High difficulty in predicting the right infrastructure for covering the business needs.
Normally, yearly budgets should provide enough information for correctly dimensioning the company’s infrastructure.
But upgrade in infrastructure is not done yearly but once at several years.
Making an accurate budget for several years is obviously complex in a big organization.
Another important aspect is related to peak periods that could hit yearly, monthly, weekly, even daily.
To cope with the increased demand in resources during peak periods, even more infrastructure should be deployed.
Therefore, the infrastructure should be dimensioned by adding, on top of the budget’s request, a certain contingency percent. This means over dimensioning.
What is even worse is the fact that this infrastructure is rarely used at its full capacity, this, actually happening only during peak periods.
Lately more and more providers changed the licensing model, which is now complex and expansive for outside cloud products.
Some products are not available at all outside the cloud.
Upgrade process for databases and applications for the in-house infrastructure are very risky.
All these do not exist in cloud.
Real life
There are companies who operate in a difficult to predict volume market. On top of this,they also have to cope with high seasonality.
We all noticed on the market a high increased in customers demand for quality and for sophisticated products.
We could say that the demand for quality in B2B segment became almost equal with B2C segments.
Big companies have million system’s interactions, of different kinds per 24 hours, which, are not evenly distributed over the day. Hours with very high demands of resources could be followed by hours with very low demand of resources.
It is hard to predict what infrastructure resource may be needed in all situations. What we know for sure is that our customers’ needs are high in terms of availability and reliability.
Therefore, automatic scalability, both up and down, for more resources and for resources saving as well, is a major factor in high seasonability businesses.
Strategic move
Decision to move to cloud architecture not with low important pieces of application but with important applications is made more and more often lately.
A PaaS (platform as a service) architecture got capability of handling very well peak issues by automatic scaling.
The product is finally mature on the market.
The process of moving important applications to cloud PaaS is complex and needs a very good requirements gathering, analyzed, fit with business and customers’ needs for a serious and responsible POC.
Then a pilot shouldn’t be missing, step two being extension of the pilot while the current infrastructure should be fully functional in case something goes wrong.
There are two main points that should also be carefully considered:
- Make automatic “scaling up and down” functioning according to the needs
- Deploy a fully redundant architecture, at the database and application level
Conclusions
Given the need for total flexibility and security, to be able to both scale up but also down to save resource, cloud PaaS could work very well to cover peak business’s situations.
High availability, scalable architecture automatically, low in house infrastructure, access to the latest versions without effort, this is what cloud PaaS is mainly providing.
There are companies who operate in a difficult to predict volume market. On top of this,they also have to cope with high seasonality.
We all noticed on the market a high increased in customers demand for quality and for sophisticated products.
We could say that the demand for quality in B2B segment became almost equal with B2C segments.
Big companies have million system’s interactions, of different kinds per 24 hours, which, are not evenly distributed over the day. Hours with very high demands of resources could be followed by hours with very low demand of resources.
It is hard to predict what infrastructure resource may be needed in all situations. What we know for sure is that our customers’ needs are high in terms of availability and reliability.
Therefore, automatic scalability, both up and down, for more resources and for resources saving as well, is a major factor in high seasonability businesses.
Strategic move
Decision to move to cloud architecture not with low important pieces of application but with important applications is made more and more often lately.
A PaaS (platform as a service) architecture got capability of handling very well peak issues by automatic scaling.
The product is finally mature on the market.
The process of moving important applications to cloud PaaS is complex and needs a very good requirements gathering, analyzed, fit with business and customers’ needs for a serious and responsible POC.
Then a pilot shouldn’t be missing, step two being extension of the pilot while the current infrastructure should be fully functional in case something goes wrong.
There are two main points that should also be carefully considered:
- Make automatic “scaling up and down” functioning according to the needs
- Deploy a fully redundant architecture, at the database and application level
Conclusions
Given the need for total flexibility and security, to be able to both scale up but also down to save resource, cloud PaaS could work very well to cover peak business’s situations.
High availability, scalable architecture automatically, low in house infrastructure, access to the latest versions without effort, this is what cloud PaaS is mainly providing.
Weekly Brief
× 
CIO Applications Europe Weekly Brief
Be first to read the latest tech news, Industry Leader's Insights, and CIO interviews of medium and large enterprises exclusively from CIO Applications Europe
