Production LLM serving must simultaneously deliver high throughput, low latency, and sufficient context capacity under non-stationary traffic and mixed request requirements. Data parallelism (DP) maximizes throughput by running independent replicas, while tensor parallelism (TP) reduces per-request latency and pools memory for long-context inference. However, existing serving stacks typically commit to a static parallelism configuration at deployment; adapting to bursts, priorities, or long-context requests is often disruptive and slow. We present Flying Serving, a vLLM-based system that enables online DP-TP switching without restarting engine workers. Flying Serving makes reconfiguration practical by virtualizing the state that would otherwise force data movement: (i) a zero-copy Model Weights Manager that exposes TP shard views on demand, (ii) a KV Cache Adaptor that preserves request KV state across DP/TP layouts, (iii) an eagerly initialized Communicator Pool to amortize collective setup, and (iv) a deadlock-free scheduler that coordinates safe transitions under execution skew. Across three popular LLMs and realistic serving scenarios, Flying Serving improves performance by up to 4.79 × under high load and 3.47 × under low load while supporting latency- and memory-driven requests.

Distributed Breadth-First Search (BFS) is fundamental to many large-scale graph applications, but its performance on parallel systems is often limited by high communication overhead. This paper presents \({\mathrm\small {CORE-BFS}}\), an extremely scalable GPU-based BFS implementation that introduces a unique rectangular 2D partitioning-based design for Frontier supercomputer. To further improve performance, we propose four key optimizations: (1) Rectangular 2D-partition specific data formats that use two compressed row and one compressed column status array bitmaps combined with a Double Compressed Sparse Row (DCSR) format per partition, reducing memory footprint and inter-rank traffic; (2) Adaptive frontier & communication strategy that unifies top-down and bottom-up traversal on the rectangular layout, uses lazy synchronization in top-down levels, and switches variants based on frontier size to minimize communication overhead; (3) Frontier-split degree-aware update that maps frontier vertices to thread-centric, wavefront-centric, and block-centric kernels based on their degree to improve GPU utilization and memory coalescing; (4) Row-reduction pipeline that overlaps bottom-up adjacency list processing with row-wise bitmap reduction to hide inter-rank latency. Together, these techniques increase parallelism while reducing memory and communication overhead. On the Graph500 benchmark, \({\mathrm\small {CORE-BFS}}\) scales up to 9,248 Frontier nodes with scale-42 graphs and reaches 160.845 TTEPS, delivering a 5.42 × speedup over our previous Frontier implementation.

Turbulence plays a crucial role in multiphysics applications, including aerodynamics, fusion, and combustion. Accurately capturing turbulence's multiscale characteristics is essential for reliable predictions of multiphysics interactions, but remains a grand challenge even for exascale supercomputers and advanced deep learning models. The extreme-resolution data required to represent turbulence, ranging from billions to trillions of grid points, pose prohibitive computational costs for models based on architectures like vision transformers. To address this challenge, we introduce a multiscale hierarchical Turbulence Transformer that reduces sequence length from billions to a few millions and a novel RingX sequence parallelism approach that enables scalable long-context learning. We perform scaling and science runs on the Frontier supercomputer. Our approach demonstrates excellent performance up to 1.1 EFLOPS on 32,768 AMD GPUs, with a scaling efficiency of 94%. To our knowledge, this is the first AI model for turbulence that can capture small-scale eddies down to the dissipative range.

Abstract The explosive demand for artificial intelligence (AI) workloads has led to a significant increase in silicon area dedicated to lower-precision computations on recent high-performance computing hardware designs. However, mixed-precision capabilities, which can achieve performance improvements of up to 8 $$\times$$ × compared to double-precision in extreme compute-intensive workloads, remain largely untapped in most scientific applications. A growing number of efforts have shown that mixed-precision algorithmic innovations can deliver superior performance without sacrificing accuracy. These developments should prompt computational scientists to seriously consider whether their scientific modeling and simulation applications could benefit from the acceleration offered by new hardware and mixed-precision algorithms. In this survey, we (1) review progress across diverse scientific domains—fluid dynamics, weather and climate, quantum chemistry, and computational genomics—that have begun adopting mixed-precision strategies; (2) examine state-of-the-art algorithmic techniques such as iterative refinement, splitting and emulation schemes, and adaptive precision solvers; (3) assess their implications for accuracy, performance, and resource utilization; and (4) survey the emerging software ecosystem that enables mixed-precision methods at scale. We conclude with perspectives and recommendations on cross-cutting opportunities, domain-specific challenges, and the role of co-design between application scientists, numerical analysts, and computer scientists. Collectively, this survey underscores that mixed-precision numerics can reshape computational science by aligning algorithms with the evolving landscape of hardware capabilities.

Large Language Models (LLMs) have shown strong capabilities in document re-ranking, a key component in modern Information Retrieval (IR) systems.However, existing LLMbased approaches face notable limitations, including ranking uncertainty, unstable top-k recovery, and high token cost due to tokenintensive prompting.To effectively address these limitations, we propose REALM, an uncertainty-aware re-ranking framework that models LLM-derived relevance as Gaussian distributions and refines them through recursive Bayesian updates.By explicitly capturing uncertainty and minimizing redundant queries, REALM achieves better rankings more efficiently.Experimental results demonstrate that our REALM surpasses state-of-the-art rerankers while significantly reducing token usage and latency, improving NDCG@10 by 0.7 -11.9 and simultaneously reducing the number of LLM inferences by 23.4 -84.4%, promoting it as the next-generation re-ranker for modern IR systems.

Diffusion models have emerged as powerful generative models, but their high computational cost in iterative sampling remains a significant bottleneck. In this work, we present an in-depth and insightful study of state-of-the-art acceleration techniques for diffusion models, including caching and quantization, and reveal their limitations in computation error and generation quality. To break these limits, this work introduces Modulated Diffusion (MoDiff), an innovative, rigorous, and principled framework that accelerates generative modeling through modulated quantization and error compensation. MoDiff not only inherits the advantages of existing caching and quantization methods but also serves as a general framework to accelerate all diffusion models. The advantages of MoDiff are supported by solid theoretical insight and analysis. In addition, extensive experiments on CIFAR-10 and LSUN demonstrate that MoDiff significantly reduces activation quantization from 8 bits to 3 bits without performance degradation in post-training quantization (PTQ). Our code implementation is available at https://github.com/WeizhiGao/MoDiff.

High performance LINPACK (HPL) remains the primary benchmark for evaluating supercomputing performance. It includes many parts with substantial internal complexity, and its performance is affected by a large number of parameters that interact in ways that are difficult to predict on large-scale heterogeneous supercomputer systems. We present a comprehensive performance analysis of HPL on Frontier, the world’s first exascale supercomputer, which achieved HPL performance of 1.35 exaflops. Through empirical parameter tuning, detailed modeling, and comparative evaluation, we uncover critical performance insights, share lessons learned, and outline best practices for effective parameter tuning on exascale systems. We introduce and evaluate two novel PDFACT strategies: a dedicated-thread (DT) variant and a GPU-based variant (GPUPDFACT) implementation using HIP cooperative groups, demonstrating that GPU-based factorization outperforms conventional CPU-based PDFACT on Frontier’s architecture. Our findings establish key performance factors for HPL on exascale systems and offer valuable guidance for future high-performance computing and benchmarking efforts.

Mixed-precision algorithms have been proposed as a way for scientific computing to benefit from some of the gains seen for artificial intelligence (AI) on recent high performance computing (HPC) platforms. A few applications dominated by dense matrix operations have seen substantial speedups by utilizing low precision formats such as FP16. However, a majority of scientific simulation applications are memory bandwidth limited. Beyond preliminary studies, the practical gain from using mixed-precision algorithms on a given HPC system is largely unclear.

Vision-based scientific foundation models hold significant promise for advancing scientific discovery and innovation. This potential stems from their ability to aggregate images from diverse sources—such as varying physical groundings or data acquisition systems—and to learn spatio-temporal correlations using transformer architectures. However, tokenizing and aggregating images can be compute-intensive, a challenge not fully addressed by current distributed methods. In this work, we introduce the Distributed Cross-Channel Hierarchical Aggregation (D-CHAG) approach designed for datasets with a large number of channels across image modalities. Our method is compatible with any model-parallel strategy and any type of vision transformer architecture, significantly improving computational efficiency. We evaluated D-CHAG on hyperspectral imaging and weather forecasting tasks. When integrated with tensor parallelism and model sharding, our approach achieved up to a 75% reduction in memory usage and more than doubled sustained throughput on up to 1,024 AMD GPUs on the Frontier Supercomputer.

The attention mechanism has become foundational for remarkable AI breakthroughs since the introduction of the Transformer, driving the demand for increasingly longer context to power frontier models such as large-scale reasoning language models and high-resolution image/video generators. However, its quadratic computational and memory complexities present substantial challenges. Current state-of-the-art parallel attention methods, such as ring attention, are widely adopted for long-context training but utilize a point-to-point communication strategy that fails to fully exploit the capabilities of modern HPC network architectures. In this work, we propose ringX, a scalable family of parallel attention methods optimized explicitly for HPC systems. By enhancing workload partitioning, refining communication patterns, and improving load balancing, ringX achieves up to 3.4 × speedup compared to conventional ring attention on the Frontier supercomputer. Optimized for both bi-directional and causal attention mechanisms, ringX demonstrates its effectiveness through training benchmarks of a Vision Transformer (ViT) on a climate dataset and a Generative Pre-Trained Transformer (GPT) model, Llama3 8B. Our method attains an end-to-end training speedup of approximately 1.5 × in both scenarios. To our knowledge, the achieved 38% model FLOPs utilization (MFU) for training Llama3 8B with a 1M-token sequence length on 4,096 GPUs represents one of the highest training efficiencies reported for long-context learning on HPC systems. Our code implementation is available at https://github.com/jqyin/ringX-attention.

Emerging expert-specialized Mixture-of-Experts (MoE) architectures, such as DeepSeek-MoE, deliver strong model quality through fine-grained expert segmentation and large top-k routing. However, their scalability is limited by substantial activation memory overhead and costly all-to-all communication. Furthermore, current MoE training systems – primarily optimized for NVIDIA GPUs – perform suboptimally on non-NVIDIA platforms, leaving significant computational potential untapped. In this work, we present X-MoE, a novel MoE training system designed to deliver scalable training performance for next-generation MoE architectures. X-MoE achieves this via several novel techniques, including efficient padding-free MoE training with cross-platform kernels, redundancy-bypassing dispatch, and hybrid parallelism with sequence-sharded MoE blocks. Our evaluation on the Frontier supercomputer, powered by AMD MI250X GPUs, shows that X-MoE scales DeepSeek-style MoEs up to 545 billion parameters across 1024 GPUs – 10x larger than the largest trainable model with existing methods under the same hardware budget, while maintaining high training throughput.

The rapid growth of large-scale datasets in fields like biology and social networks has driven the need for advanced graph analytics techniques. Community detection, a fundamental task in graph analytics, identifies closely connected groups of nodes within a network, providing valuable insights across various disciplines. This study focuses on two classic community detection methods, the Louvain algorithm and Markov Clustering (MCL), and evaluates the performance of two prominent distributed community detection algorithms: HiPDPL-GPU, our prior implementation, and HipMCL. We conduct experiments on GPU-accelerated heterogeneous HPC systems, Summit and Frontier, to assess their performance under varying conditions. Our objective is to identify the strengths and weaknesses of these algorithms in terms of scalability, and quality of solutions. We evaluate these algorithms on a diverse set of 70+ networks spanning 13 domains, with sizes ranging up to 4.2 billion edges. Our results demonstrate that HiPDPL-GPU consistently outperforms HipMCL, especially for large-scale networks. HiPDPL-GPU achieves significantly faster runtimes (47x to 1439x), higher modularity scores, and improved scalability. These findings highlight HiPDPL-GPU as a promising solution for efficient and effective large-scale graph analytics in diverse application domains, and provide insights into the feasibility of using MCL-based approaches for certain application domains.

The power & energy demands of HPC machines have grown significantly. Modern exascale HPC systems require tens of megawatts of combined power for computing resources and cooling facilities at full capacity. The current energy trend is not sustainable for future HPC systems, and there is a need to work toward the energy efficiency aspect of HPC performance. Energy awareness of the HPC applications at the job level is essential for running an efficient HPC system. This work aims to develop a pipeline to provide a production-level system-wide overview of the HPC workloads' power profile while handling evolving workloads exhibiting new power trends. We developed an open-set classification model for HPC jobs based on the properties of power profiles to continuously provide a system-wide holistic view of recently completed jobs. The pipeline helps continuously monitor the job-level power usage pattern of HPC and enables us to capture the new trends in applications' power behavior. We employed a comprehensive set of techniques to generate job-level data, custom-designed feature extraction methods to extract critical features from jobs' power profiles, clustering techniques powered by generative modeling, and open-set classification for identifying job profiles into known classes or an unknown set. With extensive evaluations, we demonstrate the effectiveness of each component in our pipeline. We provide an analysis of the resulting clusters that characterize the power profile landscape of the Summit supercomputer from more than 60K jobs executed in a year. The open-set classification classifies the known data sets into known classes with high accuracy and identifies unknown data noints with over 85% accuracy.

Transformer-based architectures have dominated various areas of machine learning in recent years. In this paper, we introduce a novel robust attention mechanism designed to enhance the resilience of transformer-based architectures. Crucially, this technique can be integrated into existing transformers as a plug-and-play layer, improving their robustness without the need for additional training or fine-tuning. Through comprehensive experiments and ablation studies, we demonstrate that our ProTransformer significantly enhances the robustness of transformer models across a variety of prediction tasks, attack mechanisms, backbone architectures, and data domains. Notably, without further fine-tuning, the ProTransformer consistently improves the performance of vanilla transformers by 19.5\%, 28.3\%, 16.1\%, and 11.4\% for BERT, ALBERT, DistilBERT, and RoBERTa, respectively, under the classical TextFooler attack. Furthermore, ProTransformer shows promising resilience in large language models (LLMs) against prompting-based attacks, improving the performance of T5 and LLaMA by 24.8\% and 17.8\%, respectively, and enhancing Vicuna by an average of 10.4\% against the Jailbreaking attack. Beyond the language domain, ProTransformer also demonstrates outstanding robustness in both vision and graph domains.

We present ExaDigiT, an open-source framework for developing comprehensive digital twins of liquid-cooled supercomputers. It integrates three main modules: (1) a resource allocator and power simulator, (2) a transient thermo-fluidic cooling model, and (3) an augmented reality model of the supercomputer and central energy plant. The framework enables the study of “what-if” scenarios, system optimizations, and virtual prototyping of future systems. Using Frontier as a case study, we demonstrate the framework’s capabilities by replaying six months of system telemetry for systematic verification and validation. Such a comprehensive analysis of a liquid-cooled exascale supercomputer is the first of its kind. ExaDigiT elucidates complex transient cooling system dynamics, runs synthetic or real workloads, and predicts energy losses due to rectification and voltage conversion. Throughout our paper, we present lessons learned to benefit HPC practitioners developing similar digital twins. We envision the digital twin will be a key enabler for sustainable, energy-efficient supercomputing.

Digital twins are an excellent tool to model, visualize, and simulate complex systems, to understand and optimize their operation. In this work, we present the technical challenges of real-time visualization of a digital twin of the Frontier supercomputer.We show the initial prototype and current state of the twin and highlight technical design challenges of visualizing such a large High Performance Computing (HPC) system. The goal is to understand the use of augmented reality as a primary way to extract information and collaborate on digital twins of complex systems. This leverages the spatio-temporal aspect of a 3D representation of a digital twin, with the ability to view historical and real-time telemetry, triggering simulations of a system state and viewing the results, which can be augmented via dashboards for details. Finally, we discuss considerations and opportunities for augmented reality of digital twins of large-scale, parallel computers.

The imbalanced I/O load on large parallel file systems affects the parallel I/O performance of high-performance computing (HPC) applications. One of the main reasons for I/O imbalances is the lack of a global view of system-wide resource consumption. While approaches to address the problem already exist, the diversity of HPC workloads combined with different file striping patterns prevents widespread adoption of these approaches. In addition, load-balancing techniques should be transparent to client applications. To address these issues, we propose Tarazu , an end-to-end control plane where clients transparently and adaptively write to a set of selected I/O servers to achieve balanced data placement. Our control plane leverages real-time load statistics for global data placement on distributed storage servers, while our design model employs trace-based optimization techniques to minimize latency for I/O load requests between clients and servers and to handle multiple striping patterns in files. We evaluate our proposed system on an experimental cluster for two common use cases: the synthetic I/O benchmark IOR and the scientific application I/O kernel HACC-I/O. We also use a discrete-time simulator with real HPC application traces from emerging workloads running on the Summit supercomputer to validate the effectiveness and scalability of Tarazu in large-scale storage environments. The results show improvements in load balancing and read performance of up to 33% and 43%, respectively, compared to the state-of-the-art.

Cancer is a leading cause of death in the US, and it results from a combination of two-nine genetic mutations. Identifying five-hit combinations responsible for several cancer types is computationally intractable even with the fastest super-computers in the USA. Iterating through nested loops required by the process presents a simplex-shaped workload with irregular memory access patterns. Distributing this workload efficiently across thousands of GPUs offers a challenge in dividing simplex-shaped (triangular/tetrahedral) workload into similar shapes with equal volume. Irregular memory access patterns create imbalanced compute utilization across nodes. We developed a generalized solution for distributing a simplex-shaped workload by partially coalescing the nested for-loops, minimizing the memory access overhead by efficiently utilizing limited shared memory, a dynamic scheduler, and loop tiling. For 4-hit combinations, we achieved a 90% − 100% strong scaling efficiency for up to 3594 V100 GPUs on the Summit supercomputer. Finally, we designed and implemented a distributed algorithm to identify 5-hit combinations for four different cancer types, and the identified combinations can differentiate between cancer and normal samples with 86.59−88.79% precision and 84.42 − 90.91% recall. We also demonstrated the robustness of our solution by porting the code to another leadership class computing platform Crusher, a testbed for the fastest supercomputer Frontier. On Crusher, we achieved 98% strong scaling efficiency on 50 nodes (400 AMD MI250X GCDs) and demonstrated the computational readiness of Frontier for scientific applications.

Since the introduction of Metropolis Monte Carlo (MC) sampling, it and its variants have become standard tools used for thermodynamics evaluations of physical systems. However, a long-standing problem that hinders the effectiveness and efficiency of MC sampling is the lack of a generic method (a.k.a. MC proposal) to update the system configurations. Consequently, current practices are not scalable. Here we propose a parallel MC sampling framework for thermodynamics evaluation—DeepThermo. By using deep learning–based MC proposals that can globally update the system configurations, we show that DeepThermo can effectively evaluate the phase transition behaviors of high entropy alloys, which have an astronomical configuration space. For the first time, we directly evaluate a density of states expanding over a range of ~e10,000for a real material. We also demonstrate DeepThermo’s performance and scalability up to 3,000 GPUs on both NVIDIA V100 and AMD MI250X-based supercomputers.

This paper explores the potential opportunities for implementing file prefetching techniques on large-scale high-performance computing (HPC) systems. Specifically, we investigate the file access patterns of various applications across multiple scientific domains using two years' worth of Darshan I/O traces obtained from the Summit supercomputer. We identify recurring trends and patterns which indicate that prefetching can be effectively leveraged to improve data access performance on HPC systems. This study serves as a valuable reference for system architects and developers in the HPC community, providing insights into the opportunities and challenges associated with enabling file prefetching on large-scale HPC systems.

The new reality of exascale computing faces many challenges in achieving optimal performance on large numbers of nodes. A key challenge is the efficient utilization of the message-passing interface (MPI), a critical component for process communication. This paper explores communication optimization strategies to harness the GPU-accelerated architectures of these supercomputers. We focus on MPI applications where processors form a two-dimensional process grid, a common arrangement in applications involving dense matrix operations. This configuration offers a unique opportunity to implement innovative strategies to improve performance and maintain effective load distribution. We study two applications— Dist-FW (Apsp:all-pair-shortest-path) and HPL-MxP (LU factorization with Mixed precision)—on two accelerated systems: Summit (IBM Power and NVIDIA V100) and Frontier (AMD EPYC and MI250X). These supercomputers are operated by the Oak Ridge Leadership Computing Facility (OLCF) and are currently ranked #1 and #5 on the Top500 list. We show how to scale up both applications to exascale levels and tackle the MPI challenges related to implementation, synchronization, and performance. We also compare the performance of several communication strategies at an unprecedented scale. Accurately predicting application performance becomes crucial for cost reduction as the computational scale grows. To address this, we suggest a hyperbolic model as a better alternative to the traditional one-sided asymptotic model for predicting future application performance at such large scales.

Large language models (LLMs) are poised to revolutionize the way we conduct scientific research. However, both model complexity and pre-training cost are impeding effective adoption for the wider science community. Identifying suitable scientific use cases, finding the optimal balance between model and data sizes, and scaling up model training are among the most pressing issues that need to be addressed. In this study, we provide practical solutions for building and using LLM-based foundation models targeting scientific research use cases. We present an end-to-end examination of the effectiveness of LLMs in scientific research, including their scaling behavior and computational requirements on Frontier, the first Exascale supercomputer. We have also developed for release to the scientific community a suite of open foundation models called FORGE with up to 26B parameters using 257B tokens from over 200M scientific articles, with performance either on par or superior to other state-of-the-art comparable models. We have demonstrated the use and effectiveness of FORGE on scientific downstream tasks. Our research establishes best practices that can be applied across various fields to take advantage of LLMs for scientific discovery.

Scientific communities are increasingly adopting deep learning (DL) models in their applications to accelerate scientific discovery processes. However, with rapid growth in the computing capabilities of HPC supercomputers, large-scale DL applications have to spend a significant portion of training time performing I/O to a parallel storage system. Previous research works have investigated optimization techniques such as prefetching and caching. Unfortunately, there exist non-trivial challenges to adopting the existing solutions on HPC supercomputers for large-scale DL training applications, which include non-performance and/or failures at extreme scale, lack of portability and generality in design, complex deployment methodology, and being limited to a specific application or dataset. To address these challenges, we propose High-Velocity AI Cache (HVAC), a distributed read-cache layer that targets and fully exploits the node-local storage or near node-local storage technology. HVAC seamlessly accelerates read I/O by aggregating node-local or near node-local storage, avoiding metadata lookups and file locking while preserving portability in the application code. We deploy and evaluate HVAC on 1,024 nodes (with over 6000 NVIDIA V100 GPUS) of the Summit supercomputer. In particular, we evaluate the scalability, efficiency, accuracy, and load distribution of HVAC compared to GPFS and XFS-on-NVMe. With four different DL applications, we observe an average 25 % performance improvement atop GPFS and 9% drop against XFS-on-NVMe, which scale linearly and are considered the performance upper bound. We envision HVAC as an important caching library for upcoming HPC supercomputers such as Frontier.

Scientific computing workloads at HPC facilities have been shifting from traditional numerical simulations to AI/ML applications for training and inference while processing and producing ever-increasing amounts of scientific data. To address the growing need for increased storage capacity, lower access latency, and higher bandwidth, emerging technologies such as non-volatile memory are integrated into supercomputer I/O subsystems. With these emerging trends, we need a better understanding of the multilayer supercomputer I/O systems and ways to use these subsystems efficiently. In this work, we study the I/O access patterns and performance characteristics of two representative supercomputer I/O subsystems. Through an extensive analysis of year-long I/O logs on each system, we report new observations in I/O reads and writes, unbalanced use of storage system layers, and new trends in user behaviors at the HPC I/O middleware stack.

Monitoring and analyzing a wide range of I/O activities in an HPC cluster is important in maintaining mission-critical performance in a large-scale, multi-user, parallel storage system. Center-wide I/O traces can provide high-level information and fine-grained activities per application or per user running in the system. Studying such large-scale traces can provide helpful insights into the system. It can be used to develop predictive methods for making predictive decisions, adjusting scheduling policies, or providing decisions for the design of next-generation systems. However, sharing real-world I/O traces to expedite such research efforts leaves a few concerns; i) the cost of sharing the large traces is expensive due to this large size, and ii) privacy concern is an issue.

The rise of machine learning (ML) applications and their use of mixed precision to perform interesting science are driving forces behind AI for science on HPC. The convergence of ML and HPC with mixed precision offers the possibility of transformational changes in computational science. The HPL-AI benchmark is designed to measure the performance of mixed precision arithmetic as opposed to the HPL benchmark which measures double precision performance. Pushing the limits of systems at extreme scale is nontrivial -little public literature explores optimization of mixed precision computations at this scale. In this work, we demonstrate how to scale up the HPL-AI benchmark on the pre-exascale Summit and exascale Frontier systems at the Oak Ridge Leadership Computing Facility (OLCF) with a cross-platform design. We present the implementation, performance results, and a guideline of optimization strategies employed for delivering portable performance on both AMD and NVIDIA GPUs at extreme scale.

In this work, we demonstrate the challenges in predicting HPC cluster power consumption in the face of significant temporal skew in power consumption behavioral patterns. Predicting large power swings that extend several megawatts has significant operational value for HPC centers, however, prediction is challenging due to the relative rarity of such events and also due to the abrupt or disjoint deviation from the average power consumption levels. To study the impact of this challenge, we have trained a recurrent neural network (RNN) as a reasonably sophisticated model to predict power consumption of the one-year worth of node power consumption data from the Summit supercomputer located in the Oak Ridge Leadership Computing Facility. By studying the prediction results, we have found that although simple usage of RNN models can provide good results on average power consumption levels, it would fail at predicting the power swings that have more operational value. With such results, we discuss potential next steps in addressing such issues aiming towards a robust usage of power prediction techniques in HPC operations.

This work seeks to advance the state of the art in HPC I/O performance analysis and interpretation. In particular, we demonstrate effective techniques to: (1) model output performance in the presence of I/O interference from production loads; (2) build features from write patterns and key parameters of the system architecture and configurations; (3) employ suitable machine learning algorithms to improve model accuracy. We train models with five popular regression algorithms and conduct experiments on two distinct production HPC platforms. We find that the lasso and random forest models predict output performance with high accuracy on both of the target systems. We also explore use of the models to guide adaptation in I/O middleware systems, and show potential for improvements of at least 15% from model-guided adaptation on 70% of samples, and improvements up to 10 x on some samples for both of the target systems.

High performance computing (HPC) is no longer solely limited to traditional workloads such as simulation and modeling. With the increase in the popularity of machine learning (ML) and deep learning (DL) technologies, we are observing that an increasing number of HPC users are incorporating ML methods into their workflow and scientific discovery processes, across a wide spectrum of science domains such as biology, earth science, and physics. This gives rise to a diverse set of I/O patterns than the traditional checkpoint/restart-based HPC I/O behavior. The details of the I/O characteristics of such ML I/O workloads have not been studied extensively for large-scale leadership HPC systems. This paper aims to fill that gap by providing an in-depth analysis to gain an understanding of the I/O behavior of ML I/O workloads using darshan - an I/O characterization tool designed for lightweight tracing and profiling. We study the darshan logs of more than 23, 000 HPC ML I/O jobs over a time period of one year running on Summit - the second-fastest supercomputer in the world. This paper provides a systematic I/O characterization of ML I/O jobs running on a leadership scale supercomputer to understand how the I/O behavior differs across science domains and the scale of workloads, and analyze the usage of parallel file system and burst buffer by ML I/O workloads.

As we approach the exascale computing era, the focused understanding of power consumption and its overall constraint on HPC architectures and applications are becoming increasingly paramount. Summit, located at the Oak Ridge Leadership Computing Facility (OLCF), is one of the fastest and largest pre-exascale platforms in operation today. This paper provides a first-order examination and analysis of power consumption at the component-level, node-level, and system-level, from all 4,626 Summit compute nodes, each with over 100 metrics at 1Hz frequency over the entire year of 2020. We also investigate the power characteristics and energy efficiency of over 840k Summit jobs and 250k GPU failure logs for further operational insights. To the best of our knowledge, this is the first systematic analysis of power data of HPC system at this scale.

Titan was the flagship supercomputer at the Oak Ridge Leadership Computing Facility (OLCF). It was deployed in late 2012, became the fastest supercomputer in the world and was retired on August 2, 2019. With Titan's mission complete, this paper provides a first-order examination of the usage of its critical resources (CPU, Memory, GPU, and I/O) over a five-year production period (2015-2019). In particular, we show quantitatively that the majority of CPU time was spent on the large-scale jobs, which is consistent with the policy of driving ground-breaking science through leadership computing. We also corroborate the general observation of the low CPU-memory usage with 95% jobs utilizing only 15% or less available memory. Additionally, we correlate the increase of total job submissions and the decrease of GPU-enabled jobs during 2016 with the GPU reliability issue which impacted the large-scale runs. We further show the surprising read/write ratio over the five-year period, which contradicts the general mindset of the large-scale simulation machines being “write-heavy”. This understanding will have potential impact on how we design our next-generation large-scale storage systems. We believe that our analyses and findings are going to be of great interest to the high-performance computing (HPC) community at large.

Parallel I/O performance is crucial to sustaining scientific applications on large-scale High-Performance Computing (HPC) systems. However, I/O load imbalance in the underlying distributed and shared storage systems can significantly reduce overall application performance. There are two conflicting challenges to mitigate this load imbalance: (i) optimizing systemwide data placement to maximize the bandwidth advantages of distributed storage servers, i.e., allocating I/O resources efficiently across applications and job runs; and (ii) optimizing client-centric data movement to minimize I/O load request latency between clients and servers, i.e., allocating I/O resources efficiently in service to a single application and job run. Moreover, existing approaches that require application changes limit wide-spread adoption in commercial or proprietary deployments. We propose iez, an “end-to-end control plane” where clients transparently and adaptively write to a set of selected I/O servers to achieve balanced data placement. Our control plane leverages realtime load information for distributed storage server global data placement while our design model leverages trace-based optimization techniques to minimize I/O load request latency between clients and servers. We evaluate our proposed system on an experimental cluster for two common use cases: synthetic I/O benchmark IOR for large sequential writes and a scientific application I/O kernel, HACC-I/O. Results show read and write performance improvements of up to 34% and 32%, respectively, compared to the state of the art.

The I/O subsystem for the Summit supercomputer, No. 1 on the Top500 list, and its ecosystem of analysis platforms is composed of two distinct layers, namely the in-system layer and the center-wide parallel file system layer (PFS), Spider 3. The in-system layer uses node-local SSDs and provides 26.7 TB/s for reads, 9.7 TB/s for writes, and 4.6 billion IOPS to Summit. The Spider 3 PFS layer uses IBM's Spectrum Scale™ and provides 2.5 TB/s and 2.6 million IOPS to Summit and other systems. While deploying them as two distinct layers was operationally efficient, it also presented usability challenges in terms of multiple mount points and lack of transparency in data movement. To address these challenges, we have developed novel end-to-end I/O solutions for the concerted use of the two storage layers. We present the I/O subsystem architecture, the end-to-end I/O solution space, their design considerations and our deployment experience.

NAND flash-based Solid State Devices (SSDs) offer the desirable features of high performance, energy efficiency, and fast growing capacity. Thus, the use of SSDs is increasing in distributed storage systems. A key obstacle in this context is that the natural unbalance in distributed I/O workloads can result in wear imbalance across the SSDs in a distributed setting. This, in turn can have significant impact on the reliability, performance, and lifetime of the storage deployment. Extant load balancers for storage systems do not consider SSD wear imbalance when placing data, as the main design goal of such balancers is to extract higher performance. Consequently, data migration is the only common technique for tackling wear imbalance, where existing data is moved from highly loaded servers to the least loaded ones. In this paper, we explore an innovative holistic approach, Chameleon, that employs data redundancy techniques such as replication and erasure-coding, coupled with endurance-aware write offloading, to mitigate wear level imbalance in distributed SSD-based storage. Chameleon aims to balance the wear among different flash servers while meeting desirable objectives of: extending life of flash servers; improving I/O performance; and avoiding bottlenecks. Evaluation with a 50 node SSD cluster shows that Chameleon reduces the wear distribution deviation by 81% while improving the write performance by up to 33%.

High Performance Computing (HPC) big data problems require efficient distributed storage systems. However, at scale, such storage systems often experience load imbalance and resource contention due to two factors: the bursty nature of scientific application I/O; and the complex I/O path that is without centralized arbitration and control. For example, the extant Lustre parallel file system-that supports many HPC centers-comprises numerous components connected via custom network topologies, and serves varying demands of a large number of users and applications. Consequently, some storage servers can be more loaded than others, which creates bottlenecks and reduces overall application I/O performance. Existing solutions typically focus on per application load balancing, and thus are not as effective given their lack of a global view of the system. In this paper, we propose a data-driven approach to load balance the I/O servers at scale, targeted at Lustre deployments. To this end, we design a global mapper on Lustre Metadata Server, which gathers runtime statistics from key storage components on the I/O path, and applies Markov chain modeling and a minimum-cost maximum-flow algorithm to decide where data should be placed. Evaluation using a realistic system simulator and a real setup shows that our approach yields better load balancing, which in turn can improve end-to-end performance.

With the increase of the scale and intensity of the parallel I/O workloads generated by those scientific applications running on high performance computing facilities, understanding the I/O dynamics, especially the root cause of the I/O performance variability and degradation in HPC environment, have become extremely critical to the HPC community. In this paper, we run extensive I/O measuring tests on a production leadership-class storage system to capture the performance variabilities of large-scale parallel I/O. Analyzing these results and its statistic correlation revealed some valuable insights into the characteristics of the storage system and the root cause of I/O performance variability. Further, we leverage these findings and propose an I/O middleware design refactoring which can improve the performance of the parallel I/O by optimizing the data striping and placement. Our preliminary evaluation results demonstrate the proposed approach can reduce the average per-process write latency by at least 80% and the maximum per-process write latency by at least 20%.

Proportional to the scale increases in HPC systems, many scientific applications are becoming increasingly data intensive, and parallel I/O has become one of the dominant factors impacting the large-scale HPC application performance. On a typical large-scale HPC system, we have observed that the lack of a global workload coordination coupled with the shared nature of storage systems cause load imbalance and resource contention over the end-to-end I/O paths resulting in severe performance degradation. I/O load imbalance on HPC systems is generally a self-inflicted wound and mostly occurs between the I/O paths and resources consumed by each individual job. In this paper, we introduce TAPP-IO, a dynamic, shared load balancing framework for mitigating resource contention. TAPP-IO extends our previous work and solves two major limitations: First, it transparently intercepts file creation calls during runtime to balance the workload over all available storage targets. The usage of TAPP-IO requires no application source code modifications and is independent from any I/O middleware. The framework can be applied to almost any HPC platform and is suitable for systems that lack a centralized file system resource manager. Second, the framework proposes a new placement strategy to support not only file-per-process I/O, but also single shared file I/O. This opens the door to a new class of scientific applications that can leverage the placement library for improved performance. We demonstrate the effectiveness of our integration on the Titan system at the Oak Ridge National Laboratory. Our experiments with a synthetic benchmark and real-world HPC workload show that, even in a noisy production environment, TAPP-IO can improve large-scale application performance significantly.

In this paper, we describe the GUIDE framework used to collect, federate, and analyze log data from the Oak Ridge Leadership Computing Facility (OLCF), and how we use that data to derive insights into facility operations. We collect system logs and extract monitoring data at every level of the various OLCF subsystems, and have developed a suite of pre-processing tools to make the raw data consumable. The cleansed logs are then ingested and federated into a central, scalable data warehouse, Splunk, that offers storage, indexing, querying, and visualization capabilities. We have further developed and deployed a set of tools to analyze these multiple disparate log streams in concert and derive operational insights. We describe our experience from developing and deploying the GUIDE infrastructure, and deriving valuable insights on the various subsystems, based on two years of operations in the production OLCF environment.

Designed for capacity and capability, HPC I/O systems are inherently complex and shared among multiple, concurrent jobs competing for resources. Lack of centralized coordination and control often render the end-to-end I/O paths vulnerable to load imbalance and contention. With the emergence of data-intensive HPC applications, storage systems are further contended for performance and scalability. This paper proposes to unify two key approaches to tackle the imbalanced use of I/O resources and to achieve an end-to-end I/O performance improvement in the most transparent way. First, it utilizes a topology-aware, Balanced Placement I/O method (BPIO) for mitigating resource contention. Second, it takes advantage of the platform-neutral ADIOS middleware, which provides a flexible I/O mechanism for scientific applications. By integrating BPIO with ADIOS, referred to as Aequilibro, we obtain an end-to-end and per job I/O performance improvement for ADIOS-enabled HPC applications without requiring any code changes. Aequilibro can be applied to almost any HPC platform and is mostly suitable for systems that lack a centralized file system resource manager. We demonstrate the effectiveness of our integration on the Titan system at the Oak Ridge National Laboratory. Our experiments with a synthetic benchmark and real-world HPC workload show that, even in a noisy production environment, Aequilibro can improve large-scale application performance significantly.

The increasing data demands from high-performance computing applications significantly accelerate the capacity, capability and reliability requirements of storage systems. As systems scale, component failures and repair times increase, significantly impacting data availability. A wide array of decision points must be balanced in designing such systems.

With the advent of big data, the I/O subsystems of large-scale compute clusters are becoming a center of focus. More applications are putting greater demands on end-to-end I/O performance. These subsystems are often complex in design. They comprise of multiple hardware and software layers to cope with the increasing capacity, capability, and scalability requirements of data intensive applications. However, the sharing nature of storage resources and the intrinsic interactions across these layers make it a great challenge to realize end-to-end performance gains. This paper proposes a topology-aware strategy to balance the load across resources, to improve the per-application I/O performance. We demonstrate the effectiveness of our algorithm on an extreme-scale compute cluster, Titan, at the Oak Ridge Leadership Computing Facility (OLCF). Our experiments with both synthetic benchmarks and a real-world application show that, even under congestion, our proposed algorithm can improve large-scale application I/O performance significantly, resulting in both a reduction in application run time as well as a higher resolution of simulation run.

In recent years, non-volatile memory devices such as SSD drives have emerged as a viable storage solution due to their increasing capacity and decreasing cost. Due to the unique capability and capacity requirements in large scale HPC (High Performance Computing) storage environment, a hybrid configuration (SSD and HDD) may represent one of the most available and balanced solutions considering the cost and performance. Under this setting, effective data placement as well as movement with controlled overhead become a pressing challenge. In this paper, we propose an integrated object placement and movement framework and adaptive learning algorithms to address these issues. Specifically, we present a method that shuffle data objects across storage tiers to optimize the data access performance. The method also integrates an adaptive learning algorithm where realtime classification is employed to predict the popularity of data object accesses, so that they can be placed on, or migrate between SSD or HDD drives in the most efficient manner. We discuss preliminary results based on this approach using a simulator we developed to show that the proposed methods can dynamically adapt storage placements and access pattern as workloads evolve to achieve the best system level performance such as throughput.

The Oak Ridge Leadership Computing Facility (OLCF) has deployed multiple large-scale parallel file systems (PFS) to support its operations. During this process, OLCF acquired significant expertise in large-scale storage system design, file system software development, technology evaluation, benchmarking, procurement, deployment, and operational practices. Based on the lessons learned from each new PFS deployment, OLCF improved its operating procedures, and strategies. This paper provides an account of our experience and lessons learned in acquiring, deploying, and operating large-scale parallel file systems. We believe that these lessons will be useful to the wider HPC community.

Although solid-state drives (SSDs) offer significant performance improvements over hard disk drives (HDDs) for a number of workloads, they can exhibit substantial variance in request latency and throughput as a result of garbage collection (GC). When GC conflicts with an I/O stream, the stream can make no forward progress until the GC cycle completes. GC cycles are scheduled by logic internal to the SSD based on several factors such as the pattern, frequency, and volume of write requests. When SSDs are used in a RAID with currently available technology, the lack of coordination of the SSD-local GC cycles amplifies this performance variance. We propose a global garbage collection (GGC) mechanism to improve response times and reduce performance variability for a RAID of SSDs. We include a high-level design of SSD-aware RAID controller and GGC-capable SSD devices and algorithms to coordinate the GGC cycles. We develop reactive and proactive GC coordination algorithms and evaluate their I/O performance and block erase counts for various workloads. Our simulations show that GC coordination by a reactive scheme improves average response time and reduces performance variability for a wide variety of enterprise workloads. For bursty, write-dominated workloads, response time was improved by 69 percent and performance variability was reduced by 71 percent. We show that a proactive GC coordination algorithm can further improve the I/O response times by up to 9 percent and the performance variability by up to 15 percent. We also observe that it could increase the lifetimes of SSDs with some workloads (e.g., Financial) by reducing the number of block erase counts by up to 79 percent relative to a reactive algorithm for write-dominant enterprise workloads.

The Earth System Grid Federation (ESGF) is a multi-agency, international collaboration that aims at developing the software infrastructure needed to facilitate and empower the study of climate change on a global scale. The ESGF's architecture employs a system of geographically distributed peer nodes, which are independently administered yet united by the adoption of common federation protocols and application programming interfaces (APIs). The cornerstones of its interoperability are the peer-to-peer messaging that is continuously exchanged among all nodes in the federation; a shared architecture and API for search and discovery; and a security infrastructure based on industry standards (OpenID, SSL, GSI and SAML). The ESGF software is developed collaboratively across institutional boundaries and made available to the community as open source. It has now been adopted by multiple Earth science projects and allows access to petabytes of geophysical data, including the entire model output used for the next international assessment report on climate change (IPCC-AR5) and a suite of satellite observations (obs4MIPs) and reanalysis data sets (ANA4MIPs).

NAND flash memory is a preferred storage media for various platforms ranging from embedded systems to enterprise-scale systems. Flash devices do not have any mechanical moving parts and provide low-latency access. They also require less power compared to rotating media. Unlike hard disks, flash devices use out-of-update operations and they require a garbage collection (GC) process to reclaim invalid pages to create free blocks. This GC process is a major cause of performance degradation when running concurrently with other I/O operations as internal bandwidth is consumed to reclaim these invalid pages. The invocation of the GC process is generally governed by a low watermark on free blocks and other internal device metrics that different workloads meet at different intervals. This results in I/O performance that is highly dependent on workload characteristics. In this paper, we examine the GC process and propose a semi-preemptive GC scheme that can preempt on-going GC processing and service pending I/O requests in the queue. Moreover, we further enhance flash performance by pipelining internal GC operations and merge them with pending I/O requests whenever possible. Our experimental evaluation of this semi-preemptive GC sheme with realistic workloads demonstrate both improved performance and reduced performance variability. Write-dominant workloads show up to a 66.56% improvement in average response time with a 83.30% reduced variance in response time compared to the non-preemptive GC scheme.

Solid-State Drives (SSDs) offer significant performance improvements over hard disk drives (HDD) on a number of workloads. The frequency of garbage collection (GC) activity is directly correlated with the pattern, frequency, and volume of write requests, and scheduling of GC is controlled by logic internal to the SSD. SSDs can exhibit significant performance degradations when garbage collection (GC) conflicts with an ongoing I/O request stream. When using SSDs in a RAID array, the lack of coordination of the local GC processes amplifies these performance degradations. No RAID controller or SSD available today has the technology to overcome this limitation. This paper presents Harmonia, a Global Garbage Collection (GGC) mechanism to improve response times and reduce performance variability for a RAID array of SSDs. Our proposal includes a high-level design of SSD-aware RAID controller and GGC-capable SSD devices, as well as algorithms to coordinate the global GC cycles. Our simulations show that this design improves response time and reduces performance variability for a wide variety of enterprise workloads. For bursty, write dominant workloads response time was improved by 69% while performance variability was reduced by 71%.

This paper studies the problem of self-deployment of heterogeneous mobile sensors using biologically-inspired principles and methodologies. The initial sensor deployment is assumed to be random, based on which two interrelated issues are investigated: the design of an optimal placement pattern of heterogeneous sensor platforms and the self configuration from the initial random state to the optimal state through intelligent sensor movement. We first develop an optimal placement algorithm based on the mosaic technique inspired by the retina mosaic pattern widely observed in both human and many animal visual systems. Different types of mobile sensors are organized into a mosaic pattern to maximize sensing coverage and to reduce network cost. Secondly, in order to converge to the optimal state, we investigate the swarm intelligence (SI)-based sensor movement strategy with the assistance of local communications, through which the randomly deployed sensors can self-organize themselves to reach the optimal placement state. The proposed algorithm is compared with the random movement and the SI-based method without direct communication using performance metrics such as sensing coverage, redundancy, convergence time, and energy consumption. Simulation results are presented to demonstrate the effectiveness of the mosaic placement and the SI-based movement with local communication.

Abstract Intrusion detection research has so far concentrated on techniques that effectively identify the malicious behaviors. No assurance can be assumed once the system is compromised. Intrusion tolerance, however, focuses on providing minimal level of services, even when some components have been partially compromised. The challenges here are how to take advantage of fault tolerant techniques in the intrusion tolerant system context and how to deal with possible unknown attacks and compromised components so as to continue providing the service. This paper presents our work on applying one important fault tolerance technique, acceptance testing, for building scalable intrusion tolerant systems. First, we propose a general methodology for designing acceptance testing. An Acceptance Monitor architecture is proposed to apply various tests for detecting the compromises based on the impact of the attacks. Second, we make a comprehensive vulnerability analysis on typical commercial‐off‐the‐shelf (COTS) Web servers. Various acceptance testing modules are implemented to show the effectiveness of the proposed approach. By utilizing the fault tolerance techniques on intrusion tolerance system, we provide a mechanism for building reliable distributed services that are more resistant to both known and unknown attacks. Copyright © 2003 John Wiley & Sons, Ltd.

Intrusion detection research has so far mostly concentrated on techniques that effectively identify malicious behavior. No assurance can be assumed once the system is compromised. Intrusion tolerance, on the other hand, focuses on providing minimal level of services even when some components have been partially compromised. The challenges here are how to take advantage of fault tolerant techniques in the intrusion tolerant system context and how to deal with possible unknown attacks and compromised components so as to continue providing the service. This paper presents our work on applying one important fault tolerance technique, acceptance testing, for building scalable intrusion tolerant systems. First, we propose a general methodology for designing acceptance tests. An acceptance monitor architecture is proposed to apply various tests for detecting compromises based on the impact of the attacks. Second, we make a comprehensive vulnerability analysis on typical commercial-off-the-shelf (COTS) Web servers. Various acceptance testing modules are implemented to show the effectiveness of the proposed approach. By utilizing the fault tolerance techniques on intrusion tolerance system, we provide a mechanism for building reliable distributed services that are more resistant to both known and unknown attacks.

This paper proposes a new approach to fault detection for a link state routing system-property oriented analysis and detection (POD). A routing system is modeled as a set of distributed processes. A property is defined as state predicate(s) over system variables. For the link state routing protocol, the high-level overall converging property P is defined as the "synchronization" among routing information bases maintained by all processes. We decompose the routing protocol into different computation phases. For each phase, we use invariant state predicates (safety property) and the liveness property as our guide for observation and analysis. The goal of the detection algorithm is to construct a validation path based on the history to determine if the fault is natural or malicious once the stable property P is rendered invalid by faults. The contribution of this paper is twofold: first, a new detection approach is proposed that differs from traditional signature-based or profile-based intrusion detection paradigms in the sense that it utilizes the stable property as a starting point, and correlates the history and future to validate changes in the system; second, by exploring the primary concerned system properties, we show that detection effort can be conducted in a more focused and systematic fashion.

The JiNao IDS project focuses on detecting intrusions, especially insider attacks against link state routing protocols such as OSPF. One important feature of the JiNao system is its integrated network management (INM) capability. Through SNMP and distributed programming interface (DPI), we can manage and control distributed JiNao IDS remotely, interoperate with other JiNao systems to do correlation analysis, and utilize both private MIB and OSPF MIB as a complementary way of doing intrusion detection. This paper describes the design and implementation of JiNao's INM architecture. Three OSPF insider attacks (maxseq, maxage, and seq++) have been developed to evaluate its effectiveness and detection capability.

This paper analyzes both the strong points and weak points of the OSPF routing protocol from the security perspective. On its strong points, we abstract its features of information least dependency and information hiding, which make it very robust and fault resilient, even when facing certain malicious attacks. On its weak points, we take a pragmatic look at various problems centering round secure routing protocols. By carefully investigating a special re-routing attacking case, we show how a home-made malicious router can easily disrupt the service. It also provides a concrete example for routing protection and intrusion detection. Finally, we present the active protection idea and its architectural flexibility and compatibility advantages.

The JiNao project at MCNC/NCSU focuses on detecting intrusions, especially insider attacks, against the OSPF (open shortest path first) routing protocol. This paper presents the implementation and experiments of JiNao's statistical intrusion detection module. Our implementation is based upon the algorithm developed in SRI's NIDES (next-generation intrusion detection expert system) project. Some modifications and improvements to NIDES/STAT are made for a more effective implementation in our environment. Also, three OSPF insider attacks (e.g., maxseq, maxage, and seq++ attacks) have been developed for evaluating the efficacy of detecting capability. The experiments were conducted on two different network routing testbeds. The results indicate that the proposed statistical mechanism is very effective in detecting these routing protocol attacks.

It is critical to protect the network infrastructure (e.g., network routing and management protocols) against security intrusions, yet dealing with insider attacks are probably one of the most challenging research problems in network security. We study the security threats, especially internal/insider threats, for the standardized routing protocol OSPF. In OSPF, a group of routers collaborate, exchange routing information, and forward packets for each other. If one (and maybe more than one) router is evil or compromised, how can this router damage the whole network? In this paper, we analyze OSPF and identify its strengths and weakness under various insider attacks. Furthermore, to confirm our analysis, we have implemented and experimented one attack, the max sequence number attack, on our OSPF routing testbed. Our attack is very successful against two independently developed router products as it will block routing updates for 60 minutes by simply injecting one bad OSPF protocol data unit.