Is GPU integration critical for Predictive Analytics?
Wed, 05 Apr 2023 18:22:40 -0000
|Read Time: 0 minutes
GPUs are getting widespread attention in the Predictive Analytics (PredAn) space. This is due to their ability to perform parallel computation on large volumes of data. GPUs leverage complex models that are tightly integrated to the simulation required to do control synthesis for real-time response in Industry 4.0 (I4) solutions. Consider the predictive maintenance use-case, where telemetry from servers in the datacenter are captured for failure analysis in the analytics cluster and control sequences generated to avoid downtime. Clearly, to be on track, the machine needs to project current data into the future and simulate fault partitions in the monitorable list to negotiate a fix, all in a tight time-window. However, prediction and simulation are inherently slow, particularly when this needs to be done on many fault partitions over multiple servers.
We argue that two things can help:
- Linearizing prediction with Koopman filters
- Leveraging generative models for control synthesis in the simulation space
We use Koopman filter to project data into a skinnier latent basis space for Dimension Reduction (DR) and embed this transformation inside an autoencoder. A probe to convert eigen-vector sections to an Eigen Value Sequence (EVS) that correlates to survival probabilities can then be transformed to Time2Failure (T2F) estimates. This failure detection can then be tagged with a reference to a pre-calibrated auto-fix script derived using Anomaly Fingerprinting (AF) while simulating the projected fault partition. Generative models (GAN) allow performance and footprint optimization, resulting in faster inferencing. In that sense, this improves inferencing throughput. We train generative models for Data, DR, T2F and AF and use them for fast inferencing. Figure 1(a) shows the inferencing flow, 1(b) shows Koopman linearization, and 1(c) shows the underlying GAN footprint.
Figure 1. Inferencing flow, Koopman linearization, and the underlying GAN footprint
In figure 1(a), T2F estimates for all faults on all servers are triaged by the scheduler using DRL in the inferencing phase. In figure 1(b), each fault-group, eigen-vector dimension is searched in the autoencoder frame by resizing the encoder depth for failure clarity in EVS. In figure 1(b), G2 is derived from G1, and G1 from G0 for each fault group. Generative model synthesis enables the mapping of complex computation to high-performance low-footprint analogues that can be leveraged at inference time.
The GPU Argument
A GPU is typically used for both training and inferencing. In the predictive maintenance testbed, we stream live telemetry from iDRACs to the analytics cluster built using Splunk services like streaming, indexing, search, tiering and data-science tools on Robin.io k8s platform. The cluster has access to Nvidia GPU resources for both training and inferencing. The plot in figure 2(a) shows that the use of asynchronous access to Multi-Instance GPU (MIG) inferencing provides performance gain over blocking alternative, measured using wall-clock estimates. The GPU scheduler manages asynchronous T2F workloads better, and blocking calls would require timeout reconfiguration in production. The plot in figure 2(b) shows that inferencing performance of generative models improved by 15% (for 12+ episodes) when selective optimization (DRL-on-CPU and T2F-Calculation-on-GPU) was opted. The direction of this trend makes sense because DRL-in-GPU requires frequent memory-to-memory transfers and so is an ideal candidate for CPU pinning, whereas T2F estimates are dense but relatively less frequent computations that do well when mapped to GPU with MIG enabled. As the gap between the plots widens, this indicates that the CPU only computation cannot keep up with data pile-up, so input sections need to be shortened. The plot in figure 2(c) shows that fewer batches (assuming fixed dataset size) shortened epochs needed to achieve the desired training accuracy in GAN. However, larger batch size requires more GPU memory implying MIG disablement for improved throughput and energy consumption. Based on this data, we argue that dedicating a GPU for training (single kernel) as opposed to switching kernels (between training and inferencing) improves throughput. This tells us that the training GPU (without MIG enablement) and the inferencing GPU (with MIG enablement) should be kept separate in I4 for optimal utilization and performance. Based on current configuration choices, this points to dual Nvidia A30 GPU preference as opposed to a single Nvidia A100 GPU attached to the Power Edge server worker node. The plot in figure 2(d) shows that single layer generative models improve inferencing performance and scales more predictively. The expectation is that multilayering would do better. The plot indicates performance improvement as the section size increases, although more work is needed to understand the impact of multilayering.
Figure 2. (a) Asynchronous calls for MIG inferencing provide performance gain over blocking calls. (b) Selective optimization provides better inferencing performance. (c) Larger training batch size (fewer batches) shortens the epochs needed to achieve acceptable accuracy. (d) Generative models improve inferencing performance.
In conclusion, predictive analytics is essential for maintenance in the era of digital transformation. We present a solution that scales with Dell server telemetry specification. It is widely accepted that for iterative error correction, linear feedback control systems perform better than their non-linear counterparts. We shaped predictions to behave linearly. By using generative models, one can achieve faster inferencing. We proposed a new way to couple generative models with Digital Twins (DT) simulation models for scheduling shaped by DRL. Our experiments indicate that GPU in analytics cluster accelerates response performance in I4 feedback loops (e.g., MIG enablement at inferencing, leveraging generative models to fast-track control synthesis).
References
- Brunton, S. L.; Budišić, M; Kaiser, E; Kutz, N. Modern Koopman Theory for Dynamical Systems. arXiv 2102.12086 2021.
- Brophy, E; Wang, Z; She, Q; Ward, T. Generative Adversarial Networks in Time Series: A Systematic Literature Survey. ACM Computing Surveys 2023, 55(10), Article 199.
- Matsuo, Y; LeCun, Y; Sahani, M; Precup, D; Silver, D; Sugiyama; M; Uchibe, E; Morimoto, J. Deep learning, reinforcement learning, and world models. Neural Networks Elsevier Press 2022, 152(2022), pp 267-275.
- Gara, S et al. Telemetry Streaming with iDRAC9. Dell White Paper May 2021.
Related Blog Posts
Choosing a PowerEdge Server and NVIDIA GPUs for AI Inference at the Edge
Fri, 05 May 2023 16:38:19 -0000
|Read Time: 0 minutes
Dell Technologies submitted several benchmark results for the latest MLCommonsTM Inference v3.0 benchmark suite. An objective was to provide information to help customers choose a favorable server and GPU combination for their workload. This blog reviews the Edge benchmark results and provides information about how to determine the best server and GPU configuration for different types of ML applications.
Results overview
For computer vision workloads, which are widely used in security systems, industrial applications, and even in self-driven cars, ResNet and RetinaNet results were submitted. ResNet is an image classification task and RetinaNet is an object detection task. The following figures show that for intensive processing, the NVIDIA A30 GPU, which is a double-wide card, provides the best performance with almost two times more images per second than the NVIDIA L4 GPU. However, the NVIDIA L4 GPU is a single-wide card that requires only 43 percent of the energy consumption of the NVIDIA A30 GPU, considering nominal Thermal Design Power (TDP) of each GPU. This low-energy consumption provides a great advantage for applications that need lower power consumption or in environments that are more challenging to cool. The NVIDIA L4 GPU is the replacement for the best-selling NVIDIA T4 GPU, and provides twice the performance with the same form factor. Therefore, we see that this card is the best option for most Edge AI workloads.
Conversely, the NVIDIA A2 GPU exhibits the most economical price (compared to the NVIDIA A30 GPU's price), power consumption (TDP), and performance levels among all available options in the market. Therefore, if the application is compatible with this GPU, it has the potential to deliver the lowest total cost of ownership (TCO).
Figure 1: Performance comparison of NVIDIA A30, L4, T4, and A2 GPUs for the ResNet Offline benchmark
Figure 2: Performance comparison of NVIDIA A30, L4, T4, and A2 GPUs for the RetinaNet Offline benchmark
The 3D-UNet benchmark is the other computer vision image-related benchmark. It uses medical images for volumetric segmentation. We saw the same results for default accuracy and high accuracy. Moreover, the NVIDIA A30 GPU delivered significantly better performance over the NVIDIA L4 GPU. However, the same comparison between energy consumption, space, and cooling capacity discussed previously applies when considering which GPU to use for each application and use case.
Figure 3: Performance comparison of NVIDIA A30, L4, T4, and A2 GPUs for the 3D-UNet Offline benchmark
Another important benchmark is for BERT, which is a Natural Language Processing model that performs tasks such as question answering and text summarization. We observed similar performance differences between the NVIDIA A30, L4, T4, and A2 GPUs. The higher the value, the better.
Figure 4: Performance comparison of NVIDIA A30, L4, T4, and A2 GPUs for the BERT Offline benchmark
MLPerf benchmarks also include latency results, which are the time that systems take to process requests. For some use cases, this processing time can be more critical than the number of requests that can be processed per second. For example, if it takes several seconds to respond to a conversational algorithm or an object detection query that needs a real-time response, this time can be particularly impactful on the experience of the user or application.
As shown in the following figures, the NVIDIA A30 and NVIDIA L4 GPUs have similar latency results. Depending on the workload, the results can vary due to which GPU provides the lowest latency. For customers planning to replace the NVIDIA T4 GPU or seeking a better response time for their applications, the NVIDIA L4 GPU is an excellent option. The NVIDIA A2 GPU can also be used for applications that require low latency because it performed better than the NVIDIA T4 GPU in single stream workloads. The lower the value, the better.
Figure 4: Latency comparison of NVIDIA A30, L4, T4, and A2 GPUs for the ResNet single-stream and multistream benchmark
Figure 5: Latency comparison of NVIDIA A30, L4, T4, and A2 GPUs for the RetinaNet single-stream and multistream benchmark and the BERT single-stream benchmark
Dell Technologies submitted to various benchmarks to help understand which configuration is the most environmentally friendly as the data center’s carbon footprint is a concern today. This concern is relevant because some edge locations have power and cooling limitations. Therefore, it is important to understand performance compared to power consumption.
The following figure affirms that the NVIDIA L4 GPU has equal or better performance per watt compared to the NVIDIA A2 GPU, even with higher power consumption. For Throughput and Perf/watt values, higher is better; for Power(watt) values, lower is better.
Figure 6: NVIDIA L4 and A2 GPU power consumption comparison
Conclusion
With measured workload benchmarks on MLPerf Inference 3.0, we can conclude that all NVIDIA GPUs tested for Edge workloads have characteristics that address several use cases. Customers must evaluate size, performance, latency, power consumption, and price. When choosing which GPU to use and depending on the requirements of the application, one of the evaluated GPUs will provide a better result for the final use case.
Another important conclusion is that the NVIDIA L4 GPU can be considered as an exceptional upgrade for customers and applications running on NVIDIA T4 GPUs. The migration to this new GPU can help consolidate the amount of equipment, reduce the power consumption, and reduce the TCO; one NVIDIA L4 GPU can provide twice the performance of the NVIDIA T4 GPU for some workloads.
Dell Technologies demonstrates on this benchmark the broad Dell portfolio that provides the infrastructure for any type of customer requirement.
The following blogs provide analyses of other MLPerfTM benchmark results:
- Dell Servers Excel in MLPerf™ Inference 3.0 Performance
- Dell Technologies’ NVIDIA H100 SXM GPU submission to MLPerf™ Inference 3.0
- Empowering Enterprises with Generative AI: How Does MLPerf™ Help Support
- Comparison of Top Accelerators from Dell Technologies’ MLPerf™
References
For more information about Dell Power Edge servers, go to the following links:
- Dell’s PowerEdge XR7620 for Telecom/Edge Compute
- Dell’s PowerEdge XR5610 for Telecom/Edge Compute
- PowerEdge XR4520c Compute Sled specification sheet
- PowerEdge XE2420 Spec Sheet
For more information about NVIDIA GPUs, go to the following links:
MLCommonsTM Inference v3.0 results presented in this document are based on following system IDs:
| ID | Submitter | Availability | System |
|---|---|---|---|
2.1-0005 | Dell Technologies | Available | Dell PowerEdge XE2420 (1x T4, TensorRT) |
2.1-0017 | Dell Technologies | Available | Dell PowerEdge XR4520c (1x A2, TensorRT) |
2.1-0018 | Dell Technologies | Available | Dell PowerEdge XR4520c (1x A30, TensorRT) |
2.1-0019 | Dell Technologies | Available | Dell PowerEdge XR4520c (1x A2, MaxQ, TensorRT) |
2.1-0125 | Dell Technologies | Preview | Dell PowerEdge XR5610 (1x L4, TensorRT, MaxQ) |
2.1-0126 | Dell Technologies | Preview | Dell PowerEdge XR7620 (1x L4, TensorRT) |
Table 1: MLPerfTM system IDs
Comparison of Top Accelerators from Dell Technologies’ MLPerf™ Inference v3.0 Submission
Fri, 21 Apr 2023 21:43:39 -0000
|Read Time: 0 minutes
Abstract
Dell Technologies recently submitted results to MLPerfTM Inference v3.0 in the closed division. This blog highlights the NVIDIA H100 PCIe GPU and compares the results to the NVIDIA A100 PCIe GPU with the PCIe form factor held constant.
Introduction
MLPerf Inference v3.0 submission falls under the benchmarking pillar of the MLCommonsTM consortium with the objective to make fair comparisons across server configurations. Submissions that are made to the closed division warrant an equitable comparison of the systems.
This blog highlights the closed division submissions Dell Technologies made with the NVIDIA A100 GPU using the PCIe (peripheral component interconnect express) form factor. The PCIe form factor is an interfacing standard for connecting various high-speed components in hardware such as a computer or a server. Servers include a certain number of PCIe slots in which to insert GPUs or other additional cards. Note that there are different physical configurations for the slots to indicate the number of lanes for data to travel to and from the PCIe card. The NVIDIA H100 GPU is truly the latest and greatest GPU with NVIDIA AI Enterprise included; it is a dual-slot air cooled PCIe generation 5.0 GPU. This GPU runs at a memory bandwidth speed of over 2,000 megabits per second and up to seven Multi-Instance GPUs at 10 gigabytes each. The NVIDIA A100 80 GB GPU is a dual-slot PCIe generation 4.0 GPU that runs at a memory bandwidth speed of over 2,000 megabits per second.
NVIDIA H100 PCIe GPU and NVIDIA A100 PCIe GPU comparison
In addition to making a submission with the NVIDIA A100 GPU, Dell Technologies made a submission with the NVIDIA H100 GPU. To make a fair comparison, the systems were identical and the PCIe form factor was held constant.
Platform | Dell PowerEdge R750xa (4x A100-PCIe-80GB, TensorRT) | Dell PowerEdge R750xa (4x H100-PCIe-80GB, TensorRT) |
Round | V3.0 | |
MLPerf System ID | R750xa_A100_PCIe_80GBx4_TRT | R750xa_H100_PCIe_80GBx4_TRT |
Operating system | CentOS 8.2 | |
CPU | Intel Xeon Gold 6338 CPU @ 2.00 GHz | |
Memory | 1 TB | 1 TB |
GPU | NVIDIA A100-PCIe-80GB | NVIDIA H100-PCIe-80GB |
GPU form factor | PCIe | |
GPU memory configuration | HBM2e | |
GPU count | 4 | |
Software stack | TensorRT 8.6 CUDA 12.0 cuDNN 8.8.0 Driver 525.85.12 DALI 1.17.0 | TensorRT 8.6 CUDA 12.0 cuDNN 8.8.0 Driver 525.60.13 DALI 1.17.0 |
Table 1: Software stack of submissions made on NVIDIA A100 PCIe and NVIDIA H100 PCIe GPUs for MLPerf Inference v3.0 on the Dell PowerEdge R750xa server
In the following figure, the per card numbers are normalized over the NVIDIA A100 GPU results to show a readable comparison of the GPUs on the same system. Across object detection, medical image segmentation, and speech to text and natural language processing, the latest NVIDIA H100 GPU outperforms its predecessor in all categories. Note the outstanding performance of the Dell PowerEdge R750xa server with NVIDIA H100 GPUs with the BERT benchmark in the high accuracy mode. With the advancements in generative artificial intelligence, the Dell PowerEdge R750xa server is a versatile, reliable, and high performing platform.
Figure 1: Normalized per GPU comparison of NVIDIA A100 and NVIDIA H100 GPUs on the Dell PowerEdge R750xa server
The following figures show absolute numbers for a comparison of the NVIDIA H100 and NVIDIA A100 GPUs.
Figure 2: Per GPU comparison of NVIDIA A100 and NVIDIA H100 GPUs for RetinaNet on the PowerEdge R750xa server
Figure 3: Per GPU comparison of NVIDIA A100 and NVIDIA H100 GPUs for 3D-Unet on the PowerEdge R750xa server
Figure 4: Per GPU comparison of NVIDIA A100 and NVIDIA H100 GPUs for RNNT on the PowerEdge R750xa server
Figure 5: Per GPU comparison of NVIDIA A100 and NVIDIA H100 GPUs for BERT on the PowerEdge R750xa server
These results can be found on the MLCommons website.
Submissions made with the NVIDIA A100 PCIe GPU
In this round of submissions, Dell Technologies submitted results on the PowerEdge R750xa server packaged with four NVIDIA A100 80 GB PCIe GPUs. In previous rounds, the PowerEdge R750xa server showed outstanding performance across all the benchmarks. For a deeper dive of a previous round's submission, check out our blog from MLPerf Inference v2.0. From the previous round of MLPerf Inference v2.1 submissions, Dell Technologies submitted results on an identical system. However, across the two rounds of submissions, the main difference is the upgrades in the software stack, as described in the following table:
Platform | Dell PowerEdge R750xa (4x A100-PCIe-80GB, TensorRT) | Dell PowerEdge R750xa (4x A100-PCIe-80GB, TensorRT) |
Round | V3.0 | V2.1 |
MLPerf System ID | R750xa_A100_PCIe_80GBx4_TRT | |
Operating system | CentOS 8.2 | |
CPU | Intel Xeon Gold 6338 CPU @ 2.00 GHz | |
Memory | 512 GB | |
GPU | NVIDIA A100-PCIe-80GB | |
GPU form factor | PCIe | |
GPU memory configuration | HBM2e | |
GPU count | 4 | |
Software stack | TensorRT 8.6 CUDA 12.0 cuDNN 8.8.0 Driver 525.85.12 DALI 1.17.0 | TensorRT 8.4.2 CUDA 11.6 cuDNN 8.4.1 Driver 510.39.01 DALI 0.31.0 |
Table 2: Software stack for submissions made on the NVIDIA A100 PCIe GPU in MLPerf Inference v3.0 and v2.1
Comparison of PowerEdge R750xa NVIDIA A100 results from Inference v3.0 and v2.1
Object detection
The RetinaNet benchmark falls under the object detection category and uses the OpenImages dataset. The results from Inference v3.0 show a less than 0.05 percent difference in the Server scenario and a 21.53 percent difference in the Offline scenario. A potential reason for this result might be NVIDIA’s optimizations, as outlined in their technical blog.
Figure 6: RetinaNet Server and Offline results on the PowerEdge R750xa server from Inference v3.0 and Inference v2.1
Medical image segmentation
The 3D-Unet benchmark performs the KiTS 2019 kidney tumor segmentation task. Across the two rounds of submission, the PowerEdge R750xa server performed consistently well with a 0.3 percent difference in both the default and high accuracy modes.
Figure 7: 3D-UNet Offline results on the PowerEdge R750xa server from Inference v3.0 and v2.1
Speech to text
The Recurrent Neural Network Transducers (RNNT) model falls under the speech recognition category. This benchmark accepts raw audio samples and produces the corresponding character transcription. In the Server scenario, the results are within a 2.25 percent difference and 0.41 percent difference in the Offline scenario.
Figure 8: RNNT Server and Offline results on the Dell PowerEdge R750xa server from Inference v3.0 and v2.1
Natural language processing
Bidirectional Encoder Representation from Transformers (BERT) is a state-of-the-art language representational model for Natural Language Processing applications. This benchmark performs the SQuAD question answering task. The BERT benchmark consists of default and high accuracy modes for the Offline and Server scenarios. For the Server scenarios, the default mode results are within a 1.69 percent range and 3.12 percent range for the high accuracy mode. For the Offline scenarios, a similar behavior is noticeable in which the default mode results are within a 0.86 percent range and 3.65 percent range in the high accuracy mode.
Figure 9: BERT Server and Offline results on the PowerEdge R750xa server from Inference v3.0 and v2.1
Conclusion
Across the various rounds of submissions to the MLPerf Inference benchmark suite, the PowerEdge R750xa server has been a consistent top performer for any machine learning tasks ranging from object detection, medical image segmentation, speech to text and natural language processing. The PowerEdge R750xa server continues to be an excellent server choice for machine learning inference workloads. Customers can take advantage of the diverse results submitted on the Dell PowerEdge R750xa server with the NVIDIA H100 GPU to make an informed decision for their specific solution needs.