Internal documentation only

This page has been marked as a draft.

CCO topic 2

Overview Copied

Review this topic/section before publish!

Capacity Planner Cloud Cost Optimisation is a set of features built on top of Capacity Planner to provide insight into cloud use, cloud costs and recommend where savings can be made. It provides options for users to compare clouds and manage a multi-cloud environments.

Cloud costs are made up of several elements:

With Capacity Planner Cloud Cost Optimisation, you can save on cloud costs by right-sizing, right-buying and optimizing.

Capacity Planner Cloud Cost Optimisation supports hybrid IT environments but cloud environments are represented in dedicated models. There is no facility to mix an on-premise environment and cloud environment within the same single model due to the significant difference in hierarchy.

Note that Capacity Planner Cloud Cost Optimisation is not a Cloud Management Platform (CMP). It does not provide management and fulfilment of service requests nor does it orchestrate change. However, the intention is for ITRS to build integration with best in class tooling that addresses this gap, taking an application deployment policy approach rather than ad-hoc real-time right-sizing.

We need to help customers run their clouds economically – but also improve to make use of the cloud effectively​. From ‘always on’ to ‘on when needed’. From ‘hard scaling’ by creating new servers to dynamic scaling, horizontally and vertically​. Understand the transactional cost of running cloud applications ​. Strike the balance between economical and effective cloud ops.

Working with granular data over long periods of time to build statistical models of utilisation allows us to​ understand not just averages and peaks, but numerous measures and behavioural patterns​. Provide a single pane of glass across multiple estates, consistent capacity reporting and predictions​.

With Capacity Planner, Rightsize, Rightbuy and Optimize your Cloud Spend.

How do we compete at this point? Copied

We have some key differentiators at this point:

The right tool for the present Copied

Cloud Cost Optimisation (CCO), a new feature available within ITRS Capacity Planner, is centred around these two processes. Based on the right-sizing and right-buying processes, the CCO solution is able to optimise cloud spend by recommending improvements to a production estate and identifying wasted capacity, and therefore money.Most importantly, ITRS Capacity Planner’s new CCO solution backs up its recommendations by presenting the data in an easy-to-visualise way. In this way, CCO is able to prove why it makes the recommendations it does, giving firms the peace of mind that they are making the best choices for their cloud estates. For example, in the cloud migration process, CCO prices up the ‘like-for-like’ migration and presents a clear comparison with the right-sized estate to demonstrate potential savings to be made.

The right tool for the future Copied

Once a firm understands and has optimised its existing environment, it then needs to be able to predict its trajectory and what demands will need to be met in the near and long-term future. ITRS Capacity Planner’s CCO solution facilitates this via application demand modelling, which determines the relationship between application volumes and resource utilisation. In this way, it can identify capacity bottlenecks and predict future platform requirements. In tandem, ‘Forward Thinking’ scenario modelling allows firms to model the impact of this demand across services. From there, they can address resource constraints and determine the end-to-end cost of servicing growth.

In this way, CCO allows businesses to identify the optimum configuration of their various applications to support not just existing but projected demand and the cost of servicing it.

To enable ongoing adaptation and optimisation as the enterprise changes or scales up, CCO is also able to integrate with a variety of service management platforms that control or automate workflows. For example, ServiceNow; if Capacity Planner suggests a change to an environment, that would go into ServiceNow as a recommendation and raise a change ticket before entering the workflow. By feeding recommendations into service management and policy-based orchestration platforms, CCO ensures that application configurations are always optimised.

More information on the Recommend panel, see /baseline-view/recommend.htm

What is right-sizing Copied

In order to accurately right-size, firms must first identify the sizes of the instances needed for each application workload.

They can do this by analysing which machines can be downsized from core count, memory and allocated storage, and which can be switched off altogether. This requires statistical analysis of highly granular resource utilization data. Using simple averages will significantly underplay resource requirements, while using peak without an understanding of how often those peaks occur and for how long can significantly overplay resource requirements. They must also understand the capacity of the on-premise estate if their Hybrid-IT allows them to bring work off-cloud if necessary. Applications or services in the Cloud can also be right-sized based on their actual use of resources.

It is about taking data and doing analytics, not aggregating data, no averages as recommendation will not be useful. Also if you do it only based on peaks, then your recommendation will not behelpful if peak occurs only occiasionally. So we are creating recommendation based on analytics and statistical proile that is built.

What is right-buying Copied

The second central tenet is right-buying. This involves accurately analysing and optimising the type of instance (container, VM or ‘bare metal), the duration the instance exists for, the location and region the instance is created in, the type of purchase (on demand or reserved capacity), and the best way to buy it. On top of this is an ongoing analysis of the billing engines of the cloud providers to identify optimal usage and policies. By undertaking a comparison of multiple cloud providers and the on-premise estate, firms will be able to select the best purchase plan for optimum price based on what their use profile actually looks like.

A combination of right-sizing and right-buying is the only way to ensure workflow demands are exactly catered for, taking peak demand into account with no costly excess headroom. These are not just one-off processes to be implemented at the cloud migration stage, but ongoing undertakings as an enterprise gets settled and expands within the world of Cloud. The process of right-sizing and right-buying one’s estate is a continuous one that should last a business’ entire cloud lifespan.

what machine to buy, what instance. eg in which region

how to best run your instance / machine, for how long, when.

What is optimizing Copied

continually working on analytics, always learning and builfing new statistical profiles, udnerstanding the demand.

Features Copied

The most prominent features of Capacity Planner are:

Getting started Copied

Getting started Copied

Cloud governance has become top priority

managing capacity in the cloud has started to become an issue

Instances in the cloud 0 they represent your vm or server in a non cloud environment Each server has a cost associated with it, it also has a size - how much work you can do on that instance. If you increase the size, the cost also increases. Usually you pay for them on an hourly basis, but there are other ways to pay. Whenever you run something in the cloud, you’re paying for it. Regardless of whether you use it or not. For example it is more expensive to run instances in some regions than others, Also you can reserve capacity for a longer period of time which is cheaper, amazon will encourage you to do it. But maybe this is not what you should be doing. Perhaps there are other more optimal ways to do it. There are also different services providers provide which makes things more complicated. a service is for example EC2. Or Lamdba (pass a piece of code, you get charged depending on how much time it takes to run that code).

In CCO we’re looking to save on costs by right sizing right buyng and optimizing.

helping customers use cloud effectively and now also economically. strike the balnce between the economical use of the cloud and effective use of the cloud.

Capacity Planner analytics apply to cloud instances just as they do to physical servers or on-prem virtual machines: long and short term projections, activity reporting, application demand modelling.

["Capacity Planner"] ["User Guide"]

Was this topic helpful?