The Private Cloud blog is all about community and we’re happy to say that we have a number of people from the community stepping up to the plate to share their knowledge about private cloud here. This week we are treated to a great post by Vincent Montalbano on the importance of the Operations Layer in private cloud architecture and design.
The Operations Layer defines the operational processes and procedures necessary to deliver Information Technology (IT) as a Service. This layer leverages IT Service Management concepts that can be found in prevailing best practices such as ITIL and MOF. The main focus of the Operations Layer is to execute the business requirements defined at the Service Delivery Layer. Cloud-like service attributes cannot be achieved through technology alone and require a high level of IT Service Management maturity.
Vincent does a great job at boiling down the key concepts you need to understand when working with the private cloud operations layer. After you read Vincent’s post, make sure to read the Private Cloud Planning Guide for Operations.
Enjoy! – Tom Shinder, the Private Cloud Guy.
The Journey to the Private Cloud Part 1 – The Operations Layer
by Vincent Montalbano
There has been a subject that has dominated our daily discussions for several months now, “Private Cloud”. We see the cloud starring in commercials on TV and in advertisements on the Internet. We watch webinars and read articles about the cloud and all of the really cool things the cloud can do and then the question that we keep asking ourselves is, “How do we get the cloud in our datacenter”?
What is the “Cloud”
Cloud computing must have the following 5 “characteristics” in your environment. I have expanded these with to include some very high level examples.
- On Demand Service– The ability to provision a service or resource to an approved user without human interaction.
- Broad Network Access– The ability of an authenticated user to access the necessary resources via computer, tablet or smartphone from anywhere.
- Resource Pooling– The combining of server, storage and network resources into one fabric for all services in the infrastructure to share.
- Rapid Elasticity – The ability quickly to add or subtract services or resources.
- Measured Service– The ability to bill for the services or resources used by the customer.
These five characteristics have been defined by the United States National Institute of Standards and Technology (NIST). The ability to provide these elements of the cloud requires an extensive review of your current IT infrastructure and the proper planning to create the roadmap that will take you to the private cloud. To initially get to the cloud you have to plan on implementing each required “element”.
Migration from the Traditional Datacenter Model to the Private Cloud
This article is written from the perspective that your company has a traditional IT datacenter and you want to move to the private cloud model. So where do you start? The starting point is to plan around the Private Cloud Reference Model Layers. These are the 6 reference layers;
- Service Delivery
- Software Layer
- Platform Layer
- Infrastructure Layer
- Management Layer
- Operations Layer
I want to start the planning discussion around layers 6 through 4. I am an Infrastructure consultant by trade and I fully believe that like any structure that is being built you must start with your foundation, (it will be hard to deliver layers 3 through 1 without the bottom 6 through 4). In this case the layers 6 through 4 involve your current IT datacenter and your operations team, (operations layer) your physical networking and server infrastructure, (Infrastructure layer) and the management of those 2 layers, (Management Layer).
Part 1 of my “journey to the cloud” series is to examine the Operations layer, (and by examining I am not suggesting re-orging, downsizing or outsourcing). I am suggesting that you analyze how your current model functions and performs its day-to-day deliveries and administration tasks.
According to Microsoft, the Operations Layer “Represents Service Management and operational processes carried out by IT operations and support staff”. The concept of successful cloud computing has the cloud reference model layers relying on each other. The cloud is a layered management model that is similar to the TCP/IP stack. The layers above and below each other are dependent on one another to deliver services to the dependent layer. The Operations layer provides the foundation for your entire IT infrastructure and is directly tied the Management layer. These 2 layers work very closely together and are very dependent on each other.
The Operations layer is your current datacenter and the staff that carries out the day-to-day tasks that keep the business in business, (IT Operations Admins, Help Desk Operators, physical security). These folks do all the monitoring, alerting, physical datacenter security, hardware maintenance, HVAC and request taking in your organization. These are the necessary functions required for this layer and you must examine your current Operations department and ask yourself do I have the following processes in place at the Operations layer:
- Do you have a Change Management (CM) approval process?
- Is the CM process approved and tracked via an application?
- Are CM owners approving planned changes?
- Is the business, (or necessary departments) being notified of the upcoming CM via the CM application?
- How to you catalog your physical computer and networking device assets?
- How do you manage your physical computers and networking device warranties?
- How is your physical asset refresh cycle managed?
- Have you gathered and stored your organizational information?
- Is this information readily available to your IT staff when needed?
- When there is a service outage is there a process of escalation to resolve the issue?
- Are your incidents recorded and reported to upper management?
- When a cost center requests a service what is your solution to fulfill that request?
- What is your Service Level Agreement, (SLA) regarding provisioning new resources?
- Do you regulate access to your datacenter, and if so how do you determine who gets the privilege?
- Can we correct system or service disruptions in a timely manner that limits the service outage time and the impact to your business?
Most businesses have some or all of these processes in place. The CM process is IT 101 and your CM process should be reviewed and refined to tie into other characteristics of the private cloud such as automated responses and maintenance. The refinement of the Operations layer is vital to your successful journey to the cloud and after answering the above questions you may have to work on refining your Operations department. The NIST“5 Characteristics” of cloud computing are very dependent on the Operations layer and without the above Operations layer processes in place it will be next to impossible to deliver On Demand Service or have the ability to provide Measured Services to your customers.
The Operations layer keeps track of all of the upcoming Change Managements, maintenance windows and general housekeeping that will occur in a datacenter. The Operations layer supports the next layer, (Management layer) by communicating and addressing service issues either through corrective measures or escalated actions, coordinating requests for new or additional services and communicating these changes to IT as a whole.
You are probably saying to yourself that you already know this but you must re-think Operations somewhat because the private cloud incorporates these additional concepts through the Operations layer:
- Using a system management application to detect and provide automated responses to service failures and requests for new services
- Automating manual processes that are performed in CM or as part of day-to-day operations
- Using workflows to alert, escalate or perform a service or task
Part 2 of this series will discuss the Management Layer of the Private Cloud Reference Model and how the Management layer will use the Operations layer to further automation of datacenter tasks, and provide the ability to dynamically provide services to customers via On Demand requests or through the traditional ticketing system.
"The opinions represented in this post and don’t necessarily represent Microsoft’s or my employer’s positions, strategies or opinions”
Infrastructure Consultant Catapult Systems, Inc.
MCITP: Enterprise Administrator
MCITP: Windows Server 2008 R2 Virtualization
MCTS:System Center Configuration Manager, System Center Operations Manager