SYSTEM ARCHITECTURE
=============================================================

CROSS-PLATFORM COMPATIBILITY
we use the C language with ANSI specifications; we will run the application on Linux, Windows, MacOS, Unix and other operating systems

ZERO ENTITY COLLISION
we store the ID of some entities in UUID format, in order to prevent collisions on entities defined across different machines

SIMULATION TIME SCALABILITY
we store the simulation time in a string format (e.g. YYYY-MM-DD-HH-NN-SS-LLL) which allows us to increase the simulation time scale (e.g. ...YYYY-MM-DD-HH-NN-SS-LLL-...)
without having to change the application code

UNIVERSAL DATA STORAGE
we use a file addressing system 

ZERO DATABASE ADMINISTRATION
we use an embedded database for data storage

ZERO DATA LOSS
we use ACID compliant transactions that protect us against hardware & software crashes; we replicate the same data across multiple machines

FAST DATA ACCESS
we use non-linear data access using B-TREEs, database connection caching and in-memory databases

ZERO DATA CONCURRENCY
we use one writer and multiple readers for each database file 

ZERO STRUCTURAL DATABASE CHANGES
we keep the object id in the Objects table so that we can allow for future changes without having to change the structure of the table; 
if we need to add new columns (e.g. UserName or LastUpdateDate) we only need to define new properties for the object (e.g. user name, last update date) 
and assign values to them; thus we do not need to change all database files across all machines when doing an upgrade

HISTORICAL DATA STORAGE
we store in Property_Values table only the current property values that have been changed since last time step; we will 
end up with property values that have not changed since creation (e.g. neuron type) and other ones that are changed at 
each time step (e.g. voltage); the Archive_Property_Values will store the property values that have been changed over time

ZERO DATA CORRUPTION IN MEMORY
we use Valgrind to identify memory leaks

CELLULAR MODELLING SCALABILITY
we use generalized data structures for storage of any cell model

UNIVERSAL INTEGRATION METHODS
we can use multiple integration methods during simulation of different objects

ZERO INSTALLATION STEPS
we use a self-contained executable that does not require external libraries; setup requires file copy only

UNIVERSAL GRAPHICS PLATFORM
we use OpenGL

SINGLE-THREADED MULTIPLE PROCESSES
we will use a single-threaded model for the application but we will run multiple instances of the module processes

ZERO APPLICATION CHANGES FOR NEW CELL MODELS
we use a modular architecture with local message queues

SINGLE POINT OF CONTROL
we will use a single point of command and control; if data is corrupted on more than x% on 
all machines in the entire set we must redo the computations on other machines; we will use redundant machines 

COMPUTATIONAL PROCESSING SCALABILITY
we use automatic process creation and destruction based on each machine's available resources; the application 
will be able to scale up to the maximum available resources

HIGH SECURITY
we use data encryption and private & public keys for data transfer; users, groups, permissions, access rights and auditing (USERS: user id, user name, password, is account locked, last connected date, creation date, created by user, modification date, modified by user)

SIMULATION TRANSACTIONS
we will use transactions on top of database transactions

ZERO DOWNTIME
we will have the core system working even when performing entity updates

ZERO APPLICATION CHANGES
once the simulator's source code will be completed, we will freeze its development and switch to the development 
of brain functions like learning, decision taking and everything else

UNIVERSAL DATA VISUALIZATION
we will be able to visualize brain processes on any machine and at any level, from molecular to system level

SIMPLE TO USE
we will build the system in such a way so that the application can be used both by researchers and by people with little or no expertise in neuroscience;
brain processes will be employed without the need to be understood