Assessment Of Technical Management And SLA Requirements For OSDS Cloud Deployment

Remote Administration

The Department of Spatial Information (DSI) has considered your assessment of the deployment model, risk management and security issues for the Online Spatial Delivery System (OSDS). They have decided that they need an additional assessment on the technical management and the SLA.

You have been assigned the task of providing DSI Executive Management with an assessment of the management requirements and the provisions of the SLA for the chosen cloud vendor. You are to:

1. Discuss the requirements for remote administration, resource management and SLA management. It may be useful to consider Morad and Dalbhanjan’s operational checklists for DSI’s OSDS. This section should be no more than two to three pages in length.

2. Discuss briefly how you will consider application resilience, backup and disaster recovery for your chosen provider in relation to OSDS. This section should be no more than two to three pages in length.

3. Use Erl’s SLA guidelines to assess the SLA for your chosen provider. This section should be no more than two to three pages in length.

4. Provide a covering one to two page executive summary of these two assessments to DSI Executive Management and summarise the major opportunities and risks that have been identified in your assessments.
 

Online Spatial Delivery System (OSDS) needs to be moved to a cloud based platform for all of the services that are covered by it. Platform as a Service (PaaS) is the model that has been recommended for the task and it is a model that provides a platform for the building of and accessing of the applications ("What is PaaS?", 2016). Remote administration, resource management, SLA management, application resilience, backup and disaster recovery are the components that are involved with the OSDS and its migration on the cloud. Each of these entities play a significant role in the successful implementation of cloud computing to OSDS.

Remote Administration is the set of processes which allow the administrator tasks possible from a remote location (cbigconsulting.com, 2016). The application of the same to OSDS is not encouraged due to certain drawbacks that are associated with it.

Scalability of the tools is the first issue that is seen. Most of the tools do not scale properly and the management of the same becomes a very complicated process for the remote administrator.

Organizations make use of various operating systems to allow the access and development of the services and applications. These operating systems are spread across the computers as well as those specific to the mobile devices. For example, Windows, Linux, Blackberry, iOS, Android and Symbian are some of the operating systems that the organizations work upon. The management of all of these and the multi-platform support is a tedious task for the remote administrator.

Resource Management

Issues such as encryption, authentication, access control and authorized access are also commonly found in case of the structures that have remote administration implemented (sans.org, 2016).

The key players in any of the project are its resources, Success or failure of the project largely depends upon these assets and the management of the same is thus an important and a very significant task. Some of the major resources that are associated with OSDS are as listed below.

• Business Owner
• Top Level Management – Project Owner, Database admin, Network admin, Operations head
• Development and design team
• Testing team
• IT Support
• Networking team
• Communications team

Some of the basic questions are listed below which need to have their respective answers before the implementation and migration process begins.

• What is the resource tagging strategy that is deveoped for OSDS?
• What is the structure and levels involved with the internal management team?
• How are the resources selected and allocated for different projects?
• How are the resources handled in case of early delivery or delay in delivery?
• Is there sufficient metadata and service specific instances for the resources? (s3.amazonaws.com, 2016)

There are a number of advantages for resource management as:

• Centralized repository can be maintained to keep all of the information around the resources at one signle place which would make it easier to manage and access ("Project Resource Management Benefits | Innotas", 2016)
• Skills of the employees can be assessed in a much better way
• There are no hidden policies between the employee and the employer regarding resource allocation
• Time sheet management and change management becomes handy (Singh, 2013)
• There is an equilibrium that is created between the employees, employer, HR departments and the company goals ("Human Resource Management (HRM) - duties, benefits", 2016).
• Better understanding of the employees and their skills would result in better and improved resource utilization leading to better efficiency and productivity. 

SLA stands for Service Level Agreement that is an agreement or a contract that is signed betwwen the parties involved that is the service provider and the end user.For OSDS, SLA is signed between the cloud service provider and DSI.

OSDS is based on cloud computing which makes use of shared resources. The primary level in this case is Service based SLA (Bose et al., 2011).

In case of OSDS, SLA Management would be based upon the following factors:

• The negotiated and agreed-upon service design betwwen DSI and the cloud provider
• Supply chain management team
• The duties that are assigned to the team and the members involved with the project
• Services that are provided by the internal partners, if applicable
• All the releavnt details around the service such as the description, cost and features
• High level process flow chart covering critical success factors and processes
• Performance indicators (mitsm.de, 2016)

In addition to this, a checklist can be prepared which can be used as a tool for all assessment pruposes i.e. measureing the sepcific cloud solution which is to be deployed.  For the given scenario, Amazon Web Services can be used as cloud cpluting service for DSI. The checklist is as follows:

• Credentials of a specific user are provided by AWS Identity and Access Management (IAM).
• After backup and recovery options are understood, Amazon Elastic Block store makes use of an appropriate storage type.
• Dynamic IP addressing will be used by AWS when Application components will start.
• For using Operating Systems and to fulfill database requirements, Amazon EBS volumes are used.
• Back up of Amazon Elastic Compute Cloud will be done in particular intervals with the help of EBS snapshots.
• Amazon Machine Images are used to test the Recovery Process of Amazon EC2 instances in specified intervals.
• Deployment of Critical components of an application is done over zones with the help of Elastic Load Balancing.
• Processes for patches, security and updates for OS, AMI services and applications are tested time to time.
• Access credentials are used by operating system and no sharing is allowed for private keys of AWS instance.
• According to the topology of hierarchy, the rules of security are applied for a group.
• Mapping of DNS with Elastic Load Balancing is used by CNAME records.
• Embedded key pairs and the authorized key files which are part of Confidential and sensitive information are deleted.
• Before the AWS-hosted application is launched, it is good to test its performance.
• AWS trusted advisor reports should be incorporated.
• AWS accounts for production are signed for enterprise support.

There are a certain High Availability (HA) requirements and specifications that are associated with OSDS. These requirements can be met by DSI with the help of a lot many steps which include use of multiple availability zones in a specific region, balancing of the load, automatic scaling of tools and applications, timely reviewing and data recovery. It is required to ensure that all the critical areas and single points of failures should be carefully assessed. The strategy would include the following:

• Design and deployment of an application around fault tolerance in cloud that would assure complete reliability and accessibility of information at all times
• Incorporation of the best practices and guidelines
• Proper and accurate data storage options
• Cloud architecture

There is a huge data the OSDS would need to handle on a per day basis. There are tools that are available to correctly manage and store this data. Two of such Big Data tools are as described below:

• Hadoop

It is an open-source tool that has been written in JAVA lnaguage. It is tool that is cost-effective in nature and also has the ability to handle thousands of terabytes of data with extreme ease. It is based on distributed file systems which makes it fast and flexible in nature. Fault tolerance is another feature that comes handy with Hadoop (ITProPortal, 2013).

• Hyperscale Storage Architecture

SLA Management

Direct-Attached Storage (DAS) is used by this architecture for handling the data. If a component suffers a breakdown the workload fails over to another node and the entire unit is replaced rather than just the component within. Hyperscale is used to easily store and manage the Big Data across several different environments (ComputerWeekly, 2016).  

There are plenty of risks related to the services of cloud and proper measures should be taken so that it may not affect and bringing back the services.

Following are the measures need to be adapted for avoiding the risks and recovering from the risks:

• Preventative – Preventative measures are the important measures of security that should be there in every organization. For e.g., antivirus should be used, firewalls must be set up, and physical measures like entry points having guard must be used.
•  Reactive – These measures are used after the event gets completed.
•  Detective – These are the measures which contain measures like intrusion detection and motion detection.
• Administrative – Administrative measures are the measures which have a specific policy or a set of guidelines related to the strategy of business.

Disaster Recovery used in Cloud Computing is a technique that performs the backing up of data and recovery which is emphasized on the copies of data stored on the cloud. ("What is cloud disaster recovery (cloud DR)? - Definition from WhatIs.com", 2016). With the help of this technique, recovery process is enabled in several locations (Kiblin, 2011). Various scenarios and techniques have been made so that efficient DR method is made. (aws.amazon.com, 2016). Disaster Recovery for the Cloud is a good option for DSI in the aspect of both finance and control. (Dix, 2016).

Conclusions – Opportunities and Risks

For any of the cloud computing service, the security of its data and information is the most critical area. The major risks that are associated with cloud revolve around these two categories only. OSDS is not different and suffers from potential risks such as unauthorized access of data, loss of important informtion, loss of confidentiality and integrity of the data, denial of services, malicious software attacks and many more ("Risks of cloud computing | Queensland Government", 2016).

OSDS is based upon the PaaS model of cloud computing for the migration of the web suite of services and applications on the cloud. Remote adminstration, resource management, SLA management, application resilience, data storage and back up along with the disaster recovery are the components in cloud computing that have a significant role in DSI. Remote administration comes with a certain set of drawbacks and is a key area that needs to be frequently monitored if implemented, Resource and asset management are required to uytilize the skills of each and every resource fully and allocate them to correct roles and responsibilities. SLA management is basically done at a service level for DSI and is otherwise expanded across three different levels. High availability requirements and their execution, data storage and recovery can be well handled by the cloud.  

References

Bose, S., Pasala, A., Ramanujam A, D., Murthy, S. and Malaiyandisamy, G. (2011). SLA Management in Cloud Computing: A Service Provider's Perspective. Cloud Computing, pp.413-436.

aws.amazon.com, (2016).

Business.qld.gov.au. (2016). Risks of cloud computing | Queensland Government.

cbigconsulting.com, (2016).

ComputerWeekly. (2016). Big data storage choices. 

Dix, J. (2016). Cloud computing causing rethinking of disaster recovery. [online] Network World. 

Innotas. (2016). Project Resource Management Benefits | Innotas. 

ITProPortal. (2013). Big data: 5 major advantages of Hadoop | ITProPortal.com.