BlueField-4 DPU

Yuke Li, Arjun Kashyap, Yanfei Guo, Xiaoyi Lu

The Data Processing Unit (DPU) (i.e., programmable SmartNICs with System-on-Chip or SoC cores) has emerged as a valuable supplementary resource to the host CPU. The DPU architecture has been attracting significant attention within High-Performance Computing (HPC) and data center clusters due to its advanced capabilities and accelerators, which include a hardware-based data compression engine. This positions the DPU as a prospective tool for accelerating and offloading compression workloads from the hosts, which can potentially speed up data-intensive applications. The convergence of Big Data, HPC, and Machine Learning (ML) systems has rendered large data volumes a major performance bottleneck in message communication and data storage. While compression can boost performance, recent studies reveal that compression techniques (e.g., lossy and lossless) are compute-intensive and time-consuming, particularly with larger data sizes. Consequently, this paper characterizes the performance of three lossy (SZ3) and lossless (DEFLATE and zlib) compression algorithms with seven real-world data sets on the popular NVIDIA’s BlueField DPUs to explore potential opportunities for offloading these workloads from the host. We find that compared to DPU’s SoC cores, DPU’s hardware compression engine can obtain up to 26.8x performance speedup. Furthermore, we discuss the challenges and opportunities associated with employing NVIDIA’s BlueField DPUs to accelerate lossy and lossless compression/decompression workloads. Our research discloses five important takeaways which shed light on future research directions for lossy and lossless compressions on DPUs.

Nick Sarkauskas, Mohammadreza Bayatpour, Tu Tran, B. Ramesh, H. Subramoni, D. Panda

Since the introduction of nonblocking collectives in the MPI-3 standard, communication has been progressed by several mechanisms. One such mechanism includes modifying the application code to periodically call MPI_ Test to enter the MPI library. Another launches an extra thread per core to progress communication asynchronously. Communication progression can also be offloaded to the Host Channel Adapter (HCA) using the latest hardware. In this paper, we explore this last option by using the Data Processing Unit (DPU) shipped with the BlueField-2 SmartNIC adapter to offload progression of non-blocking MPI_Ibcast and MPI_Iallgather collectives. For both collectives, we present several designs which take advantage of the DPU. We demonstrate the efficacy of our proposed designs through microbenchmark evaluations. At the microbenchmark level, total execution time of the osu_ibcast microbenchmark can be reduced by up to 54% using our DPU-based Ibcast designs. Total execution time of the osu_iallgather microbenchmark can be reduced by up to 43 %. To the best of our knowledge, this is the first work to optimize nonblocking broadcast and allgather collectives on emerging BlueField DPUs.

Kasim Tasdemir, Rafiullah Khan, F. Siddiqui, S. Sezer, Fatih Kurugollu, Alperen Bolat

SQL injection attacks present a significant risk to data center security. Traditional rule-based pattern matching techniques exhibit limitations, such as inability to adapt to new attack types, to give decision confidence and lower detection accuracy. Machine learning (ML) based approaches offer promising alternatives; however, their computational requirements and the increasing volume of network traffic pose challenges for their application in conventional hardware. Data Processing Units (DPUs) have emerged as the tailored computing platform for infrastructure related workloads within data centers including security. This paper evaluates the performance and efficiency of classical ML methods for SQL injection detection utilising computing resources on DPUs.In this study, 20 prominent ML models are tested against a dataset comprising 30,000 SQL payloads, and their performance is compared in a series of experiments. The results indicate that the Passive Aggressive Classifier is the most suitable model for near-real-time detection, achieving a detection latency of approximately 0.3μs/sample with an accuracy of 99.78%. This paper demonstrates that ML methods can be efficiently and effectively deployed on DPUs for SQL injection detection, providing valuable insights into threat intelligence for enhancing data center security. The codes of this study can be found at: https://github.com/gdrlab/dpu-sqli-detection.

Yuke Li, Arjun Kashyap, Weicong Chen, Yanfei Guo, Xiaoyi Lu

Data compression has become a crucial technique in addressing performance bottlenecks caused by increasing data volumes in High-Performance Computing (HPC), Big Data, and Deep Learning (DL). Despite its potential to boost system performance, recent studies have identified significant challenges with existing compression methods, mainly due to their high computational demands amidst continuously growing data sizes. Concurrently, the advent of Data Processing Units (DPUs), equipped with programmable System-on-Chip (SoC) and specialized compression accelerators, offers a promising opportunity to alter the landscape of data compression. This paper explores the complexities and potential of leveraging NVIDIA BlueField DPUs to accelerate lossy and lossless compression. Towards this, we introduce PEDAL, an innovative library that leverages the hardware capabilities of DPUs to unify and optimize data compression designs. Moreover, we seamlessly co-design PEDAL with the popular MPICH MPI library, demonstrating up to 101x speedup in compression time and 88x decrease in communication latency. Drawing on these achievements, we share our experience with various research communities about accelerating data compression on DPUs in communication-oriented HPC scenarios.

Arjun Kashyap, Yuke Li, Darren Ng, Xiaoyi Lu

Data Processing Units (DPUs) are becoming available in datacenter environments to offload/accelerate workloads from the host. However, a comprehensive analysis is required to help users determine how to effectively utilize DPUs for their workloads, considering the various configurations and generations available. To fill in this gap, we conduct a fair and rigorous characterization by performing 15 benchmarking tests to demonstrate the evolution of representative SoC-based DPUs, specifically NVIDIA’s BlueField-1, BlueField-2, and BlueField-3. Our work surfaces several idiosyncrasies across three key characterization dimensions—network, DMA engine, and memory. For network, we exhaustively test two major DPU modes—on-path (and five submodes) and off-path modes. We develop DPUDMABench, a microbenchmark suite to systematically analyze different data exchange primitives supported by DPU’s DMA engine. We also conduct two application case studies examining the DPU mode’s performance impact on TCP/IP and RDMA-based key-value stores (MICA and HERD). Based on our multi-generational DPU characterization, we identify and summarize 14 major idiosyncrasies, along with providing guidelines for optimal system and future hardware design.