindskog, Berglund, Vallhagen, Berlin, & Johansson, 2012).  While both approaches have their benefits, the more significant one is on costs and capital expenditure, since organizations aim for a better return on investment (ROI) for every investment they make on an IT enterprise system.  The cost is what this paper will focus upon (Plattner & Zeier ,2012)

This entails making equipment and software purchases and then setting it up and managing/ administering; considering the case scenario, the following equipment were sought from vendors and their costs obtained as shown in the table below;

Product	Description	Specifications	Unit Price	Quantity	Total
HPE ProLiant ML150 Gen9	Server (tower)	Intel® Xeon® E5-2600 v4 processor * Core	1439	3	4317
		8 GB RAM			0
		2.5 inch 50 GD SSD SFF hard disk	103	3	309
HPE ProLiant ML150 Gen9	Server (tower)	Intel® Xeon® E5-2600 v4 processor * Core	1199	6	7194
		8 GB RAM			0
		2.5 inch 100 GD SSD SFF hard disk	290	6	1740
HP ML350 G6 Tower Server	Quad CORE E5620 2.4Ghz 6GB RAM		799	4	3196
10 TB HDD	Seagate Enterprise Capacity	3.5 HDD 10TB, SATA 6Gb/s 7,200RPM 512N	419		0
Load balancing device	F5 BIG-IP Global Traffic Manager 1600	Big-ip Ltm Enterprise 1600 4gb	4000	2	8000
Server Software	Windows	Windows Server R 2 64 Bit	200	13	2600
Setup costs			1500	1	1500
Server administration costs			7089	1	7089
UPS	APC	SURT6000XLI, Smart-UPS, RT 6000VA, 230V, 4200 Watts	2999	13	38987
ISP		Internet service provider with dual connection	45	48	2160
Air conditioner	Tripplite	SmartRack 12,000 BTU 230V Portable Air Conditioning Unit	770	4	3080
Networking and cabling		Network cables for the servers	2500	1	2500
RAID controller		IT Mode LSI 9211-8i SAS SATA 8-port PCI-E 6Gb/s Controller Card	80	13	1040
Air conditioning		Air conditioning power consumption	3066	4	12264
Power consumption (Servers)		Servers power consumption	4030	4	16120
Software upgrade costs		Windows plus firewall and antivirus	600	4	2400
Antivirus + antimalware		Kasperski total security	12	52	624
Cisco Firewall		Cisco ASA5505-SEC-PLUS Firewall	165	3	495
Total Cost of Ownership over Four Years			115615

In making the computations, it is assumed that the servers will be running for 24 hours and that they can only be stopped for scheduled maintenance or during power outages.  It is also assumed that the premise IT system will require some form of continuity and data security so the organization will install RAID architecture for the Servers with RAID controller hardware. The power consumption rates are based on standard power consumption for Turkey/ or the EU (European Union). The power consumptions are based on the server power ratings and ignore internal power consumption by the UPS systems. The air conditioning servers will be three, one each for every set of servers (on a use basis) and each is capable of 12500 BTU and are rated at 3500 W; it is assumed that the cooling systems will on average run for 24 hours.  The server administration costs are assumed to be at the rate of 25% of the serve capital expenditure and that server maintenance costs will rise as the servers get older.  The various software, especially antivirus and server will be updated and upgraded yearly, with the upgrade costs estimated (anti-virus license must be renewed every year). It is also assumed that the power rates will not change or change significantly over the four year period.  Estimated networking costs are also included in the TCO computations. The computations assume that the IT staff are paid under general costs and so are not included in these computations.

Assumptions

For the cloud solution, the company will use various services, including Saas (software as a service), IaaS (infrastructure as a service), and PaaS (platform as a service). T hese are all charged in a single package, which includes management and backup charges; these were computed from the 1&1 cloud services calculator; note that the storage capacities offered are much higher than the requirements but have been used due to backup requirements. These were computed in a spreadsheet as shown below;

Server	Cores	RAM	HDD	Cost per month	Per Year	Servers	Total
Cloud server XL	4	8 GB	160	50	600	4	2400
Cloud server 3XL	8	8 GB	240	130	390	3	1170
Cloud server 5 XL	16	48 GB	500	255	3060	6	18360
Total per year	21930
Total Four Years	87720

There are no set up costs; the server hard disk space is much higher than what is specified in the case scenario because it is assumed the organization will want data backup through the cloud.

From the above computations and estimates, it is clear that on premise systems have a higher cost of ownership over the four year period when compared with cloud services. The bulk of the TCO for the on premise system is because of the upfront capital expenditures in purchasing the hardware and software. However, server administration costs are also significant, being estimated at being between 15 and 25 % of the total server and related capital expenditure, for which the maximum possible amount is used. The calculations show that the cloud option with a higher capacity than required in the case will cost $ 87720 while the cost for the on premise system will have a TCO of $ 115615. From the computations, the cloud services will cost just over 75% compared to the on premise solution, and still achieve very high capacity. This implies that the cloud solution will achieve a significant cost saving of 24% over the four year period.  Revisiting the advantages of both systems, it can be seen that the cloud service still is a better option and is therefore recommended as the best solution.

Conclusion and Recommendations

This paper evaluated the relative benefits of using an on remise system and cloud services for servers for a case scenario. The relative advantages of the two systems were reviewed and the TCO for both systems computed by using actual charges as of April 2017 for both the IT equipment and cloud services. The results show that the cloud option will result in significant savings of 24% over four years compared to the on premise solution and therefore, this paper recommends the cloud solution. As a future research recommendation, this paper proposes future research to investigate the TCO of a hybrid on premise system extended to the cloud.

References

Puro, N. (2016). Cloud, Colocation or Dedicated – Which one should you choose?. Esds.co.in.

Retrieved 21 April 2017, from Mahmood, Z. (2013). Cloud computing: Methods and practical approaches. London: Springer.

Plattner, H., & Zeier, A. (2012). In-memory data management: An inflection point for enterpise  applications.

data-private/d/d-id/1323089

Lindskog, E., Berglund, J., Vallhagen, J., Berlin, R., Johansson, B., & 2012 Winter Simulation Conference (WSC 2012). (December 01, 2012). Combining point cloud technologies with discrete event simulation. 1-10.

Froehlich, A. (2015). Cloud Vs. On-Premises: 6 Benefits Of Keeping Data Private –InformationWeek. Information Week. Retrieved 21 April 2017, from

‘Blue Data’. (2017). Product | Big Data Infrastructure Software | BlueData. BlueData. Retrieved  21 April 2017, from

Alexander, P., & more, R. (2017). Should You Lease or Buy Your Tech Equipment?. Entrepreneur. Retrieved 21 April 2017, from Advances in computing and information technology: Proceedings of the Second International

Conference on Advances in Computing and Information Technology (ACITY), July 13-15, 2012, Chennai, India. (2013). Berlin: Springer