Search for: All records

Foundation models have superior performance across a wide array of machine learning tasks. The training of these models typically involves model parallelism (MP) to navigate the constraints of GPU memory capacity. However, MP strategies involve transmitting model activations between GPUs, which can hinder training speed in large clusters. Previous research has examined gradient compression in data-parallel contexts, but its applicability in MP settings remains largely unexplored. In this paper, we investigate the unique characteristics of compression in MP and study why strategies from gradient compression might not be directly applicable to MP scenarios. Subsequently, to systematically understand the capabilities and limitations of Model Parallelism Compression, we present a benchmarking framework MCBench. MCBench not only includes four major categories of compression algorithms but also includes several widely used models spanning language and vision tasks on a well-established distributed training framework, Megatron-LM. We initiate the first comprehensive empirical study by using MCBench. Our empirical study encompasses both the fine-tuning and pre-training of FMs. We probe over 200 unique training configurations and present results using 10 widely used datasets. To comprehend the scalability of compression advantages with the expansion of model size and cluster size, we propose a novel cost model designed specifically for training with MP compression. The insights derived from our findings can help direct the future development of new MP compression algorithms for distributed training. Our code is available at https://github.com/uw-mad-dash/MCBench 

The recent progress of AI can be largely attributed to large language models (LLMs). However, their escalating memory requirements introduce challenges for machine learning (ML) researchers and engineers. Addressing this requires developers to partition a large model to distribute it across multiple GPUs or TPUs. This necessitates considerable coding and intricate configuration efforts with existing model parallel tools, such as Megatron-LM, DeepSpeed, and Alpa. These tools require users’ expertise in machine learning systems (MLSys), creating a bottleneck in LLM development, particularly for developers without MLSys background. In this work, we present RedCoast (Redco), a lightweight and user-friendly tool crafted to automate distributed training and inference for LLMs, as well as to simplify ML pipeline development. The design of Redco emphasizes two key aspects. Firstly, to automate model parallelism, our study identifies two straightforward rules to generate tensor parallel strategies for any given LLM. Integrating these rules into Redco facilitates effortless distributed LLM training and inference, eliminating the need of additional coding or complex configurations. We demonstrate the effectiveness by applying Redco on a set of LLM architectures, such as GPT-J, LLaMA, T5, and OPT, up to the size of 66B. Secondly, we propose a mechanism that allows for the customization of diverse ML pipelines through the definition of merely three functions, avoiding redundant and formulaic code like multi-host related processing. This mechanism proves adaptable across a spectrum of ML algorithms, from foundational language modeling to complex algorithms like meta-learning and reinforcement learning. As a result, Redco implementations exhibit significantly fewer lines of code compared to their official counterparts. RedCoast (Redco) has been released under Apache 2.0 license at https://github.com/tanyuqian/redco. 

We present Memtrade, the first practical marketplace for disaggregated memory clouds. Clouds introduce a set of unique challenges for resource disaggregation across different tenants, including resource harvesting, isolation, and matching. Memtrade allows producer virtual machines (VMs) to lease both their unallocated memory and allocated-but-idle application memory to remote consumer VMs for a limited period of time. Memtrade does not require any modifications to host-level system software or support from the cloud provider. It harvests producer memory using an application-aware control loop to form a distributed transient remote memory pool with minimal performance impact; it employs a broker to match producers with consumers while satisfying performance constraints; and it exposes the matched memory to consumers through different abstractions. As a proof of concept, we propose two such memory access interfaces for Memtrade consumers -- a transient KV cache for specified applications and a swap interface that is application-transparent. Our evaluation using real-world cluster traces shows that Memtrade provides significant performance benefit for consumers (improving average read latency up to 2.8X) while preserving confidentiality and integrity, with little impact on producer applications (degrading performance by less than 2.1%).

 

Federated learning is a promising paradigm that allows multiple clients to collaboratively train a model without sharing the local data. However, the presence of heterogeneous devices in federated learning, such as mobile phones and IoT devices with varying memory capabilities, would limit the scale and hence the performance of the model could be trained. The mainstream approaches to address memory limitations focus on width-slimming techniques, where different clients train subnetworks with reduced widths locally and then the server aggregates the subnetworks. The global model produced from these methods suffers from performance degradation due to the negative impact of the actions taken to handle the varying subnetwork widths in the aggregation phase. In this paper, we introduce a memory-adaptive depth-wise learning solution in FL called FEDEPTH, which adaptively decomposes the full model into blocks according to the memory budgets of each client and trains blocks sequentially to obtain a full inference model. Our method outperforms state-of-the-art approaches, achieving 5% and more than 10% improvements in top-1 accuracy on CIFAR-10 and CIFAR-100, respectively. We also demonstrate the effectiveness of depth-wise fine-tuning on ViT. Our findings highlight the importance of memory-aware techniques for federated learning with heterogeneous devices and the success of depth-wise training strategy in improving the global model’s performance. 

Seeds provide the basis of genetic diversity in perennial grassland communities and their traits may influence ecosystem resistance to extreme drought. However, we know little about how drought effects the community functional composition of seed traits and the corresponding implications for ecosystem resistance to drought.

We experimentally removed 66% of growing season precipitation for 4 years across five arid and semi‐arid grasslands in northern China and assessed how this multi‐year drought impacted community‐weighted means (CWMs) of seed traits, seed trait functional diversity and above‐ground net primary productivity (ANPP).

Experimental drought had limited effects on CWM traits and the few effects that did occur varied by site and year. For three separate sites, and in different years, drought reduced seed length and phosphorus content but increased both seed and seed‐coat thickness. Additionally, drought led to increased seed functional evenness, divergence, dispersion and richness, but only in some sites, and mostly in later years following cumulative effects of water limitation. However, we observed a strong negative relationship between drought‐induced reductions in ANPP and CWMs of seed‐coat thickness, indicating that a high abundance of dominant species with thick seeds may increase ecosystem resistance to drought. Seed trait functional diversity was not significantly predictive of ANPP, providing little evidence for a diversity effect.

Our results suggest that monitoring community composition with a focus on seed traits may provide a valuable indicator of ecosystem resistance to future droughts despite inconsistent responses of seed trait composition overall. This highlights the importance of developing a comprehensive seed and reproductive traits database for arid and semi‐arid grassland biomes.

Read the freePlain Language Summaryfor this article on the Journal blog.

 

Large neural networks can be pruned to a small fraction of their original size, with little loss in accuracy, by following a time-consuming "train, prune, re-train" approach. Frankle & Carbin conjecture that we can avoid this by training lottery tickets, i.e., special sparse subnetworks found at initialization, that can be trained to high accuracy. However, a subsequent line of work presents concrete evidence that current algorithms for finding trainable networks at initialization, fail simple baseline comparisons, e.g., against training random sparse subnetworks. Finding lottery tickets that train to better accuracy compared to simple baselines remains an open problem. In this work, we resolve this open problem by proposing Gem-Miner which finds lottery tickets at initialization that beat current baselines. Gem-Miner finds lottery tickets trainable to accuracy competitive or better than Iterative Magnitude Pruning (IMP), and does so up to 19x faster. 

Distributed model training suffers from communication bottlenecks due to frequent model updates transmitted across compute nodes. To alleviate these bottlenecks, practitioners use gradient compression techniques like sparsification, quantization, low rank updates etc. The techniques usually require choosing a static compression ratio, often requiring users to balance the trade-off between model accuracy and per-iteration speedup. In this work, we show that such performance degradation due to choosing a high compression ratio is not fundamental and that an adaptive compression strategy can reduce communication while maintaining final test accuracy.Inspired by recent findings on critical learning regimes, in which small gradient errors can have irrecoverable impact on model performance, we propose ACCORDION a simple yet effective adaptive compression algorithm. While ACCORDION maintains a high enough compression rate on average, it avoids detrimental impact by not compressing gradients too much whenever in critical learning regimes, detected by a simple gradient-norm based criterion. Our extensive experimental study over a number of machine learning tasks in distributed environments indicates that ACCORDION, maintains similar model accuracy to uncompressed training, yet achieves up to 5.5×better compression and up to 4.1×end-to-end speedup over static approaches. We show that ACCORDION also works for adjusting the batch size, another popular strategy for alleviating communication bottlenecks. Our code is available at https://github.com/uw-mad-dash/Accordion 

Federated learning allows edge devices to collaboratively learn a shared model while keeping the training data on device, decoupling the ability to do model training from the need to store the data in the cloud. We propose the Federated matched averaging (FedMA) algorithm designed for federated learning of modern neural network architectures e.g. convolutional neural networks (CNNs) and LSTMs. FedMA constructs the shared global model in a layer-wise manner by matching and averaging hidden elements (i.e. channels for convolution layers; hidden states for LSTM; neurons for fully connected layers) with similar feature extraction signatures. Our experiments indicate that FedMA not only outperforms popular state-of-the-art federated learning algorithms on deep CNN and LSTM architectures trained on real world datasets, but also reduces the overall communication burden.