Recently, a friend of mine came up to me and asked “Can you succinctly define Federated Cloud Ecosystems?”. That lead to an instantaneous brainstorming between three of us and we came up with a few characteristics that defines Federated Cloud Ecosystems like NIST did for cloud computing. I tweaked the gist from our discussions to add a bit of my flavor to it and thought I will put it out in public for scrutiny. I strongly encourage you to chip in with your comments below or contact me directly. I would love to have the discussion going and standardize the definition.
First, some motivation
To me, handful of cloud providers serving the world’s computing needs is a shortsighted idea. The needs of the world are diverse and even the regulations are diverse. Handful of cloud service providers cannot satisfy these needs and requirements. Also, I would consider it as a dangerous idea to begin with. We saw the impact of monopoly in traditional software days and people in US are seeing the impact of small number of telecom providers controlling the wireless market. The idea of consolidation is for people who believe in the principle of economics of scarcity and, as someone who is from open source background, I see value in the economics of abundance. I see competition through abundant participation in the market, rather than consolidation, as the correct realization of capitalistic ideals. The abundance factor may continuously commoditize the products but it forces vendors to monetize on higher order systems (Check out Simon Wardley‘s thesis on this topic for more info), driving continuous innovation at a much rapid pace than the alternative thesis. Having made my economic beliefs clear, I see the evolution of federated cloud ecosystems as the right thing to happen in the cloud marketplace.
Now, the definition
Federated Cloud Ecosystems have the following characteristics:
- Multitude of players: This is the critical piece to the very definition of federated cloud ecosystems. This will ensure that the market vulnerability, monopoly, doesn’t develop in the system. One of the necessary conditions for ensuring such a participation by service providers is to have at least one open source cloud platform.
- Heterogeneity of cloud platforms: Another important characteristic is the presence of heterogeneous platforms in the ecosystem. A federated cloud ecosystem can be realized just with one cloud platform as long as the platform supports federation but it will just shift the monopoly from the service providers to the platform. Monopoly at any layer is bad for the users and this characteristic ensures that we avoid monopoly at the platform layer. I don’t support monopoly even if the underlying platform is open source. The idea of heterogeneity not only helps avoid any monopoly, it is critical to handle the needs of long tail users. More importantly, as I will explain in a future post or research report, it will help cloud customers avoid the so called cascading failures (Cloud is a complex adaptive system and heterogeneity is critical to avoid cascading failures in such systems) in IT.
- Interoperability and Portability: For a federated system to satisfy the definition of cloud computing and to avoid lock-in, the cloud platforms must be interoperable. Also, users should be able to move their workloads and data between different providers. I don’t care if it works through standardization of APIs and formats or through third party services/libraries but it is a necessary characteristic.
- No vendor lock-in: The consequences of the above three conditions will ensure that there is no vendor lock-in. However, I would like to state this as a separate characteristic behavior of federated cloud ecosystems.
- Geographical distribution: This is also an essential characteristic of federated cloud ecosystems. Even the requirement of multiple providers in the market is not enough to meet the world’s computing needs, especially the ones that fall in the long tail region. Also, the regulatory regimes are not going away anytime soon (definitely, not in our lifetime). In order to meet these needs/requirements, there is also a need for geographical distribution of vendors/datacenters worldwide.
I think I have captured the essential characteristics of federated cloud ecosystems. I would love to hear your thoughts so that we can refine this definition in terms of both characteristics and, more importantly, the language. I sincerely hope widespread participation in this effort and if your affiliation in a particular organization is preventing you from public discussion, get in touch with me privately and I will bring those thoughts to the forefront while keeping your identity secret. Help me!!
Krishnan,
I’m in agreement here – I think that increasingly organisations and individuals will be consuming services which are delivered utilising resources owned by multiple providers in a federated network.
To me the key facets of a federated cloud are autonomy and cooperaion. The organisations who own the resources being utilised, and the services they deliver, are in ultimate control of their own environments and cooperate through agreement with other parties. At Arjuna, we argue that service agreements will be the glue that connects this network together. Those agreements will define not only the form of the service being delivered and its QoS but also potentially a myriad of other aspects, such as support levels, data secuirty, legal juridiction, pricing etc. I’ve blogged recently on this here – http://blog.arjuna.com/2011/10/11/service-agreements-in-the-cloud/
Steve Caughey, CEO, Arjuna technologies