An official website of the United States government
Writing large amounts of data concurrently to stable storage is a typical I/O pattern of many HPC workflows. This pattern introduces high I/O overheads and results in increased storage space utilization especially for workflows that need to capture the evolution of data structures with high frequency as checkpoints. In this context, many applications, such as graph pattern matching, perform sparse updates to large data structures between checkpoints. For these applications, incremental checkpointing techniques that save only the differences from one checkpoint to another can dramatically reduce the checkpoint sizes, I/O bottlenecks, and storage space utilization. However, such techniques are not without challenges: it is non-trivial to transparently determine what data has changed since a previous checkpoint and assemble the differences in a compact fashion that does not result in excessive metadata. State-of-art data reduction techniques (e.g., compression and de-duplication) have significant limitations when applied to modern HPC applications that leverage GPUs: slow at detecting the differences, generate a large amount of metadata to keep track of the differences, and ignore crucial spatiotemporal checkpoint data redundancy. This paper addresses these challenges by proposing a Merkle tree-based incremental checkpointing method to exploit GPUs’ high memory bandwidth and massive parallelism. Experimental results at scale show a significant reduction of the I/O overhead and space utilization of checkpointing compared with state-of-the-art incremental checkpointing and compression techniques.
more »
« less
DeepFreeze: Towards Scalable Asynchronous Checkpointing of Deep Learning Models
In the age of big data, deep learning has emerged as a powerful tool to extract insight and exploit its value, both in industry and scientific applications. One common pattern emerging in such applications is frequent checkpointing of the state of the learning model during training, needed in a variety of scenarios: analysis of intermediate states to explain features and correlations with training data, exploration strategies involving alternative models that share a common ancestor, knowledge transfer, resilience, etc. However, with increasing size of the learning models and popularity of distributed data-parallel training approaches, simple checkpointing techniques used so far face several limitations: low serialization performance, blocking I/O, stragglers due to the fact that only a single process is involved in checkpointing. This paper proposes a checkpointing technique specifically designed to address the aforementioned limitations, introducing efficient asynchronous techniques to hide the overhead of serialization and I/O, and distribute the load over all participating processes. Experiments with two deep learning applications (CANDLE and ResNet) on a pre-Exascale HPC platform (Theta) shows significant improvement over state-of-art, both in terms of checkpointing duration and runtime overhead.
more »
« less
- Award ID(s):
- 1664142
- PAR ID:
- 10393365
- Date Published:
- Journal Name:
- 20th IEEE/ACM International Symposium on Cluster, Cloud and Internet Computing (CCGRID)
- Page Range / eLocation ID:
- 172 to 181
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Aguilera, Marcos; Yadgar, Gala (Ed.)Training Deep Neural Networks (DNNs) is a resource-hungry and time-consuming task. During training, the model performs computation at the GPU to learn weights, repeatedly, over several epochs. The learned weights reside in GPU memory, and are occasionally checkpointed (written to persistent storage) for fault-tolerance. Traditionally, model parameters are checkpointed at epoch boundaries; for modern deep networks, an epoch runs for several hours. An interruption to the training job due to preemption, node failure, or process failure, therefore results in the loss of several hours worth of GPU work on recovery. We present CheckFreq, an automatic, fine-grained checkpointing framework that (1) algorithmically determines the checkpointing frequency at the granularity of iterations using systematic online profiling, (2) dynamically tunes checkpointing frequency at runtime to bound the checkpointing overhead using adaptive rate tuning, (3) maintains the training data invariant of using each item in the dataset exactly once per epoch by checkpointing data loader state using a light-weight resumable iterator, and (4) carefully pipelines checkpointing with computation to reduce the checkpoint cost by introducing two-phase checkpointing. Our experiments on a variety of models, storage backends, and GPU generations show that CheckFreq can reduce the recovery time from hours to seconds while bounding the runtime overhead within 3.5%.more » « less
-
With the emergence of versatile storage systems, multi-level checkpointing (MLC) has become a common approach to gain efficiency. However, multi-level checkpoint/restart can cause enormous I/O traffic on HPC systems. To use multilevel checkpointing efficiently, it is important to optimize checkpoint/restart configurations. Current approaches, namely modeling and simulation, are either inaccurate or slow in determining the optimal configuration for a large scale system. In this paper, we show that machine learning models can be used in combination with accurate simulation to determine the optimal checkpoint configurations. We also demonstrate that more advanced techniques such as neural networks can further improve the performance in optimizing checkpoint configurations.more » « less
-
Microsecond I/O will make data serialization a major bottleneck for datacenter applications. Serialization is fundamentally about data movement: serialization libraries coalesce and flatten in-memory data structures into a single transmittable buffer. CPU-based serialization approaches will hit a performance limit due to data movement overheads and be unable to keep up with modern networks. We observe that widely deployed NICs possess scatter-gather capabilities that can be re-purposed to accelerate serialization's core task of coalescing and flattening in-memory data structures. It is possible to build a completely zero-copy, zero-allocation serialization library with commodity NICs. Doing so introduces many research challenges, including using the hardware capabilities efficiently for a wide variety of non-uniform data structures, making application memory available for zero-copy I/O, and ensuring memory safety.more » « less
-
null (Ed.)Deep learning has been shown as a successful method for various tasks, and its popularity results in numerous open-source deep learning software tools. Deep learning has been applied to a broad spectrum of scientific domains such as cosmology, particle physics, computer vision, fusion, and astrophysics. Scientists have performed a great deal of work to optimize the computational performance of deep learning frameworks. However, the same cannot be said for I/O performance. As deep learning algorithms rely on big-data volume and variety to effectively train neural networks accurately, I/O is a significant bottleneck on large-scale distributed deep learning training. This study aims to provide a detailed investigation of the I/O behavior of various scientific deep learning workloads running on the Theta supercomputer at Argonne Leadership Computing Facility. In this paper, we present DLIO, a novel representative benchmark suite built based on the I/O profiling of the selected workloads. DLIO can be utilized to accurately emulate the I/O behavior of modern scientific deep learning applications. Using DLIO, application developers and system software solution architects can identify potential I/O bottlenecks in their applications and guide optimizations to boost the I/O performance leading to lower training times by up to 6.7x.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Conference Paper:
- https://doi.org/10.1109/CCGrid49817.2020.00-76
