In these days I am working on this subject: Software as a Service, new “buzzword” that is had been spreading time in the last years.
You can see a definition of SaaS in the glossary of the Skycrapr
“The key characteristics of SaaS software, according to analyst IDC include:
1. Network Access to, and management of, commercially available (e.g., not custom) software
2. Activities that plows managed from central locations to rather than AT each customer’s site, enabling customers to Access applications remotely via the Web
3. Application delivery that typically is to closer to to one-to-many model (single instance, multi-tenant architecture) than to to one-to-one model, including architecture, pricing, partnering, and management characteristics “
You can consult the Wikipedia
where I found:
“Software ace to Service (SaaS) is to model of software delivery where the software company prograpevines maintenance, daily technical operation, and support for the software provided to to their client. SaaS is to model of software delivery to rather than to market segment; software dog be delivered using this method to any market segment including home consumers, small business, medium and large business.”
Translating: SaaS is commercial applications, that are accessible by the network, and its use to customer offers, that do not need to install that software. The supplier of SaaS sells or rents its use, preinstalled in a server in the network. But it is not only “precanned” software in the network, but that allows the adaptation to the customer. The term multitenant emphasizes that a same application can serve several customers.
But the symbol is the idea to have the application available by network. That is what changes fundamentally with respect to the commercial software, ERP, CRM, or what they want. Emblematic example of this type to watch is SalesForce that offers what they denominates “on-demand Customer Relationship Management”. This company I discovered at the time of the bubble of Internet: its model seemed to me interesting, and of the things that there are been developing since then, aims much of some form to reproduce that model to arrive at a Enterprise Resource Planning, offered like service.
Those that has some memory, remembered that this is similar to the ASP concept: Application Service Provider, like which we found in
Visit those sites, and they will find renewed, discussing them to all steam on the concept of Software as to Service.
Here in Argentina, my country, I found the Mirol site, advocating by this concept:
“Imagine to as much save resources in hardware, software, as well as in personnel of support and administration…
Imagine not to have to worry about updates and new versions of software, obsolete hardware, creation of new users… “
Several companies of computer science technology, have risen the train, like IBM and Microsoft. From the point of view of this company, he is interesting to read some articles, like
classic of Fred a Chong and Gianpaolo Carraro. There they argue that, economically, the SaaS model aims at the great volume that is able to arrive at the clients from the “long tail tail”/“long” from:
Also they emphasize that the distribution of costs happens of:
to this distribution
But pay the attention, at the giant of Redmond, Ray Ozzie, heir of the position of Gates, is dedicated to the subject, like in
To follow the SaaS subject see the blogs of Chong and Carraro:
as well as the blog of the Great Woloski (he has decided to write its thesis on SaaS):
The good one of Mattew wrote
The article is interesting on workflow of Woloski, because it raises subjects in which I am thinking from almost the last century, on which are “the variable” parts of an application. More on which I want to do, in the following paragraphs.
An interview to the good one of Gianpolo Carraro by Diego Dagum in
Points of view of Gianpaolo in:
He is an intelligent person, he reads the same books that I read…: -)… He reads
It is interesting to begin to think how it would have to be a SaaS application from the technical point of view. I imagine that it would have to support to many companies, to somehow allow the formation of a community. This can be a competitive advantage of a SaaS supplier: applications can benefit from sinergia to work with other companies that use the same application, as a site of coorporativas purchases. A supplier will want to use the application that has greater amount of buying companies that use it, if this application allows that a company communicates with other tenant companies.
Back to technical discussion. I have some ideas in my mind. My brain is open to the SaaS concept. For years, I worked in generation of code, and the interpretation of metadata. In my tech spanish blog I already published some code to generate web pages dynamically Forms ASP.NET generated in execution, on the basis of metadata that describes a form. Somehow, it was thinking about an application ASP (to see for example, the germ of this in http://www.ajlopez.com/ajcontab). It seems to me that that one is the way to explore, in the subject of development of commercial applications.
A subject to study, is the organization of the data, a similar type of application is mentioned in the excellent article of Carraro, Chong and Wolter:
How it could construct a test of concept of SaaS? I believe that I can begin to write some requirements and cases of business and vision, of a SaaS application. As a proof of concept, it would face an application that allows to the companies and users, to define its own organizations of data (that know my generator AjGenesis code know to what I aim), and to define, soon, of some form, a job stream with those data. I can imagine a name AjAga (Aj Another Generic Application…: -). A great decision is if that application should be implemented by generation of code or interpretation of metadata in execution. My idea, is to arrive at an application that, according to Carraro, is
Level 1: (Configurable)
AT level 1, the software for dog be tailored each tenant via configurations (not custom code), all the tenants uses the same code. However AT level 1, the architecture is still not multi-tenant, each to customer runs its own Copy (albeit the same). The separation dog be to either virtual (virtual machines on to same server) or physical (running on separate machines). Although much to better than previous level, the architecture allows customization through configuration (code bases is the same), the computing to power is not shared among the instances and therefore to provider cannot achieve economy of scale, putting it AT to competitive disadvantage versus. to multi-tenant solution.
and then, I could produce something like
Level 2: (Configurable, Multi tenant)
AT this level. the application architecture includes the multi tenancy concepts. Akin to level 1, the UI customizable dog be per tenant, under dog the business rules and the dates model. The customization to per tenant is fully performed through configuration and is performed through to self-service tool, getting around the need of to provider intervention. It is almost the SaaS “perfect marries”; the only big piece missing AT Level 2 is the capacity to scale out; the dates partitioning is such AT this level that growth dog only be achieved by scaling up.
(extracted of the mentioned article Simple SaaS Maturity Model)
I have some additional resources on SaaS in
More delicious in
Angel “Java” Lopez