You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hydra is a C++11 compliant and header only framework designed to perform commom data analysis tasks on massively parallel platforms. Hydra provides a collection of containers and algorithms commonly used in HEP data analysis, which can deploy transparently OpenMP, CUDA and TBB enabled devices, allowing the user to re-use the same code across a large range of available multi-core CPU and accelerators. The framework design is focused on performance and precision.
34
+
Hydra is a C++14 compliant and header only framework designed to perform common data analysis tasks on massively parallel platforms. Hydra provides a collection of containers and algorithms commonly used in HEP data analysis, which can deploy transparently OpenMP, CUDA and TBB enabled devices, allowing the user to re-use the same code across a large range of available multi-core CPU and accelerators. The framework design is focused on performance and precision.
28
35
29
-
The core algorithms follow as close as possible the implementations widely used in framworks like ROOT and libraries
36
+
The core algorithms follow as close as possible the implementations widely used in frameworks like ROOT and libraries
30
37
like GSL.
31
38
32
39
Main features
33
40
-------------
34
41
35
42
Currently Hydra supports:
36
43
37
-
* Generation of phase-space Monte Carlo samples with any number of particles in the final states. Sequential decays, calculation of integrals of models
38
-
over the corresponding phase-space and production of weighted and unweighted samples, which can be flat or distributed following a model provided by the user.
44
+
* Generation of phase-space Monte Carlo samples with any number of particles in the final states. Sequential decays, calculation of integrals of models over the corresponding phase-space and production of weighted and unweighted samples, which can be flat or distributed following a model provided by the user.
39
45
* Sampling of multidimensional pdfs.
40
46
* Multidimensional maximum likelihood fits using binned and unbinned data sets.
41
-
* Calculation of S-Plots, a popular technique for statistical unfolding of populations contributing to a sample.
42
-
* Evaluation of multidimensional functions over heterogeneos data sets.
47
+
* Multi-layered simultaneous fit of different models, over different datasets, deploying different parallelization strategies for each model.
48
+
* Calculation of S-Plots, a popular technique for statistical unfolding of populations contributing to a sample.
49
+
* Evaluation of multidimensional functions over heterogeneous data sets.
43
50
* Numerical integration of multidimensional functions using self-adaptive Monte Carlo and quadrature methods.
44
-
* Multidimensional sparse and dense histogramming of large samples.
51
+
* Multidimensional sparse and dense histogramming of large samples.
52
+
* Object-based interface to FFTW and CuFFT for performing Fast Fourier Transform in CPU and GPU.
53
+
* FFT based one-dimensional convolution for arbitrary signal and kernel shapes.
54
+
* Booststrap and real cubic spiline for datasets on CPU and GPU.
55
+
* Sobol low discrepance sequences up to 3667 dimensions.
45
56
46
-
Hydra also provides a bunch of custom types, optimized containers and a number of algorithms and constructs to maximaze performance, avoiding unecessary usage of memory and without losing the flexibility and portability to compile and run the same code across different platforms and deployment scenarios.
57
+
Hydra also provides a bunch of custom types, optimized containers and a number of algorithms and constructs to maximize performance, avoiding unnecessary usage of memory and without losing the flexibility and portability to compile and run the same code across different platforms and deployment scenarios.
47
58
48
59
49
-
For example, just changing .cu to .cpp in any source code writen only using the Hydra and standard C++11 is enough
60
+
For example, just changing .cu to .cpp in any source code written only using the Hydra and standard C++14 is enough
50
61
to compile your application for OpenMP or TBB compatible devices using GCC or other compiler in a machine without a NVIDIA GPU installed.
51
62
52
-
In summary, by using Hydra the user can transparently typical bottle-neck calculations to a suitable parallel device and get speed-up factors ranging from dozens to hundreds.
63
+
In summary, by using Hydra the user can transparently implement and dispatch typical bottle-neck calculations to a suitable parallel device and get speed-up factors ranging from dozens to hundreds.
53
64
54
65
55
66
56
67
Hydra and Thrust
57
68
----------------
58
69
59
-
Hydra is implemented on top of the [Thrust library](https://thrust.github.io/) and relies strongly on Thrust's containers, algorithms and backend managment systems.
60
-
The official version of Thrust supports tuples with maximum ten elements. In order to overcome this limitation, Hydra uses the
70
+
Hydra is implemented on top of the [Thrust library](https://thrust.github.io/) and relies strongly on Thrust's containers, algorithms and backend management systems.
71
+
The official version of Thrust supports tuples with maximum ten elements. In order to overcome this limitation, Hydra uses an
61
72
[unofficial version, forked from the original, by Andrew Currigan and collaborators](https://github.com/andrewcorrigan/thrust-multi-permutation-iterator).
62
-
This version implements variadic tuples and related classes, as well as provides some additional functionalities, which are missing in the official Thrust.
73
+
This version implements variadic tuples and related classes, as well as provides some additional functionality, which are missing in the official Thrust.
74
+
In order to keep Hydra uptodated with the latest bug-fixes and architetural improvements in Thrust, at each Hydra release, the official [Thrust library](https://thrust.github.io/) is patched with the Currigan's variadic tuple implementation.
63
75
64
-
The version of Thrust distributed with Hydra is maintained by [MultithreadCorner](https://github.com/MultithreadCorner). It is basically a fork of Currigan's repository, which was merged with the latest official release available in github (Thrust 1.8.3).
76
+
So, version of Thrust distributed with Hydra is maintained by [MultithreadCorner](https://github.com/MultithreadCorner). It is basically a fork of Currigan's repository, which was merged with the latest official release available in GitHub (Thrust 1.9.7).
65
77
66
-
***Hydra does not depend or conflits with the official Thrust library distributed with the CUDA-SDK.***
78
+
***Hydra does not depend or conflict with the official Thrust library distributed with the CUDA-SDK.***
The available "host" and "device" backends can be freely combined.
88
100
Two important features related to Hydra's design and the backend configuration:
89
101
90
-
* If CUDA backend is not used, [NVCC and the CUDA runtime](https://developer.nvidia.com/cuda-toolkit) are not necessary. The programs can be compiled with GCC, CLang or other host compiler compatible with C++11 directly.
102
+
* If CUDA backend is not used, [NVCC and the CUDA runtime](https://developer.nvidia.com/cuda-toolkit) are not necessary. The programs can be compiled with GCC, Clang or other host compiler compatible with C++14 support directly.
91
103
* Programs written using only Hydra, Thrust, STL and standard c++ constructs, it means programs without any raw CUDA code or calls to the CUDA runtime API, can be compiled with NVCC, to run on CUDA backends, or a suitable host compiler to run on OpenMP , TBB and CPP backends. **Just change the source file extension from .cu to .cpp, or something else the host compiler understands.**
92
104
93
105
94
106
The Latest Version
95
107
------------------
108
+
The latest release can be downloaded [here](https://github.com/MultithreadCorner/Hydra/releases/latest).
96
109
97
110
Documentation
98
111
-------------
@@ -110,7 +123,7 @@ It is also possible to browse the documentation by class, file or name using the
The Hydra User's Guide is available in the following formats:
116
129
@@ -123,11 +136,12 @@ Installation and requirements
123
136
124
137
Hydra is a header only library, so no build process is necessary to install it.
125
138
Just place the `hydra` folder and its contents where your system can find it.
126
-
The library run on Linux systems and requires at least a host compiler supporting C++11. To use NVidia's GPUs, CUDA 8 or higher is required.
127
-
A suite of examples demonstrating the basic features of the library are included in the `examples` folder.
139
+
The framework runs on Linux systems and requires at least a host compiler supporting C++14. To use NVidia's GPUs, CUDA 9.2 or higher is required.
140
+
A suite of examples demonstrating the basic features of the framework is included in the `examples` folder.
128
141
All the examples are organized in .inl files, which implements the `main()` function. These files are included by .cpp and .cu
129
142
files, which are compiled according with the availability of backends. TBB and CUDA backends requires the installation of the corresponding libraries and runtimes.
130
143
These code samples uses, but does not requires [ROOT](https://root.cern.ch/) for graphics, and [TCLAP](http://tclap.sourceforge.net/) library for process command line arguments.
144
+
Some functionality in Hydra requires Eigen, GSL, CuFFT and FFTW.
131
145
132
146
Examples
133
147
--------
@@ -141,7 +155,7 @@ The examples are built using [CMAKE](https://cmake.org/) with the following inst
141
155
6.`make`
142
156
143
157
144
-
The compiled examples will be placed in the build/examples folder. The sub-directories are named according to the functionalities they ilustrates.
158
+
The compiled examples will be placed in the build/examples folder. The sub-directories are named according to the functionalities they illustrate.
145
159
146
160
The examples are listed below:
147
161
@@ -151,10 +165,12 @@ The examples are listed below:
9.__root_macros__ : macros to run examples in ROOT
156
172
157
-
Each compiled example executable will have an postfix (ex.: _cuda, _omp, _tbb) to indicate the deployed device backend.
173
+
Each compiled example executable will have an postfix (ex.:_cpp,_cuda, _omp, _tbb) to indicate the deployed device backend.
158
174
All examples use CPP as host backend.
159
175
160
176
@@ -167,6 +183,24 @@ Recent publications and presentations at conferences and workshops
167
183
4.[A. A. Alves Junior, *Hydra* - HSF-HEP analysis ecosystem workshop, 22-24 May 2017 Amsterdam, Netherlands](https://indico.cern.ch/event/613842/)
168
184
5.[A. A. Alves Junior, *MCBooster and Hydra: two libraries for high performance computing and data analysis in massively parallel platforms* -Perspectives of GPU computing in Science September 2016, Rome, Italy](http://www.roma1.infn.it/conference/GPU2016/pdf/talks/AlvesJr.pdf)
0 commit comments