Over the past several weeks we have been discussing some of the mistakes made in early cloud deployments. As a refresher, here are the issues we outlined:
There is a common assumption that leveraging a Cloud operating model will save money. We won't debate the accuracy of this statement (tune into our next installment for that discussion!). For the purposes of this discussion, we'll assume it's true. But if that's the only benefit who, besides your CFO, really cares? Everyone wants to save money, but if you told a critical business customer that you were going to adopt a new computing model that would save money but introduce risk, do you think they would be supportive? Probably not: they wouldn't find the risk/reward ratio
acceptable. In order to shift that ratio, you need to improve the quality of the service(s) offered; otherwise you've simply lowered the cost of an increasingly commoditized service.
Even today, many enterprises rely on their end users to notify them of service outages or slowdowns (sometimes referred to as brownouts). When they finally become aware of a critical issue, the typical diagnostic approach entails getting the individuals responsible for each technology silo to circle up on a conference bridge until the problem is isolated and resolved. The Cloud makes this remediation process difficult to execute: good luck getting your Cloud provider on that bridge! Cloud makes the underlying infrastructure supporting an application/service fluid, rendering traditional resource-based monitoring tools useless in a diagnostic process.
Original Article - https://cloudcomputing.sys-con.com/node/1475427
- Not understanding the business value
- Assuming server virtualization is enough
- Not understanding service dependencies
- Leveraging traditional monitoring
- Not understanding internal/external costs
There is a common assumption that leveraging a Cloud operating model will save money. We won't debate the accuracy of this statement (tune into our next installment for that discussion!). For the purposes of this discussion, we'll assume it's true. But if that's the only benefit who, besides your CFO, really cares? Everyone wants to save money, but if you told a critical business customer that you were going to adopt a new computing model that would save money but introduce risk, do you think they would be supportive? Probably not: they wouldn't find the risk/reward ratio
acceptable. In order to shift that ratio, you need to improve the quality of the service(s) offered; otherwise you've simply lowered the cost of an increasingly commoditized service.Even today, many enterprises rely on their end users to notify them of service outages or slowdowns (sometimes referred to as brownouts). When they finally become aware of a critical issue, the typical diagnostic approach entails getting the individuals responsible for each technology silo to circle up on a conference bridge until the problem is isolated and resolved. The Cloud makes this remediation process difficult to execute: good luck getting your Cloud provider on that bridge! Cloud makes the underlying infrastructure supporting an application/service fluid, rendering traditional resource-based monitoring tools useless in a diagnostic process.
Original Article - https://cloudcomputing.sys-con.com/node/1475427
Join Us: http://bit.ly/joincloud
Posts
1 comment:
Great bblog post
Post a Comment