January 17, 2013 | Written by: Maciej Sztukiewicz
Share this post:
If you survived my previous posts, let us explore one more dangerous and unproved path in a cloud world – cloud brokerage. If you are instead having first symptoms of insanity, I propose to watch another episode of “The Penguins of Madagascar,” relax and come back to my text with normal activity of your brain waves.
Don’t be afraid. The title of this article does not indicate that in a cloud business people usually are going broke, although it may be the case in today’s time of prosperity. It may however highlight a new opportunity that appears on the horizon. The thing is about “cloud brokers” that will shape tomorrow’s IT marketplace.
The cloud brokerage topic is under primary exploration, so let’s check what is being done today and then let’s play the futurologist game in the area. As I’ve heard, the good thing about being a futurologist is that you can be constantly wrong and still get a decent pay, so why not to try this.
Cloud computing subject draws attention of standardization bodies. National Institute of Standards and Technology (NIST) in the US developed cloud computing reference architecture, identifying actors and their roles in this domain. The one of particular interest for us is the cloud broker actor, described as “an entity that manages the use, performance and delivery of cloud services, and negotiates relationships between cloud providers and cloud consumers.”
According to NIST, several service areas can be supported by the cloud broker including:
- Service aggregation: The cloud broker combines and integrates multiple existing services into new service, carrying responsibility of data integration between cloud consumer and cloud providers.
- Service intermediation: A cloud broker provides value-added service, enhancing an existing service by improving some of its capabilities.
- Service arbitrage: This is similar to service aggregation but with flexible dynamic choice of service providers based on the broker’s internal evaluations.
So this is formal approach. Let’s verify it with market expectations. Is the cloud brokerage phenomenon really needed? Why is the appearance of such entities almost guaranteed? The simple answer is because cloud services, despite marketing truths, are usually difficult to digest by customers, especially if we are considering collection of interdependent cloud services taken from separate providers. Therefore external help is needed.
A common expectation is that cloud brokers can successfully help in the following ways:
- Normalize multiple cloud services available on the market, creating “apples-to-apples” comparison in functionality, pricing, and service levels.
- Get better discounts and access more information from cloud providers.
- Integrate outputs from multiple cloud service providers.
- Select the best cloud provider for specified requirements.
- Provide additional approval workflow functionality.
- Provide enhanced control, compliance, and security (as always isn’t it?).
Well, these are only the examples related to current state of knowledge about this immature market. The future will tell what other functions can be performed by such cloud brokerage entities. Talking about the future – let’s play the futurologist role for a while, generating bunch of potential new type cloud brokers firms that can shape the future IT industry.
The first possible entities of the new type that come to my mind we call “ghosts” and have no real infrastructure, data centers and servers, but provide complex products rapidly responding to urgent customer needs. They use cloud environments coming from established cloud providers to run virtual servers, store data, access programming environments but on top of them they rapidly create new products – applications and services – instantly deployed and delivered to final customer. The example of such a product can be marketing application promoting a new line of products for the customer who is launching a new marketing campaign. The real asset of this new company type is the ability to quickly react to customer needs and the knowledge of how to convert existing cloud market offerings into something new and appealing. Customer will receive online access to the complete product tailored to specified requirements – without needing to purchase software, servers, and storage from multiple providers. Using NIST terminology, ghost companies will provide service aggregation and service intermediation functions making use of existing cloud services and enhancing them with their value-added services.
My second thought is about “service multiplexers” – new types of companies providing customers with enhanced service, based on commodity offerings easily found on the market. An example is when a service multiplexer company provides customers with reasonably priced storage service, with agreed SLAs, taking the service itself remotely from multiple providers and converting it to new level of attractive offering, with more service guarantees and SLA levels. The “multiplexer” company would benchmark its providers regularly, finding the cheapest source of gigabytes of storage, directing customer data to a new place, and hiding all these operations from the customer. In the future it might even be possible to relocate customer data from one provider to the other seamlessly without customer notification. It might also be that for better access time or service level guarantee, the “multiplexer” company will store customer data at several providers offering coherent output to its customer.
Virtual cloud providers
Another entity type that appears on the IT horizon is the virtual cloud provider, which is similar to virtual mobile phone operator. A typical scenario here is that a company of that type does not invest in its own infrastructure but rather uses the infrastructure of another provider to deliver services under its “own” brand, and promoting this brand with help of dedicated marketing campaign. The difference to ghost companies is that virtual cloud providers do not aggregate or enhance existing services. They only resell them under a different brand.
Cloud transformation agents
We can also imagine a situation in which several market players specialize in transformation activities that help their customers to migrate from one cloud provider to another. Such a transformation agent can have running contracts with many cloud providers, access to their APIs, and ability to convert data from one cloud provider format to another.
The truth is that we cannot predict the future, taking as fundamental the things that we know now and extrapolating their capabilities as only imaginable. If this was the case, we would be still driving steam-powered wooden-finished vehicles running at millions of kilometers per hour now. Reality is that civilization is changed by unrecognized concepts and abstractions that did not exist before. The future reality can be that, because of usage of new materials (for example, graphene or next big thing) and developments in information processing technologies space, the cost of processing and storing information will be minimal. That will increase importance of brokers as entities shaping new products and creating new value for their customers.
Ok, so now you know what to do – I guess you are queuing to create a cloud broker company in your garage. By the way if you become a new Bill Gates of the modern cloud era, don’t hesitate to drop me a note thanking for a tip. Bye for now; as always please share your thoughts and let’s meet at my next post!
National Institute of Standards and Technology – NIST Cloud Computing Reference Architecture – Special Publication 500-292: http://www.nist.gov/customcf/get_pdf.cfm?pub_id=909505