NVIDIA HierarchicalKV(Beta)

About HierarchicalKV

HierarchicalKV is a part of NVIDIA Merlin and provides hierarchical key-value storage to meet RecSys requirements.

The key capability of HierarchicalKV is to store key-value (feature-embedding) on high-bandwidth memory (HBM) of GPUs and in host memory.

You can also use the library for generic key-value storage.

Benefits of HierarchicalKV

When building large recommender systems, machine learning (ML) engineers face the following challenges:

• GPUs are needed, but HBM on a single GPU is too small for the large DLRMs that scale to several terabytes.
• Improving communication performance is getting more difficult in larger and larger CPU clusters.
• It is difficult to efficiently control consumption growth of limited HBM with customized strategies.
• Most generic key-value libraries provide low HBM and host memory utilization.

HierarchicalKV alleviates these challenges and helps the machine learning engineers in RecSys with the following benefits:

• Supports training large RecSys models on HBM and host memory at the same time.
• Provides better performance by full bypassing CPUs and reducing the communication workload.
• Implements table-size restraint strategies that are based on LRU or customized strategies. The strategies are implemented by CUDA kernels.
• Operates at a high working-status load factor that is close to 1.0.

HierarchicalKV makes NVIDIA GPUs more suitable for training large and super-large models of search, recommendations, and advertising. The library simplifies the common challenges to building, evaluating, and serving sophisticated recommenders models.

API Documentation

The main classes and structs are below, and it's recommended to read the comments in the source code directly:

• class HashTable
• class EvictStrategy
• struct HashTableOptions
• Struct HashTable::Vector

For regular API doc, please refer to API Docs

Usage restrictions

• The key_type and meta_type must be uint64_t.
• The keys of 0xFFFFFFFFFFFFFFFC, 0xFFFFFFFFFFFFFFFD, 0xFFFFFFFFFFFFFFFE, and 0xFFFFFFFFFFFFFFFF are reserved for internal using.

Contributors

HierarchicalKV is co-maintianed by NVIDIA Merlin Team and NVIDIA product end-users, and also open for public contributions, bug fixes, and documentation. [Contribute]

How to build

Basically, HierarchicalKV is a headers only library, the commands below only create binaries for benchmark and unit testing.

with cmake

git clone --recursive https://github.com/NVIDIA-Merlin/HierarchicalKV.git
cd HierarchicalKV && mkdir -p build && cd build
cmake -DCMAKE_BUILD_TYPE=Release -Dsm=80 .. && make -j

For Benchmark:

./merlin_hashtable_benchmark

For Unit Test:

./merlin_hashtable_test

with bazel

git clone --recursive https://github.com/NVIDIA-Merlin/HierarchicalKV.git
cd HierarchicalKV && bazel build --config=cuda //...

For Benchmark:

./benchmark_util

Your environment must meet the following requirements:

• CUDA version >= 11.2
• NVIDIA GPU with compute capability 8.0, 8.6, 8.7 or 9.0

Benchmark & Performance(W.I.P)

• GPU: 1 x NVIDIA A100 80GB PCIe: 8.0
• Key Type = uint64_t
• Value Type = float32 * {dim}
• Key-Values per OP = 1048576
• Hit rate = 0.60
• find* means the find API that directly returns the addresses of values.
• Throughput Unit: Billion-KV/second

On pure HBM mode:

• dim = 4, capacity = 64 Million-KV, HBM = 32 GB, HMEM = 0 GB

load_factor	insert_or_assign	find	find_or_insert	assign	find*	insert_and_evict
0.50	1.402	2.958	1.743	1.954	3.632	1.178
0.75	1.072	1.629	0.617	0.914	1.851	0.906
1.00	0.352	0.826	0.342	0.552	0.895	0.303

• dim = 64, capacity = 64 Million-KV, HBM = 16 GB, HMEM = 0 GB

load_factor	insert_or_assign	find	find_or_insert	assign	find*	insert_and_evict
0.50	0.925	1.584	0.890	1.128	3.645	0.795
0.75	0.665	1.115	0.541	0.834	1.849	0.569
1.00	0.323	0.640	0.314	0.512	0.896	0.179

On HBM+HMEM hybrid mode:

• dim = 64, capacity = 128 Million-KV, HBM = 16 GB, HMEM = 16 GB

load_factor	insert_or_assign	find	find_or_insert	assign	find*
0.50	0.121	0.149	0.120	0.147	3.397
0.75	0.116	0.145	0.115	0.143	1.800
1.00	0.087	0.125	0.087	0.114	0.883

• dim = 64, capacity = 1024 Million-KV, HBM = 56 GB, HMEM = 200 GB

load_factor	insert_or_assign	find	find_or_insert	assign	find*
0.50	0.036	0.054	0.035	0.045	2.809
0.75	0.035	0.055	0.034	0.047	1.930
1.00	0.034	0.051	0.031	0.047	0.855

Support and Feedback:

If you encounter any issues or have questions, go to https://github.com/NVIDIA-Merlin/HierarchicalKV/issues and submit an issue so that we can provide you with the necessary resolutions and answers.

Acknowledgment

We are very grateful to external initial contributors @Zhangyafei and @Lifan for their design, coding, and review work.

License

Apache License 2.0