FinOps tools in the Cloud: An insufficient first line of defense

FinOps tools   in the Cloud: 

 

An insufficient first line of defense

Once seen as an economic Eldorado, the Cloud is now a source of anxiety for companies facing exploding costs.


Flexera's renowned annual " State of the Cloud " survey, based on 753 IT decision-makers across various types of organizations, reveals that cloud spending is the number one concern.
This is quite understandable given that the cloud market is projected to reach $1.2 trillion by 2028  Grand View Research ).
Some studies  indicate that nearly 50% of cloud spending is simply wasted! 

 

FinOps tools have emerged as a relevant response to this problem.

Thus, 49% of respondents to Flexera's survey indicated that they had adopted a FinOps solution to analyze and optimize cloud spending. This optimization relies primarily on streamlining cloud infrastructure and measurement. 


We believe it is possible to go much further to  solve this perpetual drift in a sustainable way.

 

 

The role of FinOps  tools  in management

Cloud costs  

Faced with the uncontrolled rise in Cloud bills, companies are adopting cost management tools to monitor their spending.

The solutions offered by hyperscalers are Azure Cost Management, Google Cloud Cost Management, and AWS Cloud Financial Management. They allow users to track cost trends and send alerts before costs become excessive.

 

However, 89% of respondents to Flexera's survey  are multi-cloud, which limits the scope of these solutions, unless you juggle between several tools continuously, which is not necessarily simple. 

And then there's a feeling that these hyperscalers are both judge and jury for having implemented billing systems perceived as opaque!

 

Third-party tools for multi-cloud enterprises offer a consolidated and more detailed view of cloud spending. Examples include Datadog and Apptio Cloudability.
These tools help identify unnecessary expenses related to oversized infrastructure and unused or underutilized services.

We can also mention Turbonomic, which automates the starting and stopping of applications based on demand, or Densify, which dynamically adjusts the size of Kubernetes clusters and VMware instances.

Why are these FinOps solutions insufficient?

 

Simply tracking expenses or optimizing infrastructure sizing, without addressing the root causes of cost inflation, is like putting a band-aid on a broken leg!
FinOps tools are a good first line of defense for controlling cloud costs.

However, they must be accompanied by an ambition for continuous optimization of the information stored, often wrongly, for many excellent reasons!  

This is what we offer with  {openAudit} .

 

 

Usage   and  traceability  to isolate unnecessary flows and sustainably reduce costs  !

Identify unused data through dynamic log analysis. 

 

Log analysis by  {openAudit}  allows continuous identification of data that is not consulted or used over a given period.

Identify the upstream "branches" of this unused data, using a technical data lineage.

 

{openAudit}  identifies the sources and transformations that generate unused data through technical and multi-technology data lineage,  up to a hypothetical branch that will be the source of a useful flow.

Link these "branches" to cost logs

 

{openAudit}  will link the identified processes to the cost logs exposed by the hyperscalers to assess their financial impact. This will also allow for prioritizing optimization actions by targeting the most expensive flows.  Storage and compute represent the largest associated costs in billing.

Measuring progress

 

Implementing a dashboard to track cost evolution will help to positively motivate the team responsible for the subject.

 

Commentaires

Posts les plus consultés de ce blog

Power BI libère les utilisateurs… Mais comment garder la maîtrise de sa plateforme dans le temps ?

De la source à la cellule du dashboard : Cartographier le SI pour le reconstruire intelligemment

Migrer de SAP BO vers GCP Looker - Garder ses données en source ? Possible ?