SAF Use Cases

[Stavros.Isaiadis@uk.fujitsu.com]

Hi all,

I hope all our US colleagues had a good time on thanksgiving last week.

A reminder that it would be nice to have a number of use cases to discuss so that we can drive the cloud profile forward with more concrete requirements. It doesn't have to be detailed at this point.

So, in an effort to "provoke" further discussion and submissions, I am posting here a use case in the domain of stock market. It is by no means complete or detailed, but I think it can potentially demonstrate the cross domain integration, and the combination of both live feeds and static historical data.

Feel free to tear this to pieces so that subsequent use cases are more complete :-)


===============
Overview

The goal is for the end user/client to integrate (directly or indirectly through a broker) stock market services and combine and analyze data streams. The client will use cloud provider resources to scale up and down as necessary to support high computational demands and elastic storage capacity. Computational demands pertain to the analysis of the combined data streams looking for undervalued combinations of stock prices, hedging opportunities, well priced options, and more.

Symptoms might include the changes in market prices, alerts for very steep changes, changes in commission from the brokers/intermediaries, volatility of currency exchanges, and various other pricing and options related events.

Such events could also be combined with on demand fetching and integration of relevant historical data from static repositories, for data mining and pattern matching to identify other potential revenue sources based on longer term statistical analysis and quantitative models.

Roles

A number of service providers can participate in such use case, some of which might be in the cloud, others more traditional SaaS. These provide services such as querying stock prices, subscriptions to live price feeds, statistical analysis tools, etc.  A cloud service broker might act as the single entry point for the end user.

A cloud provider will provide typical IaaS capabilities.

The end user/cloud consumer can be a trader, hedge fund, or investment firm. The end user makes use of the cloud to accommodate the computational and storage needs, and uses the various service in his application to be deployed in the cloud.

So, the emitters are the various services provided, which could all be part of an integrated PaaS (in which case we have only a single emitter in the PaaS cloud provider??). The diagnostician is (probably?) the user application that provides the catalogue of what needs to be monitored and how to react. The practitioner is the user application or a trading platform which responds to the symptoms emitted by making recommendations on how to proceed in the market. 

==============