horn-dev mailing list archives

Site index · List index
Message view « Date » · « Thread »
Top « Date » · « Thread »
From edwardy...@apache.org
Subject incubator-horn git commit: Update readme directly
Date Tue, 17 Nov 2015 09:47:34 GMT
Repository: incubator-horn
Updated Branches:
  refs/heads/master 3d06e5de8 -> c10b6f67d

Update readme directly

Project: http://git-wip-us.apache.org/repos/asf/incubator-horn/repo
Commit: http://git-wip-us.apache.org/repos/asf/incubator-horn/commit/c10b6f67
Tree: http://git-wip-us.apache.org/repos/asf/incubator-horn/tree/c10b6f67
Diff: http://git-wip-us.apache.org/repos/asf/incubator-horn/diff/c10b6f67

Branch: refs/heads/master
Commit: c10b6f67dc9bd2e4c67cfa32e38f05ebb9a91469
Parents: 3d06e5d
Author: Edward J. Yoon <edwardyoon@apache.org>
Authored: Tue Nov 17 18:47:19 2015 +0900
Committer: Edward J. Yoon <edwardyoon@apache.org>
Committed: Tue Nov 17 18:47:19 2015 +0900

 README.md | 21 ++++++++++++++++++++-
 1 file changed, 20 insertions(+), 1 deletion(-)

diff --git a/README.md b/README.md
index c155637..fd3ca9b 100644
--- a/README.md
+++ b/README.md
@@ -1,7 +1,26 @@
 # Apache Horn
-Apache Horn is a neuron-centric programming model and execution framework, inspired by Google's
DistBelief, supports both data and model parallelism for training large models with massive
datasets, currently in Incubator status.
+The Apache Horn is an Apache Incubating project, a neuron-centric programming model and Sync
and Async hybrid distributed training framework, supports both data and model parallelism
for training large models with massive datasets. Unlike most systems having matrix approach
to neural network training, Horn adopted the the neuron-centric model which enables training
large-scale deep learning on highly scalable CPU cluster. In the future, we plan also to support
GPU accelerations for heterogeneous devices.
+## Tensor vs. Neuron
+While tensor-based models would require an large memory consumption or parallel computational
complexity to calibrate a large number of model parameters, the neuron-centric model has advantages
like below:
+ * More intuitive programming APIs
+ * An effective partition and parallelization strategy for large model
+ * Easy to understand how groups of neurons communicate 
+|             | Tensor           | Neuron  |
+| ------------- |:-------------:|:-----:|
+| Computation model	| tensor/matrix-based computation model | neuron-based iterative computation
model |
+| Partitioning models | Vector or Submatrix (block) | Subgraph components (densely connected
areas) |
+| Communication overhead | Large |  Small |
+## High Scalability
+The Apache Horn is an Sync and Async hybrid distributed training framework. Within single
BSP job, each task group works asynchronously using region barrier synchronization instead
of global barrier synchronization, and trains large-scale neural network model using assigned
data sets in synchronous way.
 ## Getting Involved
 Horn is an open source volunteer project under the Apache Software Foundation. We encourage
you to learn about the project and contribute your expertise.

View raw message