Tuesday, July 24, 2018

A Complex Enterprise Change: AWS Cloud Transformation Management




“When it comes to cloud adoption, the biggest challenge isn't technology - it's the people and processes that must change and adapt” - Forbes
The impact of cloud adoption will affect not only your application workloads but also your organization, business models, HR and procurement processes. Organizations need to examine the influence cloud adoption has on its people, culture, and processes. Culture can be defined as the values, beliefs, and practices that exist in an organization. Culture is critical when it comes to cloud adoption because it will affect an organization’s management and the economics of its IT services.
“Cultural issues are at the root of many failed business transformations, yet most organizations do not assign explicit responsibility for culture” - Gartner
The first step in influencing cultural change is to identify the values required for the organization’s cloud adoption. Once these values are defined, it is crucial to communicate the new operating behaviors and reinforce them explicitly and implicitly by using a reward system. The last step then becomes hiring or selecting the right individuals that are excited to embrace the change and would align the values that you want to drive.
The obstacles that organizations might face as they embark on the journey to the cloud are:
●      Lack of visible and active leadership
●      Lack of change management resourcing
●      Resistance from project teams, middle-level management and fellow employees resulting from the organization’s inertia to change.
Many IT organizations are not used to driving projects with change management in mind. Organizational change management can be defined as applying tools & processes to the people side of change, to transition from a current state to a future state to successfully achieve a particular outcome.
Successful change across an enterprise does not just happen organically. Organizational change first requires individual changes and its normal to experience some reluctance and resistance to change across an organization. The scope and type of change should inform your change management plan, which comes down to selecting the appropriate toolsets for the job at hand.
There are three types of change efforts, namely:
  1. Improvement: Focuses in on the changing toolsets.
  2. Transitional: This change effort focuses mainly on changing both the toolsets and skillsets.
  3. Transformational: The key difference here is that we are looking at the impact and effect changing toolsets & skillsets has on the changing mindsets.
Across the three change efforts, communication, change & project management, training & change leadership are all required solution drivers. Elements such as training & orientation, organization design, HR strategy & support are necessary elements required for a more robust transformational change.
There are 3 phases to tactically applying change management approach to accelerate your cloud transformation.
In the first phase, your team and senior leadership have to be mobilized. The team to lead the change and build the momentum is formed and referred to as the Cloud Center of Excellence (CCoE). The CCoE can consist of a cloud engineering team and a cloud business office team. The program governance structure is shaped, and it is essential to assess and align the change leadership roles.
The second phase is about defining your vision and engaging your organization. The third and final phase is about implementing your changes. After the selection process, the leaders need to articulate and communicate the vision & roadmap for transitioning into the cloud. At this phase, it is essential for leaders to address how the change would affect the employees in the organizations.
Finally, the last phase is about enabling the capacity and ensuring successful cloud transition into the cloud. The objective of this phase is to align IT organization structure, roles, and processes with the AWS platform and too also ensure that cloud benefits and goals are achieved.  

Friday, July 20, 2018

10 Tips for Managing Cloud Costs

1. Monitor spending on a daily basis.
2. Shut down unused or unnecessary instances.
3. Require tags company-wide.
4. Rely on automation rather than manual processes.
5. Consider a standalone cloud cost management tool.
6. Invest in a hybrid cloud management tool if your needs are more complex.
7. Look for a solution with machine learning capabilities.
8. Take a systematic approach to cloud cost management.
9. Watch out for vendor lock-in.
10. Optimize private cloud costs.

Evaluation of Cloud Cost Optimization Tools

We’ve looked at a number of cloud cost optimization tools in house and for customers.  This is a brief breakdown of some of the characteristics and features of the programs.  Please realize that this is a very fast moving part of the industry.  If you decision hinges on a single feature, it’s availability may have changed by the time you read this sentence.
package
direct cloud support
import data support
deployment method
cost aggregation
shared access
RI calcs.
cap. planning
general monitoring
project breakdown
price basis
AWS, GCE
no, on roadmap for enterprise
SaaS
yes
 
yes
yes- premium product
cost monitoring only
AWS and GCE tags
free, premium $240 up to $200k spend
AWS, Google, Azure
yes, using Dropbox
SaaS, AWS install
yes
yes, via email
yes
no
cost report/daily email
yes,can split cost as well as gather
free for the basic cost monitor with a tiered pricing for more functions.
AWS, Rackspac, Azure, more...
 
SaaS
yes
 
yes
yes
 
organization, application, environment (production, test, development)
free basic tier
EC2, CloudStack, Eucalyptus, HP, Logicworks, OpenStack, RackSpace, Terremark, vCloud Director
no
Saas, on-premise
 
yes
 
yes
yes
  
AWS/EC2
 
SaaS
yes
yes
yes
yes
no
yes by tag
three tiers, free/pro/ent - pro based on cloud spend tracked
AWS, Heroku, VMWare private clouds
VMware Vsphere, Citrix XenServer, Microsoft HyperV or Private Cloud platforms like vCloud Director, OpenStack or CloudStack
SaaS
in enterprise
in enterprise
  
alerts
yes
scales on spend tracked
AWS, AWS GovCloud
no
SaaS
yes
yes
yes
yes
alerts based on thresholds, regualr scans of "best-practices"
yes
tiered fixed cost based on total spend tracked
Column key:
package - Name of the package.
URL - click the name of the package (link)
direct cloud support - The clouds that it can track costs on it’s own without any data munging by you.
import data support - Can it import records from outside sources, such as munged data or private clouds which might not be directly accessible.
deployment method - Where does the code run?
cost aggregation - Does it aggregate basic costs?
shared access - Does it support more than one user logging in? DOes it support reporting to more than one person?
RI calculations - Does it do modeling to determine where Reserved Instances would save money? (RIs are pre-purchased computing available in AWS).
cap. planning
general monitoring - Does the product do any type of alerting other than cost based? (on example: unused but paid for storage)
project breakdown - Can the product do cost breakdown by projects or groups of resources?
price basis - What is the basic concept by which they charge?