MirrorHub/construct
0
0
Fork 0
mirror of https://github.com/matrix-construct/construct synced 2024-11-25 08:12:37 +01:00
Code Issues Releases Wiki Activity

doc: Update architecture summary.

Browse source
This commit is contained in:
Jason Volk 2019-09-27 20:26:02 -07:00
parent f6992ca3e1
commit 0d523f95e3
1 changed files with 27 additions and 69 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

96
doc/ARCHITECTURE.md
View file

@ -4,25 +4,19 @@
##### Single-threaded✝
The design of `libircd` is fully-asynchronous, single-thread-oriented. No code
in the library _blocks_ the process. All operations are conducted on top of
a single `boost::asio::io_service` which must be supplied by the executable
linking to `libircd`. That `io_service` must be orchestrated by the executable
at its discretion; typically the embedder's call to `ios.run()` is the only
place the process will _block_.
The design of `libircd` is fully-asynchronous, oriented around a single-thread
event-loop. No code in the library _blocks_ the process. All operations are
conducted on top of a single `boost::asio::io_service` which must be supplied
by the executable linking to `libircd`. That `io_service` must be run by the
executable at its discretion; typically the embedder's call to `ios.run()` is
the only place the process will _block_.
> Generally, applications are limited by one or more of the following bounds:
> Computing, Memory (Space), Memory (Peripheral I/O)
>
> `libircd` is dominated by the **I/O bound**.
Our effort is rooted in the above assumption. The single-threaded approach
ensures there is an _uninterrupted_, _uncontended_, _predictable_ execution
which is easy for developers to reason about intuitively with
sequential-consistency in a cooperative coroutine model. If there are periods
of execution which are computationally intense like parsing, hashing,
cryptography, etc: this is absorbed in lieu of thread synchronization and bus
contention.
The single-threaded approach ensures there is an _uninterrupted_, _uncontended_,
_predictable_ execution which is easy for developers to reason about intuitively
with sequential-consistency. This is ideal for the I/O-bound application being
facilitated. If there are periods of execution which are computationally intense
like parsing, hashing, cryptography, etc: this is absorbed in lieu of thread
synchronization and bus contention.
This system achieves scale through running multiple independent instances which
synchronize at the application-logic level through passing the application's own
@ -37,14 +31,12 @@ but it is heterogeneous.
##### Introduces userspace threading
IRCd presents an interface introducing stackful coroutines, a.k.a. userspace
context switching, a.k.a. green threads, a.k.a. fibers. The library avoids callbacks
as the way to break up execution when waiting for events. Instead, we harken back
to the simple old ways of synchronous programming where control flow and data are
easy to follow.
If there are certain cases where we don't want a stack to linger which may
jeopardize the c10k'ness of the daemon the asynchronous pattern is still used,
thus this is a hybrid system.
context switching, a.k.a. green threads, a.k.a. fibers. The library avoids
callbacks as the way to break up execution when waiting for events. Instead, we
harken back to the simple old ways of synchronous programming where control
flow and data are easy to follow. If there are certain cases where we don't
want a stack to linger which may jeopardize the c10k'ness of the daemon the
asynchronous pattern is still used (this is a hybrid system).
Consider coroutines like "macro-ops" and asynchronous callbacks like
"micro-ops." The pattern tends to use a coroutine to perform a large and
@ -53,15 +45,7 @@ approach relegates the asynchronous callback pattern to simple tasks contained
within specific units which require scale, encapsulating the complexity away
from the rest of the project.
##### Can be embedded in your application with very minimal overhead.
Linking to libircd from your executable allows you to customize and extend the
functionality of the server and have control over its execution, or, simply use
library routines provided by the library without any daemonization. Users of
the library should never pay for what they don't use. The library should also
minimize conflicts with other libraries sharing the address space.
##### Runs only one server at a time.
##### Runs only one server at a time
Keeping with the spirit of simplicity of the original architecture, `libircd`
continues to be a "singleton" object which uses globals and keeps actual server
@ -70,38 +54,12 @@ within a process's address space at a time.** Whether or not this was a pitfall
of the original design, it has emerged over the decades as a very profitable
decision for making IRCd an accessible open source internet project.
##### Leverages formal grammars
##### Formal grammars, RTTI, exceptions
We utilize the `boost::spirit` system of parsing and printing through formal grammars,
rather than writing our own parsers manually. In addition, we build several tools
on top of such formal devices like a type-safe format string library acting as a
drop-in for `::sprintf()`, but accepting objects like `std::string` without `.c_str()`
and prevention of outputting unprintable/unwanted characters that may have been
injected into the system somewhere prior.
##### Modular design
`libircd` is designed specifically as a shared object library. The purpose of its
shared'ness is to facilitate IRCd's modular design: IRCd ships with many other
shared objects which introduce the "business logic" and features of the daemon. If
`libircd` was not a shared object, every single module would have to include large
amounts of duplicate code drawn from the static library. This would be a huge drag
on both compilation and the runtime performance.
```
|-------------|
---------------------- | | < ---- (module)
| | | |
| User's executable | <---- | libircd | < ---- (module)
| | | |
---------------------- | | < ---- (module)
|-------------|
```
The user (which we may also refer to as the "embedder" elsewhere in
documentation) only deals directly with `libircd` and not the modules.
`libircd` is generally loaded with its symbols bound globally in the executable
and on most platforms cannot be unloaded (or even loaded) manually and has not
been tested to do so. As an aside, we do not summarily dismiss the idea of
reload capability and would like to see it made possible.
We utilize the `boost::spirit` system of parsing and printing through
compile-time formal grammars, rather than writing our own parsers manually.
In addition, we build several tools on top of such formal devices like a
type-safe format string library acting as a drop-in for `::sprintf()`, but
accepting objects like `std::string` without `.c_str()` and prevention of
outputting unprintable/unwanted characters that may have been injected into
the system somewhere prior.