1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
|
<center><table border="0" cellpadding="4">
<tr>
<td align="center"> <a href="https://github.com/ned14/afio">AFIO</a><br><a href="https://github.com/ned14/afio">on GitHub</a> </td>
<td align="center"> <a href="http://my.cdash.org/index.php?project=Boost.AFIO">CTest summary</a><br><a href="http://my.cdash.org/index.php?project=Boost.AFIO">dashboard</a> </td>
<td align="center"> <a href="https://travis-ci.org/ned14/afio">Linux and MacOS CI:</a><img src="https://travis-ci.org/ned14/afio.svg?branch=master"/> </td>
<td align="center"> <a href="https://ci.appveyor.com/project/ned14/afio/branch/master">Windows CI:</a><img src="https://ci.appveyor.com/api/projects/status/680b1pt9srnoprs3/branch/master?svg=true"/> </td>
</tr>
</table></center>
Herein lies my proposed zero whole machine memory copy async file i/o and filesystem
library for Boost and the C++ standard, intended for storage devices with ~1 microsecond
4Kb transfer latencies.
It is a complete rewrite after a Boost peer review in August 2015. Its github
source code repository lives at https://github.com/ned14/boost.afio.
<table width="100%" border="0" cellpadding="4">
<tr>
<th colspan="3">Why you might need AFIO<hr></th>
</tr>
<tr>
<td valign="top" width="33%">
Manufacturer claimed 4Kb transfer latencies for the physical hardware:
- Spinning rust hard drive latency @ QD1: **9000us**
- SATA flash drive latency @ QD1: **800us**
- NVMe flash drive latency @ QD1: **300us**
- RTT UDP packet latency over a LAN: **60us**
- NVMe Optane drive latency @ QD1: **60us**
- `memcpy(4Kb)` latency: **5us** (main memory) to **1.3us** (L3 cache)
- RTT PCIe latency: **0.5us**
</td>
<td valign="top" width="33%">
100% read QD1 4Kb direct transfer latencies for the software with AFIO:
- < 99% spinning rust hard drive latency: Windows **187,231us** FreeBSD **9,836us** Linux **26,468us**
- < 99% SATA flash drive latency: Windows **290us** Linux **158us**
- < 99% NVMe drive latency: Windows **37us** FreeBSD **70us** Linux **30us**
Worst case AFIO read overhead benchmarked by author: **0.19us** (5.3M IOPS @ QD1, approx 20Gb/sec)
</td>
<td valign="top" width="33%">
75% read 25% write QD4 4Kb direct transfer latencies for the software with AFIO:
- < 99% spinning rust hard drive latency: Windows **48,185us** FreeBSD **61,834us** Linux **104,507us**
- < 99% SATA flash drive latency: Windows **1,812us** Linux **1,416us**
- < 99% NVMe drive latency: Windows **50us** FreeBSD **143us** Linux **40us**
Worst case AFIO write overhead benchmarked by author: **0.18us** (5.6M IOPS @ QD1, approx 21Gb/sec)
</td>
</tr>
</table>
\note Note that this code is of late alpha quality. It's quite reliable on Windows and Linux, but be careful when using it!
These compilers and OS are regularly tested:
- GCC 7.0 (Linux 4,x x64)
- clang 4.0 (Linux 4.x x64)
- clang 5.0 (Windows 10 x64)
Other compilers, architectures and OSs may work, but are not tested regularly. You will need a Filesystem TS
implementation in your STL and C++ 14. See https://github.com/ned14/afio/blob/master/programs/fs-probe/fs_probe_results.yaml
for a database of latencies for various previously tested OS, filing systems and storage devices.
Todo list for already implemented parts: https://ned14.github.io/afio/todo.html
To build and test (make, ninja etc):
~~~
mkdir build
cd build
cmake ..
cmake --build .
ctest -R afio_sl
~~~
To build and test (Visual Studio, XCode etc):
~~~
mkdir build
cd build
cmake ..
cmake --build . --config Release
ctest -C Release -R afio_sl
~~~
## v2 architecture and design implemented:
| NEW in v2 | Boost peer review feedback | |
| --------- | -------------------------- | --- |
| ✔ | ✔ | Universal native handle/fd abstraction instead of `void *`.
| ✔ | ✔ | Perfectly/Ideally low memory (de)allocation per op (usually none).
| ✔ | ✔ | noexcept API throughout returning error_code for failure instead of throwing exceptions.
| ✔ | ✔ | AFIO v1 handle type split into hierarchy of types:<ol><li>handle - provides open, close, get path, clone, set/unset append only, change caching, characteristics<li>path_handle - a race free anchor to a subset of the filesystem<li>io_handle - adds synchronous scatter-gather i/o, byte range locking<li>file_handle - adds open/create file, get and set maximum extent<li>async_file_handle - adds asynchronous scatter-gather i/o</ol>
| ✔ | ✔ | Cancelable i/o (made possible thanks to dropping XP support).
| ✔ | ✔ | All shared_ptr usage removed as all use of multiple threads removed.
| ✔ | ✔ | Use of std::vector to transport scatter-gather sequences replaced with C++ 20 `span<>` borrowed views.
| ✔ | ✔ | Completion callbacks are now some arbitrary type `U&&` instead of a future continuation. Type erasure for its storage is bound into the one single memory allocation for everything needed to execute the op, and so therefore overhead is optimal.
| ✔ | ✔ | Filing system algorithms made generic and broken out into public `afio::algorithms` template library (the AFIO FTL).
| ✔ | ✔ | Abstraction of native handle management via bitfield specified "characteristics".
| ✔ | | Storage profiles, a YAML database of behaviours of hardware, OS and filing system combinations.
| ✔ | | Absolute and interval deadline timed i/o throughout (made possible thanks to dropping XP support).
| ✔ | | Dependency on ASIO/Networking TS removed completely.
| ✔ | | Four choices of algorithm implementing a shared filing system mutex.
| ✔ | | Uses CMake, CTest, CDash and CPack with automatic usage of C++ Modules or precompiled headers where available.
| ✔ | | Far more comprehensive memory map and virtual memory facilities.
| ✔ | | Much more granular, micro level unit testing of individual functions.
| ✔ | | Much more granular, micro level internal logging of every code path taken.
| ✔ | | Path views used throughout, thus avoiding string copying and allocation in `std::filesystem::path`.
| ✔ | | Paths are equally interpreted as UTF-8 on all platforms.
| ✔ | | We never store nor retain a path, as they are inherently racy and are best avoided.
Todo:
| NEW in v2 | Boost peer review feedback | |
| --------- | -------------------------- | --- |
| ✔ | | clang AST assisted SWIG bindings for other languages.
| ✔ | | Statistical tracking of operation latencies so realtime IOPS can be measured.
## Planned features implemented:
| NEW in v2 | Windows | POSIX | |
| --------- | --------| ----- | --- |
| ✔ | ✔ | ✔ | Native handle cloning.
| ✔ (up from four) | ✔ | ✔ | Maximum possible (seven) forms of kernel caching.
| | ✔ | ✔ | Absolute path open.
| | ✔ | ✔ | Relative "anchored" path open enabling race free file system.
| ✔ | ✔ | | Win32 path support (260 path limit).
| | ✔ | | NT kernel path support (32,768 path limit).
| ✔ | ✔ | ✔ | Synchronous universal scatter-gather i/o.
| ✔ (POSIX AIO support) | ✔ | ✔ | Asynchronous universal scatter-gather i/o.
| ✔ | ✔ | ✔ | i/o deadlines and cancellation.
| | ✔ | ✔ | Retrieving and setting the current maximum extent (size) of an open file.
| | ✔ | ✔ | Retrieving the current path of an open file irrespective of where it has been renamed to by third parties.
| | ✔ | ✔ | statfs_t ported over from AFIO v1.
| | ✔ | ✔ | utils namespace ported over from AFIO v1.
| ✔ | ✔ | ✔ | `shared_fs_mutex` shared/exclusive entities locking based on lock files
| ✔ | ✔ | ✔ | Byte range shared/exclusive locking.
| ✔ | ✔ | ✔ | `shared_fs_mutex` shared/exclusive entities locking based on byte ranges
| ✔ | ✔ | ✔ | `shared_fs_mutex` shared/exclusive entities locking based on atomic append
| | ✔ | ✔ | Memory mapped files and virtual memory management (`section_handle` and `map_handle`)
| ✔ | ✔ | ✔ | `shared_fs_mutex` shared/exclusive entities locking based on memory maps
| ✔ | ✔ | ✔ | Universal portable UTF-8 path views.
| | ✔ | ✔ | "Hole punching" and hole enumeration ported over from AFIO v1.
| | ✔ | ✔ | Directory handles and very fast directory enumeration ported over from AFIO v1.
| ✔ | ✔ | ✔ | `shared_fs_mutex` shared/exclusive entities locking based on safe byte ranges
Todo to reach feature parity with AFIO v1:
| NEW in v2 | Windows | POSIX | |
| --------- | --------| ----- | --- |
| | | | Hard links and symlinks.
| | | | Set random or sequential i/o (prefetch).
| | | | BSD and OS X kqueues optimised `io_service`
Todo thereafter:
| NEW in v2 | Windows | POSIX | |
| --------- | --------| ----- | --- |
| ✔ | | | Extended attributes support.
| ✔ | | | Coroutines TS integration for `async_file_handle`. Use example: https://gist.github.com/anonymous/a67ba4695c223a905ff108ed8b9a342f
| ✔ | | | Linux KAIO support for native async `O_DIRECT` i/o
| ✔ | | | Reliable directory hierarchy deletion algorithm.
| ✔ | | | Reliable directory hierarchy copy algorithm.
| ✔ | | | Reliable directory hierarchy update (two and three way) algorithm.
| ✔ | | | Algorithm to replace all duplicate content with hard links.
| ✔ | | | Algorithm to figure out all paths for a hard linked inode.
Max bandwidth for the physical hardware:
- DDR4 2133: **30Gb/sec** (main memory)
- x4 PCIe 4.0: **7.5Gb/sec** (arrives end of 2017, the 2018 NVMe drives will use PCIe 4.0)
- x4 PCIe 3.0: **3.75Gb/sec** (985Mb/sec per PCIe lane)
- 2017 XPoint drive (x4 PCIe 3.0): **2.5Gb/sec**
- 2017 NVMe flash drive (x4 PCIe 3.0): **2Gb/sec**
- 10Gbit LAN: **1.2Gb/sec**
|
