Designing for the Edge: PowerEdge and NEBS
Download PDFMon, 16 Jan 2023 15:33:02 -0000
|Read Time: 0 minutes
Summary
Compute infrastructures are evolving to meet the demand for low latency, distributed computing outside the data center. These edge environments can present unique challenges that traditional servers do not address. Dell PowerEdge has led the way in designing and delivering reliable servers built for the edge.
Introduction
As computing expands beyond the data center, there is a need for enterprise-grade servers that offer reliable performance in environmentally challenging facilities. A traditional data center can provide power redundancy, climate control, and physical security. However, a server deployed in a telephone network's central office, a manufacturing facility, or in a retail backroom may be more exposed to the effects of natural disasters, extreme temperatures, high humidity, airborne contaminants, high altitude, lightning, impact shock, vibration, or EMI emissions. Dell Technologies understands the unique challenges of these environments, and our engineering teams design our edge servers to be certified rugged for NEBS.
What is NEBS Testing?
The North American telecom industry requires service providers and edge computing providers to be Network Equipment-Building System (NEBS) compliant to ensure network integrity, compatibility, and safety. Being NEBS compliant indicates that the products and equipment operate reliably at the edge. Therefore, network operators need to invest in suppliers who ensure their performance through rigorous testing.
With the adoption of 5G, rapid network expansion, and the need for carriers to successfully manage their infrastructure during extreme weather events, the demand for NEBS-compliant devices is only increasing.
NEBS Test Levels
NEBS compliance ensures that a server meets the GR-63-CORE and GR-1089- CORE standards and is made up of various levels that distinguish certain aspects of testing. Each one verifies a different performance specification with operational requirements. For example, the lowest level of NEBS compliance, Level 1, is used for prototypes in laboratory trials. By contrast, the highest level, Level 3, is typically required for equipment deployed in a communications network.
Many standards fall within the scope of NEBS. The standards most used are:
- GR-1089-CORE - Electromagnetic compatibility and electrical safety
- GR-63-CORE - Physical protection
The NEBS Levels are described below. Note that successive levels incorporate previous level requirements.
- NEBS Level 1: Addresses equipment safety measures and requirements for GR-63-CORE and GR-1089-CORE standards. Typically used by service providers for prototype equipment for trial and limited deployment equipment.
- NEBS Level 2: Addresses equipment operability in controlled environments such as data centers. Level 2 includes all requirements of Level 1 with some added level of operability reliability.
- NEBS Level 3: Determines that the equipment meets all the requirements of GR-63-CORE and GR-1089-CORE. This provides the highest assurance of product operability. Most TCGs require Level 3 before acceptance/installation on the networks.
NEBS Level 1 | NEBS Level 2 | NEBS Level 3 |
GR-63
GR-1089
|
GR-63
GR-1089
|
GR-63
GR-1089
|
Table 1. NEBS Levels
NEBS Testing
NEBS testing is designed to simulate conditions at the edge. The following section shows the test areas in more detail.
Thermal and Altitude Exposure
Servers deployed at the edge can be exposed to extreme temperatures, humidity, or high altitude. These conditions can occur during transport, storage, or operation. This testing ensures that a server functions reliably during and after exposure.
Test | Server State | Test Conditions |
Non-Operational Test | Off |
|
Operational Test | On and running system stress |
|
Simulated Fan Failure | On and running system stress |
|
Altitude Exposure | On and running system stress |
|
Table 2. Thermal and Altitude Testing
Flame Resistance
In rare instances, a server can malfunction and produce sparks or fire. This testing ensures that the flames do not escape from the server chassis and damage adjacent people, equipment, or facilities. The fire must be confined to a failing chassis, and the chassis should self-extinguish and dissipate the smoke. In addition, the server materials must be flame retardant to minimize the spread of flames.
Shipping Impact Survivability
While shipping partners do their best to handle equipment carefully, accidents happen. When a packaged server is dropped, it should not sustain significant damage. NEBS testing ensures the server packaging material protects the server during typical shipping scenarios. This testing includes drops from heights up to 1M (weight dependent) on all axes.
Seismic and Vibration Robustness
A server must withstand transportation vibration and earthquakes up to 8.3 on the Richter Scale. Drives, risers, memory, PCIe devices, and other components should not dislodge, break, become loose, or stop functioning after experiencing seismic or transportation vibrations. A server is subjected to a prescribed motion waveform that simulates typical earthquake motion. This motion occurs in multiple axes over time, and once complete, the server is checked for damage and proper functionality. The server must not sustain permanent damage and should function normally after the test.
Airborne Contaminants
Servers deployed to edge sites such as factory floors or cell towers may encounter contaminants not seen in a clean, climate-controlled data center. In this testing, servers are exposed to various contaminants over approximately ten days and should function properly after the test. The fan filter must block contaminants and allow the server to deliver reliable performance. The filter must be replaceable while the server is operational. The exposure testing includes gas with corrosive material and hygroscopic sand-like contaminants
Acoustic Noise
High-performance fans move an enormous amount of air and can be loud. Therefore, NEBS testing includes checking the sound power while running fans at maximum speed to ensure OSHA compliance. The server should be less than 78dB at max fan speed in Telco Room testing and less than 83dB at max fan speed in Power Room testing.
ESD Robustness
Electrostatic discharge happens. Servers must function as designed after exposure to ESD. This exposure testing includes close and open chassis scenarios that simulate a field repair scenario. ESD testing includes 8kV contact discharge and 15kV air discharge
EMI Emissions and Immunity
Electromagnetic interference, or EMI, is present in areas with electronic equipment. EMI can be exceptionally high when lots of electrical equipment is placed in confined spaces. This testing ensures that a server does not radiate emissions beyond the max allowed and that it can survive exposure to emissions from other devices.
Lightning and Surge Robustness
Edge servers may encounter periodic site lightning. Given proper grounding, the server must survive a power surge from lightning. This test simulates the power surge generated from a lightning strike and the server's ability to function correctly afterward.
Designing for the Edge
Dell works with customers and partners to understand the unique challenges of operating outside the data center. We use this knowledge and experience to build best-in-class, enterprise-grade servers that offer reliable performance at the edge. Some edge servers offered by Dell include the PowerEdge XR11, XR12, and XE2420. The following section highlights some of the design elements of our edge servers.
Figure 1. Dell PowerEdge XE2420 with optional bezel filter
Rugged, Flexible, Compact Chassis Options
- Hot and cold aisle access options
- AC and DC power supply support
- Short chassis depth for confined spaces
- Support for full-size PCIe cards
- Individual locking mechanisms prevent dislodging of add-on cards
- Optional bezel air filter to protect against airborne contaminants
Optimized Thermal Performance
- High-performance fans
- Minimal in-chassis airflow obstructions
- Unique chassis designs eliminate preheated air across add-in devices like GPU accelerators
- Configurable thermal management with iDRAC
Figure 2. Dell PowerEdge XR12
Conclusion
With the need for organizations to integrate products from many vendors into large, robust systems, they need to be confident that these products are resilient. NEBS testing ensures that these products meet a high standard of reliability and longevity at the edge. So, choosing a Dell PowerEdge server that is certified rugged for NEBS gives you the peace of mind that your server can perform in rough environmental conditions at the edge, with the guaranteed reliability of Dell PowerEdge.
References
- Telcordia. NEBS Requirements: Physical Protection. Generic Requirements, GR-63-CORE, Issue 5, December 2017
- Telcordia. Electromagnetic Compatibility (EMC) and Electrical Safety - Generic Criteria for Network Telecommunications Equipment. Generic Requirements, GR-1089-CORE, Issue 7, December 2017
Related Documents
Cloud Vs On Premise: Putting Leading AI Voice, Vision & Language Models to the Test in the Cloud & On Premise
Thu, 14 Mar 2024 16:49:21 -0000
|Read Time: 0 minutes
| DEPLOYING LEADING AI MODELS ON PREMISE OR IN THE CLOUD
The decision to deploy workloads either on premise or in the cloud, hinges on four pivotal factors: economics, latency, regulatory requirements, and fault tolerance. Some might distill these considerations into a more colloquial framework: the laws of economics, the laws of the land, the laws of physics, and Murphy's Law. In this multi-part paper, we won't merely discuss these principles in theory. Instead, we'll delve deeper, testing and comparing leading AI models across voice, computer vision, and large language models both on premise and in the cloud.
In part one we’ll put leading CPUs to the test, with 4th Generation Intel® Xeon® Scalable Processor both in the cloud and on premise.
| LEVERAGING INTEL® DISTRIBUTION OF OPENVINO™ TOOLKIT & CORE PINNING FOR ENHANCED PERFORMANCE
To ensure enhanced performance across the cloud and on premise, we are using the Intel® Distribution of OpenVINO™ Toolkit because it offers enhanced optimizations of AI models runs and across a broad range of platforms and leading AI frameworks.
To further enhance performance, we conducted core pinning, a process used in computing to assign specific CPU cores to specific tasks or processes.
| AWS INSTANCE SELECTION
We have selected the AWS EC2 M7i Instance, specifically the m7i.48xlarge model, part of Amazon general-purpose instances that offers a substantial amount of computing resources making it comparable to Dell™ PowerEdge™ 760xa, the on-premise solution we selected.
- Processing Power and Memory: The m7i.48xlarge Instance is equipped with 192 virtual CPUs (vCPUs) and 768 GiB of memory. This high level of processing power and memory capacity is ideal for CPU-based machine learning.
- Networking and Bandwidth: This instance provides a bandwidth of 50 Gbps, facilitating efficient data processing and transfer, essential for high-transaction and latency-sensitive workloads.
- Performance Enhancement: The M7i Instances, including the m7i.48xlarge, are powered by custom 4th Generation Intel® Xeon® Scalable Processors, also known as Sapphire Rapids.
As of November 2023, the pricing for the AWS EC2 M7i Instance, specifically the m7i.48xlarge model, starts at US$9.6768 per hour.
| HARDWARE SELECTION CONSIDERATIONS
For the cloud instance, we selected the top AWS EC2 M7i Instance with 192 virtual cores. For on premise, Dell™ PowerEdge™ portfolio offered more choice and we selected 112 physical core processor with 224 hyper threaded cores. While cloud offerings offer significant choice, Dell™ PowerEdge™ portfolio offered great choice of processors, memory, and networking.
In our analysis, we are providing performance insights as well as cost of compute comparisons. For deployment you will also want to consider the following factors:
- Operational expenditures including power and maintenance costs,
- Network costs including data transfer to cloud and local connectivity,
- Data storage costs including cloud cost versus local storage,
- Network latency requirements including lower latency as data is processed locally,
- Security and compliance costs.
| AI MODELS SELECTION
- LLama-2 7B Chat • OpenAI Whisper Base • YOLOv8n Instance Segmentation
To ensure we have a broad range of AI workloads tested on premise and in the cloud we opted for three of the leading models in their domains:
- VISION | YOLOv8n-seg
YOLOv8n-seg is model variant of YOLOv8 that is designed for instance segmentation and has 3.2 million parameters for the nano version. Unlike basic object detection instance segmentation identifies the objects in an image as well as the segments of each object and provides outlines and confidence scores.
- LANGUAGE | Llama 2 7B Chat
Llama-2 7B-chat is a member of the Llama family of large language models offered by Meta, trained on 2 trillion tokens and well suited for chat applications.
- VOICE | OpenAI Whisper base 74M
OpenAI Whisper is a deep learning model developed by OpenAI for speech recognition and transcription, capable of transcribing speech in English and multiple other languages and translating several non-English languages into English.
EDGE HARDWARE | DELL™ POWEREDGE™ R760XA RACK SERVER
The system we selected is Dell™ PowerEdge™ R760xa hardware powered by 4th Generation Intel® Xeon® Scalable Processors.
The Air-cooled design with front-facing accelerators enables better cooling Cyber Resilient Architecture for Zero Trust IT environment.
Operations Security is integrated into every phase of Dell™ PowerEdge™ lifecycle, including protected supply chain and factory-to-site integrity assurance.
Silicon-based root of trust anchors provide end-to-end boot resilience complemented by Multi-Factor Authentication (MFA) and role-based access controls to ensure secure operations. iDRAC delivers seamless automation and centralize one-to-many management.
*Performance varies by use case, model, application, hardware & software configurations, the quality of the resolution of the input data, and other factors. This performance testing is intended for informational purposes and not intended to be a guarantee of actual performance of an AI application.
| PERFORMANCE INSIGHTS
The results selected for YOLOv8n Instance Segmentation running 12 processes as that threshold achieved targeted performance of >30 images per second. Llama-2 7B Chat was selected running 2 processes as it achieved targeted sub 100 ms per token user latency. OpenAI Whisper selected running 64 processes targeting user reading speed. Across vision, language, and voice, the on premise offering exceeded the cloud instance, including offering lower latency AI performance. From a computational cost comparison the on premise solution offered a payback period of nearly a year based on dell.com pricing indicating a TCO win for on premise as well.
| RETAIL USE CASE
- Drive-thru Pharmacy Pick-up
To demonstrate the practical application of these models, we designed a solution architecture accompanied by a demo that simulates a drive-through pharmacy scenario. In this use case, the vision model identifies the car upon its arrival, the language model gathers the client's information, and communication is facilitated via the voice model. As you can discern, factors such as latency, privacy, security, and cost play crucial roles in this scenario, emphasizing the importance of the decision to deploy either in the cloud or on premise.
In our drive-thru pharmacy pick-up scenario, we utilize a comprehensive architecture to optimize the customer experience. The Video AI module employs an Intel® OpenVINO™ optimized YOLOv8n Instance Segmentation model to accurately detect and track cars in the drive-thru zone. The Audio AI segment captures and transcribes human speech into text using an Intel® OpenVINO™ optimized OpenAI whisper-base model. This transcribed text is then processed by our Large Language Models segment, where an application leverages the Intel® OpenVINO™ optimized LLama 2 7B Chat model to generate intuitive, human-like responses.
| RETAIL USE CASE ARCHITECTURE
| SUMMARY
In this analysis, we put the leading voice, language, and vision models to the test on Dell™ PowerEdge™ and AWS on CPUs. Dell™ PowerEdge™ R760xa Rack Server exceeded the cloud instances on all performance tests and offers a payback period of nearly one year based on Dell™ public pricing. The drive-through pharmacy use case showcased the advantages of an on premise deployment to maintain customer privacy, HIPPA compliance, and ensure fault tolerance and low latency. Finally, in both instances we showcased enhanced CPU performance with Intel® OpenVINO™ and core pinning. In part II, we’ll compare GPU workloads in the cloud versus on premise.
APPENDIX | PERFORMANCE TESTING DETAILS
Performance Insights | 4th Generation Intel® Xeon® Scalable Processors
- Yolov8n Instance Segmentation with Intel® OpenVINO™ & Core Pinning
| Test Methodology
YOLOv8n Instance Segmentation FP32 model is exported into the Intel® OpenVINO™ format using ultralytics 8.0.43 library and then tested for object segmentation (inference) using Intel® OpenVINO™ 2023.1.0 runtime.
For performance tests, we used a source video of 53 sec duration with resolution of 1080p and a bitrate of 1906 kb/s. The initial 30 inference samples were treated as warm-up and excluded from calculating the average inference metrics. The time collected includes H264 encode-decode using PyAV 10.0.0 and model inference time.
Output | Video file with h264 encoding (without segmentation post processing)
*Performance varies by use case, model, application, hardware & software configurations, the quality of the resolution of the input data, and other factors. This performance testing is intended for informational purposes and not intended to be a guarantee of actual performance of an AI application.
Performance Insights | 4th Gen Intel® Xeon® Scalable Processors
- Llama 2 7B Chat with Intel® OpenVINO™ & Core Pinning
| Test methodology
The Llama-2 7B Chat FP32 model is exported into the Intel® OpenVINO™ format and then tested for text generation (inference) using Hugging Face Optimum 1.13.1. Hugging Face Optimum is an extension of Hugging Face transformers and Diffusers and provides tools to export and run optimized models on various ecosystems including Intel® OpenVINO™. For performance tests, 25 iterations were executed for each inference scenario out of which initial 5 iterations were considered as warm-up and were discarded for calculating Inference time (in seconds) and tokens per second. The time collected includes encode-decode time using tokenizer and LLM inference time.
Input | Discuss the history and evolution of artificial intelligence in 80 words.
Output | Discuss the history and evolution of artificial intelligence in 80 words or less.
Artificial intelligence (AI) has a long history dating back to the 1950s when computer scientist Alan Turing proposed the Turing Test to measure machine intelligence. Since then, AI has evolved through various stages, including rule-based systems, machine learning, and deep learning, leading to the development of intelligent systems capable of performing tasks that typically require human intelligence, such as visual recognition, natural language processing, and decision-making.
Base Model | https://huggingface.co/meta-llama/Llama-2-7b-chat-hf
*Performance varies by use case, model, application, hardware & software configurations, the quality of the resolution of the input data, and other factors. This performance testing is intended for informational purposes and not intended to be a guarantee of actual performance of an AI application.
PERFORMANCE INSIGHTS | 4TH GEN INTEL® XEON® SCALABLE PROCESSORS
- OpenAI Whisper-base model with Intel® OpenVINO™ & Core Pinning
| Test methodology
The OpenAI Whisper base 74M FP32 model is exported into the Intel® OpenVINO™ format and then tested for inference using Intel® OpenVINO™. For performance tests, 25 iterations were executed for each inference scenario out of which initial 5 iterations were considered as warm-up and were discarded for calculating Inference time (in seconds) and tokens per second. The time collected includes encode-decode time using tokenizer and LLM inference time.
Input | MP3 file with 28.2 sec audio
Output | Generative AI has revolutionized the retail industry by offering a wide array of innovative use cases that enhance customer experiences and streamline operations. One prominent application of Generative AI is personalized product recommendations. Retailers can utilize advanced recommendation algorithms to analyze customer data and generate tailored product suggestions in real time. This not only drives sales but also enhances customer satisfaction by presenting them with items that align with their preferences and purchase history.
| 74 words transcribed.
Base Model | https://github.com/openai/whisper#available-models-and-languages
***Performance varies by use case, model, application, hardware & software configurations, the quality of the resolution of the input data, and other factors. This performance testing is intended for informational purposes and not intended to be a guarantee of actual performance of an AI application.
| About Scalers AI™
Scalers AI™ specializes in creating end-to-end artificial intelligence (AI) solutions to fast-track industry transformation across a wide range of industries, including retail, smart cities, manufacturing, insurance, finance, legal and healthcare. Scalers AI™ industry offering include predictive analytics, generative AI chatbots, stable diffusion, image and speech recognition, and natural language processing. As a full stack AI solutions company with solutions ranging from the cloud to the edge, our customers often need versatile common off the shelf (COTS) hardware that works well across a range of workloads.
- Fast track development & save hundreds of hours in development with access to the solution code.
As part of this effort, Scalers AI™ is making the solution code available. Reach out to your Dell™ representative or contact Scalers AI™ at contact@scalers.ai for access to GitHub repo.
Optimizing Performance Per Watt with Dell PowerEdge XR Servers
Thu, 14 Mar 2024 16:48:00 -0000
|Read Time: 0 minutes
Executive Summary
With power and cooling costs accounting for increasingly large portions of IT budgets, IT departments looking to minimize total cost of ownership (TCO) are making power efficiency a priority when choosing server hardware. This white paper will examine the power efficiency of Dell Edge servers in the multi-node, 2U form factor, a form factor that is one of the most popular in many Edge and Telecom use cases because of the balance it strikes between density and expandability. This white paper will present and analyze power efficiency results for several Dell current-generation PowerEdge XR servers and also illustrate how those results compare on various parameters with a prior-generation Dell Edge server.
The environmental conditions for telecom edge computing are typically vastly different than those at centralized data centers. Telecom edge computing sites might, at best, consist of little more than a telecommunications closet with minimal or no HVAC. Thus, ruggedized, front-access servers are ideal for such deployments. The Dell PowerEdge servers checks all of the boxes.
Dell Technologies commissioned Tolly to evaluate the power efficiency of Dell’s XR8000, XR4000, XR5610, and XR11 servers using the industry standard Standard Performance Evaluation Corporation (SPEC) SPECPower benchmark and compare those to each other. The SPECPower benchmark measures server-side Java (SSJ) throughput and system power consumption. The benchmark calculates SSJ operations per watt of system power consumed. All analysis was based on public data submitted to the SPEC and published on their website.[1]
The Dell PowerEdge XR8000, XR4000, XR5610, and XR11 are all highly-capable edge servers but offer customers different options with respect to form factor, CPU specifications, and power efficiency/cost. The following summary tables provide insights into the value each offers from a different perspective of performance, cost, and energy usage.
The first table, below, summarizes the raw performance results calculated by SPECPower. As one would expect, the newer systems deliver higher performance per watt the older systems. The XR5610[2] and XR11 were measured on 32 cores where the other two systems were measured on 64 cores.
Table 1. SPECPower - Performance/Watt
The second table, below, analyzes results on a “per core” basis as the various Dell systems have either 32 or 64 cores. The XR4000 results are 26% higher than the XR11 results, the XR8000 results are 42% higher than the XR11, and the XR5610 results are 62% higher than the XR11 roughly tracking the results shown in the previous table for the entire systems.
Table 2. SPECPower - Performance/CPU Core
The third table, below, calculates watts consumed per CPU core without reference to performance.The XR4000 and the XR11 results are within 2% of each other. The XR8000 results are13% better than the XR11 and the XR5610 results are 7% better than the XR11. Note that the XR11 is powered by an Intel 3rd Gen Xeon SP CPU while the XR4000 is powered by an Intel Xeon-D CPU.
Table 3. SPECPower - Watts/CPU Core
The fourth table, below, factored in the cost of the CPU into the perf/watt equation. Thus, lower cost CPUs will have higher values in this table when the raw performance is the same as higher cost CPUs. The XR4000 results are 120% better than the XR11 results, the XR8000 results are 110% better than the XR11, and the XR5610 results are 104% better than the XR11.
Table 4. SPECPower - Perf/Watts/CPU Cost
The fifth table, below, provides links to details of each of the CPUs evaluated.
Table 5. Dell PowerEdge Server Systems - Intel CPU Detail Links
System | CPU | Intel Reference Link |
Dell PowerEdge XR8000 (XR8620T) & Dell PowerEdge 5610 | Intel Xeon Gold 6421N, 1.80 GHz | |
Dell PowerEdge XR4000 | Intel Xeon D-2776NT, 2.10 GHz | |
Dell PowerEdge XR11 | Intel Xeon Gold 6338N, 2.2 GHz |
Competitive Positioning
Based on the publicly available data from spec.org/power, we can see high capacity data intensive workload targeted HPE and Supermicro servers. Although these are not direct competitors to Dell PowerEdge XR servers, it is worthwhile to note that the perf/watt/CPU$ for XR8000 is better than both HPE ProLiant DL360 Gen11 (Intel Xeon Platinum 8480+ 2.0 GHz), HPE ProLiant DL380 Gen 1 (Intel Xeon Platinum 8480+ 2.0 GHz), as well as the Supermicro SYS-621C-TN12R (Intel Xeon Platinum 8490H 1.90GHz).
Dell XR servers provide solutions for various edge workloads in a short form factor, edge optimized with power efficiency consideration taken into account.
Air Cooling
Dell created Multi-Vector Cooling (MVC) to maximize the potential of air cooling. It includes control algorithms, thermal and power sensors, component mapped fan zoning and airflow channeling shrouds to balance and intelligently direct airflow across the systems’ components.
New high-performance fans and heatsinks, as well as special airflow-optimized configurations, ensure even high-power CPUs are supported without throttling.
For more information, go to https://www.dell.com/en-us/blog/better-ways-to-cool-your-poweredge-servers, read this “Direct from Development” (DfD) note https://infohub.delltechnologies.com/p/understanding-thermal-design-and-capabilities-for-the-poweredge-xr8000-server, or view a video on the topic at: https://www.youtube.com/watch?v=-rHEXJsX75Y&ab_channel=DellTechnologies.
Telecom Edge Computing
Wireless telecom providers world-wide have at least two things in common: seemingly endless growth, and the rapid migration from specialized, proprietary radio access network (RAN) hardware to scalable, software-based vRAN solutions. Over two dozen system operators and nearly 300 related companies and academic institutions are part of the Open RAN Alliance (O-RAN) working together to bring an open solution to the industry.[3]
The telecom edge, thus, needs ruggedized servers built to resist demanding environmental conditions while delivering significant compute power with cost-efficient use of electric power.
Dell, an acknowledged information technology leader, builds servers that are designed for both the processing requirements and physical deployment requirements of edge servers with a particular focus on telecom applications. In particular, the Dell PowerEdge XR8000 and Dell PowerEdge XR4000 edge servers provides a powerful and flexible selection of configurations focused on the particular needs of the telecom edge.[4]
- Built to withstand extreme heat & dust; operating temperature range from -5 to 55C
- Efficient use of electric power
- Suitable for shock and vibration of factory floors & construction site
- Can be deployed in distributed telecom and other extreme environments
- Short depth (355mm), small form factor
- Ruggedized; tested for NEBS and MIL-STD
- Multi-node capable
PowerEdge XR4000: Scalability and Flexibility with HCI Capabilities
The Dell PowerEdge XR4000 Edge Server is part of Dell’s family of purpose-built, ruggedized servers. The PowerEdge XR4000 is built for environments like telecom edge deployment or factory floors where the servers could be subjected to demanding conditions including high temperatures, dust, shock and vibrations.
The high-performance, multinode XR4000 server was purpose built to address the demands of today’s retail, manufacturing and defense customers. It was designed around a unique chassis and compute sled(s) concept. The actual computer resides in modular 1U or 2U sled form factors. The only shared component between the sleds is power. The server is also designed to support hyperconverged infrastructure (HCI).
The XR4000 is available in two 14" depth “rackable” and “stackable” chassis form factors. The optional nano server sled replaces the need for a virtual witness node. The in-chassis witness node allows for native, two-node vSAN clusters in the stackable server chassis.
The servers are small form factor, short depth units that can be deployed alone or in multi-node configurations.
The XR4000 used for this test was an XR4520c 2U compute sled. See table below for key specifications.
Table 6. Dell PowerEdge XR4520 Compute Sled Key Specifications
PowerEdge XR8000: Flexible, Innovative, Sled-based RAN-Optimized Server
The Dell PowerEdge XR8000 Edge Server is the newest addition Dell’s family of purpose-built, ruggedized servers. The PowerEdge XR4000 is built for environments like telecom edge environments where the servers could be subjected to demanding conditions including high temperatures, dust, shock and vibrations.
The short-depth XR8000 server, which comes in a sledded server architecture (with 1U and 2U single-socket form factors), is optimized for total cost of ownership (TCO) and performance in O-RAN (radio access network) applications. It is RAN optimized with integrated networking and 1/0 PTP/SyncE support. And its front-accessible design radically simplifies sled serviceability in the field.
The XR8000 offers options for multiple sled form factors with up to four nodes per chassis that can work together or independently. The 2U half-width sled configuration accommodates general purpose compute at the edge / far edge, while the 1U half-width sled configuration is ideal for dense compute and network edge-optimized workloads.
Table 7. Dell PowerEdge XR8620 Compute Sled Key Specifications
The XR8000 delivers extended tolerance to heat and cold with enhanced heatsinks and optimized airflow design. The system supports Sapphire Rapids SP and Edge Enhanced (EE) processors with Intel vRAN Boost, on-chip acceleration and includes both DC and AC power supply options and five total power supply unit (PSU) variants
PowerEdge XR5610: All-Purpose, Rugged 1U Edge Server
The Dell PowerEdge XR8000 Edge Server is a new addition Dell’s family of purpose-built, ruggedized servers. As with the PowerEdge XR8000 and PowerEdge XR4000, the PowerEdge XR5610 is built for environments where the servers could be subjected to demanding conditions including high temperatures, dust, shock and vibrations. The XR5610 is the upgraded successor to the XR11 that is also covered in this report.
The PowerEdge XR5610 is a 1U, single-socket server designed for target workloads in networking and communication, enterprise edge, military, and defense. It is well suited for 5G vRAN and ORAN telecom workloads, as well as military and defense deployments and retail AI including video monitoring, IoT device aggregation and PoS analytics. The design specification supports continuous operation in extreme temperatures ranging from -5C to 55C. The design is ruggedized, compliant, and compact.
The server features a filtered smart bezel for dust reduction and the server has undergone MIL810H and NEBS Level 3 testing for handling shocks and vibrations.
Table 8. Dell PowerEdge XR5610 Key Specifications
SPECPower Workload & Results
The Standard Performance Evaluation Corporation (SPEC), according to their website, “is a non-profit corporation formed to establish, maintain and endorse standardized benchmarks and tools to evaluate performance and energy efficiency for the newest generation of computing systems. SPEC develops benchmark suites and also reviews and publishes submitted results from our member organizations and other benchmark licensees.”
SPEC has established benchmarks, to date, in some nine different areas. In addition to power, the focus of this report, the benchmarks include Machine Learning, High Performance Computing, Virtualization, and more.
Server vendors run the benchmark tests in their own labs according to the SPEC benchmark specifications. Vendors may use the results internally and/or they can submit the results to SPEC for review and publication. Once published, the results are freely available and can be used by others in public reports so long as that use complies with the SPEC “Fair Use Policy” for the given benchmark.
SPECPower_ssj2008 Benchmark
As evidenced by its name, the SPECPower benchmark was issued in 2008. The workload, represented in the name by “ssj,” is “Server Side Java (SSJ).“ The benchmark drives the load on the target server while also measuring the power consumption of the server.
While the benchmark allows for different java virtual machines (JVM) to be used in the benchmark, the Oracle JVM is used almost exclusively for the tests. The results document CPU and memory configurations of the systems and reports “submeasurements” of SSL operations at 100% CPU, average watts consumed at 100%, and average watts at idle. The result reported is the overall SSJ operations divided by the watts consumed.
It is important to note that the test is run at 10 different loads from 10% to 100% in increments of 10% load. Only the 100% results are displayed in the SPECPower results table but the SPECPower “result” value is an average of all ten tests.
Raw Results
All results referenced in this report are available to the general public on the SPEC site at: https://www.spec.org/power_ssj2008/results. The information in the following tables is excerpted from the public results. The table, below, contains the submeasurements and the final result for each system discussed in the paper. All other results in this paper are calculated using the the SPECPower raw results below.
Table 9. SPEC SPECPower_ssj2008 Results
Server Specifications
The table, below, contains the server system specifications as shown on the SPEC results website. All systems were tested using Oracle Corporation’s JVM.
Table 10. Server System Specifications
System BIOS Settings
The tests used Dells recommended BIOS settings for power efficiency. The Dell PowerEdge XR8000 and Dell PowerEdge XR4000 systems both used the following BIOS settings.
Table 11. Server System BIOS Settings
XR Series Price/Power Efficiency Claims
The charts below visualize the tabular results presented in the Executive Summary section earlier in this report.
Performance/Watt (Performance-to-Power-Ratio)
Performance/CPU Core
Watt/CPU Core
Performance/Watt/CPU Cost
About Tolly
The Tolly Group companies have been delivering world-class IT services for more than 30 years. Tolly is a leading global provider of third-party validation services for vendors of IT products, components and services.
You can reach the company by E-mail at sales@tolly.com, or by telephone at +1 561.391.5610.
Visit Tolly on the Internet at: http://www.tolly.com
Terms of Usage
This document is provided, free-of-charge, to help you understand whether a given product, technology or service merits additional investigation for your particular needs. Any decision to purchase a product must be based on your own assessment of suitability based on your needs. The document should never be used as a substitute for advice from a qualified IT or business professional. This evaluation was focused on illustrating specific features and/or performance of the product(s) and was conducted under controlled, laboratory conditions. Certain tests may have been tailored to reflect performance under ideal conditions; performance may vary under real-world conditions. Users should run tests based on their own real-world scenarios to validate performance for their own networks.
Reasonable efforts were made to ensure the accuracy of the data contained herein but errors and/or oversights can occur. The test/audit documented herein may also rely on various test tools the accuracy of which is beyond our control. Furthermore, the document relies on certain representations by the sponsor that are beyond our control to verify. Among these is that the software/hardware tested is production or production track and is, or will be, available in equivalent or better form to commercial customers. Accordingly, this document is provided "as is", and Tolly Enterprises, LLC (Tolly) gives no warranty, representation or undertaking, whether express or implied, and accepts no legal responsibility, whether direct or indirect, for the accuracy, completeness, usefulness or suitability of any information contained herein. By reviewing this document, you agree that your use of any information contained herein is at your own risk, and you accept all risks and responsibility for losses, damages, costs and other consequences resulting directly or indirectly from any information or material available on it. Tolly is not responsible for, and you agree to hold Tolly and its related affiliates harmless from any loss, harm, injury or damage resulting from or arising out of your use of or reliance on any of the information provided herein.
Tolly makes no claim as to whether any product or company described herein is suitable for investment. You should obtain your own independent professional advice, whether legal, accounting or otherwise, before proceeding with any investment or project related to any information, products or companies described herein. When foreign translations exist, the English document is considered authoritative. To assure accuracy, only use documents downloaded directly from Tolly.com. No part of any document may be reproduced, in whole or in part, without the specific written permission of Tolly. All trademarks used in the document are owned by their respective owners. You agree not to use any trademark in or as the whole or part of your own trademarks in connection with any activities, products or services which are not ours, or in a manner which may be confusing, misleading or deceptive or in a manner that disparages us or our information, projects or developments.
Tolly Report #223124
August 2023
© 2023 TOLLY ENTERPRISES, LLC www.tolly.com
[2] At publication time the XR5610 results were being prepared for submission to SPEC and should appear later in Q3 2023.