Why PGASP is more cost efficient than other stacks

PGASP is open-source and free. No licencing costs.

Leveraging Postgres which is open-source and free. Leveraging Apache which is open-source and free. Leveraging Linux which is open-source and free.

Making use of existing Postgres and PL/pgSQL skills, no need to learn new languages or technologies.

Smaller code base, fewer bugs, requires less testing.

Native SQL compilation, “bad” dynamic SQL is made impossible by design, reducing testing time even further.

Easier collaboration between graphics designers and software developers. Saving time and money.

Potentially reduced hardware and hosting/cloud costs due to less resources used by PGASP compared to other stacks.

Why PGASP is faster than other stacks

There is no “middle man” (or even multiple “middle men”).

Apache connects directly to Postgres.

No data marshalling.

No network trips, or any extra trips. Just one request from Apache to Postgres, and one reply.

No data transformation. Reduced CPU time, allowing more time for real tasks.

No multiple copies of the same data. Reduced memory usage, allowing more memory for cache and buffers, which in turn significantly reduces disk I/O.

Why PGASP is more secure than other stacks

There is no source code stored in the file system.

There is no authentication information stored in the code, only in apache config file (owned by root).

Allowed pages/functions must be explicitely declared in the config file (owned by root).

Parameters are treated as native PL/pgSQL variables, no risk of SQL injection.

Getting started

As of Jan 2015, the environment is Ubuntu Server 14.10 with standard LAMP stack plus Postgres 9.4 from official Ubuntu repositories.

The PGASP compiler and the Apache mod_pgasp module are written in C.

Helper functions for Postgres procedures are written in PL/pgSQL.

Other software used (optional): git, pgAdmin III.