Four generations of SaaS
Cloud computing together with Enterprise 2.0 and the mobile Internet were clearly among the leaders of public opinion following the results of the past year and in the forecasts of IT development for 2009. It goes without saying that the debate over SaaS (software as a service) —the progenitor of cloud computing — has modestly become only one of its branches in this direction * [ 1 , 2 , 3 , 4 , 5 , 6 ]. A blessing in disguise, but under the influence of the outbreak of the global recession, which once again showed how connected everything is in our world, even violent proponents of the uniqueness of the Russian market stopped saying that " SaaS is not for us ."
However, on both sides of the ocean, many still do not quite clearly represent, and what is SaaS. And not surprisingly, in less than 10 years, four generations have changed in this area. Almost like Drosophila in genetics :). Not really, however, have changed: the last generation of SaaS, it seems, will live together peacefully for a long time. Let's try to define SaaS, namely, through the features of each of these generations.
The first generation (late 90s - early 2000s). Its name coincides with the name of the group of web providers that provide relevant services - Application Service Provider (ASP) . ASP is nothing more than a service provided by a provider to a customer and / or its customers through some web application program that the provider itself has placed on its server. Is this program its own development or did it just buy it (a license to use the program), is a secondary matter. The main points here are two points characteristic of SaaS as a whole. First, the provider removed from the customer all the problems associated with the installation of the program and its maintenance. In other words, the traditional web hosting provider has started to additionally provide another service - to provide users with an application program that they have. And second, the payment for such a complex service was charged only after the customer subscribed to the program for temporary rental of the program (on-demand) , that is, the customer did not need to buy the program itself in personal possession (on-premise). I would call the first generation "pre-SaaS".
')
The second generation (appeared somewhere in late 2003). This is the beginning, in fact, of SaaS generations, and SalesForce.com is their first, well-known representative. Then it went for a walk around the world this, perhaps, not quite a beautiful abbreviation of SaaS. There was one, but very fundamental difference between the second generation and the first. If each client in ASP was provided with a separate sample of the rented web program, then in SaaS such program worked in the collective lease mode (multi-tenant) . And this already required special development “for SaaS” , which was not needed by the programs whose services were provided by ASP providers (in principle, they could have installed any web program). At the same time, I want to emphasize once again that the right to use the leased program of any generation of SaaS (including ASP) automatically extended to the tenant’s clients, who had already determined their working conditions. Unless, of course, the program provided for work in multi-user mode (not to be confused with collective rent).
The third generation (somewhere at the turn of 2005 and 2006). This generation was born within the framework of a more general direction of computing in the clouds, along with Amazon S2 - the first most prominent representative of this direction. The main feature of SaaS of this generation was that externally (at the functional level) collective rentals seemed to remain, but at the logical (and sometimes at the physical) level, each customer received his own separately working sample program for the subscription period. In other words, if with ASP the possible physical or virtual server allocation (dedicated server) occurred in a static mode, then in the third generation of SaaS we are already dealing with a dynamic or elastic allocation of such a “server”, on which the standalone programs. You can compare the second generation of SaaS with a "transparent" communal apartment, in which tenants (tenants) and their guests (customers) do not see other tenants (and guests). Although they know that they exist very close. In this regard, the third generation of SaaS is already an apartment building, where a separate apartment is promptly created for the tenant, which disappears at the end of the lease.
The fourth direction (closer to 2007). It differs from the third only in the model of payment for the service. If the payment in the third generation of SaaS, as in all previous ones, was made on a temporary basis (subscription for a week, month, year), then in the fourth generation the customer is charged only when he and / or his clients directly use the services of the leased program (pay -as-you-do) .
The characteristic features of the last two (purely cloudy) SaaS generations, as under certain conditions and its second generation, can be used in various combinations, and the SaaS systems themselves, in general, work on different platforms in the clouds. At the same time, the proposal to withdraw from the SaaS term is now increasingly heard, replacing it with Application-as-a-Service (AaaS). This, in the opinion of many, with whom it is impossible to disagree, more accurately reflects the place of SaaS (AaaS) in the system of directions associated with cloud computing. Indeed, in most cases, one of the main components of the “platform as a service” (PaaS) itself is also a program (software system), but already operational and, perhaps, instrumental.
* In the list of my notes you will find all the necessary links to other sources.
However, on both sides of the ocean, many still do not quite clearly represent, and what is SaaS. And not surprisingly, in less than 10 years, four generations have changed in this area. Almost like Drosophila in genetics :). Not really, however, have changed: the last generation of SaaS, it seems, will live together peacefully for a long time. Let's try to define SaaS, namely, through the features of each of these generations.
The first generation (late 90s - early 2000s). Its name coincides with the name of the group of web providers that provide relevant services - Application Service Provider (ASP) . ASP is nothing more than a service provided by a provider to a customer and / or its customers through some web application program that the provider itself has placed on its server. Is this program its own development or did it just buy it (a license to use the program), is a secondary matter. The main points here are two points characteristic of SaaS as a whole. First, the provider removed from the customer all the problems associated with the installation of the program and its maintenance. In other words, the traditional web hosting provider has started to additionally provide another service - to provide users with an application program that they have. And second, the payment for such a complex service was charged only after the customer subscribed to the program for temporary rental of the program (on-demand) , that is, the customer did not need to buy the program itself in personal possession (on-premise). I would call the first generation "pre-SaaS".
')
The second generation (appeared somewhere in late 2003). This is the beginning, in fact, of SaaS generations, and SalesForce.com is their first, well-known representative. Then it went for a walk around the world this, perhaps, not quite a beautiful abbreviation of SaaS. There was one, but very fundamental difference between the second generation and the first. If each client in ASP was provided with a separate sample of the rented web program, then in SaaS such program worked in the collective lease mode (multi-tenant) . And this already required special development “for SaaS” , which was not needed by the programs whose services were provided by ASP providers (in principle, they could have installed any web program). At the same time, I want to emphasize once again that the right to use the leased program of any generation of SaaS (including ASP) automatically extended to the tenant’s clients, who had already determined their working conditions. Unless, of course, the program provided for work in multi-user mode (not to be confused with collective rent).
The third generation (somewhere at the turn of 2005 and 2006). This generation was born within the framework of a more general direction of computing in the clouds, along with Amazon S2 - the first most prominent representative of this direction. The main feature of SaaS of this generation was that externally (at the functional level) collective rentals seemed to remain, but at the logical (and sometimes at the physical) level, each customer received his own separately working sample program for the subscription period. In other words, if with ASP the possible physical or virtual server allocation (dedicated server) occurred in a static mode, then in the third generation of SaaS we are already dealing with a dynamic or elastic allocation of such a “server”, on which the standalone programs. You can compare the second generation of SaaS with a "transparent" communal apartment, in which tenants (tenants) and their guests (customers) do not see other tenants (and guests). Although they know that they exist very close. In this regard, the third generation of SaaS is already an apartment building, where a separate apartment is promptly created for the tenant, which disappears at the end of the lease.
The fourth direction (closer to 2007). It differs from the third only in the model of payment for the service. If the payment in the third generation of SaaS, as in all previous ones, was made on a temporary basis (subscription for a week, month, year), then in the fourth generation the customer is charged only when he and / or his clients directly use the services of the leased program (pay -as-you-do) .
The characteristic features of the last two (purely cloudy) SaaS generations, as under certain conditions and its second generation, can be used in various combinations, and the SaaS systems themselves, in general, work on different platforms in the clouds. At the same time, the proposal to withdraw from the SaaS term is now increasingly heard, replacing it with Application-as-a-Service (AaaS). This, in the opinion of many, with whom it is impossible to disagree, more accurately reflects the place of SaaS (AaaS) in the system of directions associated with cloud computing. Indeed, in most cases, one of the main components of the “platform as a service” (PaaS) itself is also a program (software system), but already operational and, perhaps, instrumental.
* In the list of my notes you will find all the necessary links to other sources.
Source: https://habr.com/ru/post/50314/
All Articles