Hetzner - ARM64 4 Core 8 GB Ram DE
Comprehensive VPS benchmark results from Germany tested on Jun 05, 2023. Learn our methodology.
System Specifications
Additional Benchmarks
UnixBench
System Performance Index
WordPress Bench
Web Application Performance
Benchmark Steps
| # | Suite | Run At | Raw Output |
|---|---|---|---|
| 1 |
Bench
|
Jun 05, 2023 07:27 |
View Raw Output-------------------- A Bench.sh Script By Teddysun ------------------- Version : v2022-06-01 Usage : wget -qO- bench.sh | bash ---------------------------------------------------------------------- CPU Model : CPU model not detected CPU Cores : 4 AES-NI : Enabled VM-x/AMD-V : Disabled Total Disk : 75.0 GB (3.2 GB Used) Total Mem : 7.6 GB (135.8 MB Used) System uptime : 0 days, 1 hour 18 min Load average : 0.00, 0.25, 0.98 OS : Ubuntu 20.04.6 LTS Arch : aarch64 (64 Bit) Kernel : 5.4.0-150-generic TCP CC : bbr Virtualization : Dedicated Organization : AS24940 Hetzner Online GmbH Location : Gunzenhausen / DE Region : Bavaria ---------------------------------------------------------------------- I/O Speed(1st run) : 1.1 GB/s I/O Speed(2nd run) : 1.3 GB/s I/O Speed(3rd run) : 1.4 GB/s I/O Speed(average) : 1297.1 MB/s ---------------------------------------------------------------------- Node Name Upload Speed Download Speed Latency Speedtest.net 649.10 Mbps 580.73 Mbps 3.05 ms Los Angeles, US 508.31 Mbps 3172.61 Mbps 159.82 ms Dallas, US 620.26 Mbps 4872.36 Mbps 130.99 ms Montreal, CA 664.15 Mbps 937.92 Mbps 102.18 ms Paris, FR 4427.09 Mbps 7353.27 Mbps 18.44 ms Amsterdam, NL 5700.72 Mbps 9223.78 Mbps 11.65 ms Shanghai, CN 367.52 Mbps 3537.05 Mbps 221.44 ms Hongkong, CN 340.44 Mbps 3671.22 Mbps 199.48 ms Singapore, SG 483.80 Mbps 342.53 Mbps 157.29 ms Tokyo, JP 348.05 Mbps 950.48 Mbps 225.52 ms ---------------------------------------------------------------------- Finished in : 5 min 6 sec Timestamp : 2023-06-05 07:27:59 UTC ---------------------------------------------------------------------- |
| 2 |
PassMark
|
Jun 05, 2023 07:27 |
View Raw OutputPassMark PerformanceTest Linux Neoverse-N1 (aarch64) 4 cores @ 0 MHz | 7.6 GiB RAM Number of Processes: 4 | Test Iterations: 2 | Test Duration: Very Long -------------------------------------------------------------------------- CPU Mark: 3198 Integer Math 18792 Million Operations/s Floating Point Math 14369 Million Operations/s Prime Numbers 35.4 Million Primes/s Sorting 12260 Thousand Strings/s Encryption 640 MB/s Compression 22718 KB/s CPU Single Threaded 1289 Million Operations/s Physics 669 Frames/s Extended Instructions (NEON) 2977 Million Matrices/s Memory Mark: 1814 Database Operations 2898 Thousand Operations/s Memory Read Cached 15025 MB/s Memory Read Uncached 10545 MB/s Memory Write 11312 MB/s Available RAM 6769 Megabytes Memory Latency 71 Nanoseconds Memory Threaded 39532 MB/s -------------------------------------------------------------------------- |
| 3 |
Speedtest Global
|
Jun 05, 2023 07:47 |
View Raw Output---------------------------- network-speed.xyz ----------------------------
A simple script to test network performance using speedtest-cli
---------------------------------------------------------------------------
Version : v2023.05.28
Global Speedtest : wget -qO- network-speed.xyz | bash
Region Speedtest : wget -qO- network-speed.xyz | bash -s -- -r <region>
---------------------------------------------------------------------------
Basic System Info
---------------------------------------------------------------------------
CPU Model : CPU model not detected
CPU Cores : 4
AES-NI : ✔ Enabled
VM-x/AMD-V : ❌ Disabled
Total Disk : 75.0 GB (3.2 GB Used)
Total RAM : 7.6 GB (138.6 MB Used)
System uptime : 0 days, 1 hour 31 min
Load average : 0.06, 0.14, 0.53
OS : Ubuntu 20.04.6 LTS
Arch : aarch64 (64 Bit)
Kernel : 5.4.0-150-generic
Virtualization : KVM
---------------------------------------------------------------------------
Basic Network Info
---------------------------------------------------------------------------
Primary Network : IPv6
ISP : Hetzner Online GmbH
ASN : AS24940 Hetzner Online GmbH
Host : Hetzner
Location : Falkenstein, Saxony-SN, Germany
Location (IPv4) : Gunzenhausen, Bavaria, DE
---------------------------------------------------------------------------
Speedtest.net (Region: GLOBAL)
---------------------------------------------------------------------------
Location Latency Loss DL Speed UP Speed Server
ISP: Hetzner Online
Nearest 3.08 ms 0.0% 87.80 Mbps 691.66 Mbps WebseitenDesigner.com - Hannover
Kochi, IN 166.39 ms 0.0% 3572.63 Mbps 467.80 Mbps Asianet Broadband - Cochin
Bangalore, IN 138.01 ms 0.0% 5151.46 Mbps 434.28 Mbps Bharti Airtel Ltd - Bangalore
Chennai, IN 143.06 ms N/A 5477.89 Mbps 543.41 Mbps Jio - Chennai
Mumbai, IN 117.70 ms 0.0% 6446.33 Mbps 725.28 Mbps i3D.net - Mumbai
Delhi, IN 176.41 ms 0.0% 4860.72 Mbps 483.66 Mbps Tata Play Fiber - New Delhi
Los Angeles, US 155.87 ms 0.0% 4932.69 Mbps 535.60 Mbps ReliableSite Hosting - Los Angeles, CA
Dallas, US 128.11 ms 0.0% 6192.37 Mbps 635.51 Mbps Hivelocity - Dallas, TX
New York, US 85.49 ms 0.0% 842.17 Mbps 947.74 Mbps Clouvider Ltd - New York, NY
Seattle, US 172.53 ms 0.0% 1193.35 Mbps 465.35 Mbps DediPath - Seattle, WA
Miami, US 125.61 ms 0.0% 2745.45 Mbps 627.36 Mbps AT&T - Miami, FL
Toronto, CA 105.53 ms 0.0% 4965.03 Mbps 751.50 Mbps Rogers - Toronto, ON
Paris, FR 18.41 ms 0.0% 4891.98 Mbps 4050.08 Mbps ORANGE FRANCE - Paris
Amsterdam, NL 10.48 ms 0.0% 8436.14 Mbps 7400.54 Mbps Clouvider Ltd - Amsterdam
Warsaw, PL 20.18 ms 0.0% 9225.83 Mbps 4003.00 Mbps UPC Polska - Warszawa
London, UK 26.18 ms 0.0% 9060.42 Mbps 3360.75 Mbps VeloxServ Communications - London
Frankfurt, DE 5.69 ms 0.0% 9286.48 Mbps 7144.55 Mbps 23M GmbH - Frankfurt Am Main
Bucharest, RO 33.21 ms 0.0% 4069.11 Mbps 2458.20 Mbps Vodafone Romania Fixed - Bucharest - Bucharest
Dubai, AE 142.20 ms 0.0% 5627.66 Mbps 573.89 Mbps du - Dubai
Fujairah, AE 115.05 ms 0.0% 6980.60 Mbps 707.03 Mbps ETISALAT-UAE - Fujairah
Jeddah, KSA 81.56 ms 0.0% 8768.71 Mbps 990.98 Mbps Saudi Telecom Company
Shanghai, CU-CN 221.45 ms 0.0% 3612.55 Mbps 368.69 Mbps China Unicom 5G - ShangHai
Nanjing, CT-CN FAILED
Hong Kong, HKG 164.12 ms N/A 715.31 Mbps 495.22 Mbps STC - Hong Kong
Singapore, SG 160.37 ms 0.0% 837.95 Mbps 502.34 Mbps i3D.net - Singapore
Jakarta, ID 183.87 ms 0.0% 3921.20 Mbps 405.10 Mbps PT. Telekomunikasi Indonesia - Jakarta
Tokyo, JP FAILED
---------------------------------------------------------------------------
Avg DL Speed : 4876.08 Mbps
Avg UL Speed : 1590.78 Mbps
Total DL Data : 170.84 GB
Total UL Data : 48.97 GB
Total Data : 219.81 GB
---------------------------------------------------------------------------
Duration : 12 min 5 sec
System Time : 05/06/2023 - 07:47:23 UTC
Total Script Runs : 3107
---------------------------------------------------------------------------
Result : https://cdn1.frocdn.ch/TTdzspBXHQokqms.txt
---------------------------------------------------------------------------
|
| 4 |
UnixBench
|
Jun 05, 2023 09:22 |
View Raw Output------------------------------------------------------------------------
Benchmark Run: Mon Jun 05 2023 09:22:35 - 09:50:29
4 CPUs in system; running 1 parallel copy of tests
Dhrystone 2 using register variables 41012008.6 lps (10.0 s, 7 samples)
Double-Precision Whetstone 8421.3 MWIPS (9.7 s, 7 samples)
Execl Throughput 5400.7 lps (30.0 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 960287.5 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 267088.5 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 1968487.0 KBps (30.0 s, 2 samples)
Pipe Throughput 1633720.1 lps (10.0 s, 7 samples)
Pipe-based Context Switching 92068.1 lps (10.0 s, 7 samples)
Process Creation 5747.0 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 6197.6 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 3079.8 lpm (60.0 s, 2 samples)
System Call Overhead 1344009.9 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 41012008.6 3514.3
Double-Precision Whetstone 55.0 8421.3 1531.1
Execl Throughput 43.0 5400.7 1256.0
File Copy 1024 bufsize 2000 maxblocks 3960.0 960287.5 2425.0
File Copy 256 bufsize 500 maxblocks 1655.0 267088.5 1613.8
File Copy 4096 bufsize 8000 maxblocks 5800.0 1968487.0 3393.9
Pipe Throughput 12440.0 1633720.1 1313.3
Pipe-based Context Switching 4000.0 92068.1 230.2
Process Creation 126.0 5747.0 456.1
Shell Scripts (1 concurrent) 42.4 6197.6 1461.7
Shell Scripts (8 concurrent) 6.0 3079.8 5133.0
System Call Overhead 15000.0 1344009.9 896.0
========
System Benchmarks Index Score 1445.5
------------------------------------------------------------------------
Benchmark Run: Mon Jun 05 2023 09:50:29 - 10:18:25
4 CPUs in system; running 4 parallel copies of tests
Dhrystone 2 using register variables 164189609.3 lps (10.0 s, 7 samples)
Double-Precision Whetstone 33700.6 MWIPS (9.8 s, 7 samples)
Execl Throughput 13180.8 lps (29.9 s, 2 samples)
File Copy 1024 bufsize 2000 maxblocks 2306274.7 KBps (30.0 s, 2 samples)
File Copy 256 bufsize 500 maxblocks 641027.5 KBps (30.0 s, 2 samples)
File Copy 4096 bufsize 8000 maxblocks 6029398.7 KBps (30.0 s, 2 samples)
Pipe Throughput 6557523.8 lps (10.0 s, 7 samples)
Pipe-based Context Switching 895239.5 lps (10.0 s, 7 samples)
Process Creation 20480.9 lps (30.0 s, 2 samples)
Shell Scripts (1 concurrent) 20838.2 lpm (60.0 s, 2 samples)
Shell Scripts (8 concurrent) 3513.4 lpm (60.0 s, 2 samples)
System Call Overhead 5416214.1 lps (10.0 s, 7 samples)
System Benchmarks Index Values BASELINE RESULT INDEX
Dhrystone 2 using register variables 116700.0 164189609.3 14069.4
Double-Precision Whetstone 55.0 33700.6 6127.4
Execl Throughput 43.0 13180.8 3065.3
File Copy 1024 bufsize 2000 maxblocks 3960.0 2306274.7 5823.9
File Copy 256 bufsize 500 maxblocks 1655.0 641027.5 3873.3
File Copy 4096 bufsize 8000 maxblocks 5800.0 6029398.7 10395.5
Pipe Throughput 12440.0 6557523.8 5271.3
Pipe-based Context Switching 4000.0 895239.5 2238.1
Process Creation 126.0 20480.9 1625.5
Shell Scripts (1 concurrent) 42.4 20838.2 4914.7
Shell Scripts (8 concurrent) 6.0 3513.4 5855.7
System Call Overhead 15000.0 5416214.1 3610.8
========
System Benchmarks Index Score 4723.9
|
Onidel Cloud VPS
Performance-first cloud hosting for teams that need consistent benchmarks and transparent pricing.
- 99.99% uptime SLA
- Proven availability for critical workloads
- Plans from $4.95/month
- Predictable pricing and simple scaling
- AMD EPYC + NVMe stack
- Modern hardware with built-in DDoS protection
- 5 global regions
- Singapore, Sydney, Amsterdam, Ho Chi Minh City, more coming soon
Ideal for: demanding SaaS workloads, managed hosting agencies, and teams migrating off legacy VPS providers.
Launch faster with concierge onboarding
Start 14-day TrialCPU Performance (Geekbench)
CPU benchmarks measure processing power for compute-intensive tasks. Learn how we test CPU performance.
I/O Speed Summary (dd)
IOPS Performance by Block Size
Combined IOPS Performance
Total IOPS across all block sizes (4K, 64K, 512K, 1M)
FIO Benchmark Details
Detailed I/O performance test results
Test 1 - Block Size: 4k (mixed_rw_50_50)
Test 2 - Block Size: 64k (mixed_rw_50_50)
Test 3 - Block Size: 512k (mixed_rw_50_50)
Test 4 - Block Size: 1m (mixed_rw_50_50)
Detailed I/O Benchmark (FIO)
IOPS Performance by Block Size
Bandwidth Performance by Block Size
Detailed Test Results
| Block Size | Read IOPS | Write IOPS | Read MB/s | Write MB/s |
|---|---|---|---|---|
| 4k | 32,700 | 32,700 | 131.1 | 131.1 |
| 64k | 17,800 | 18,300 | 1,140.0 | 1,170.0 |
| 512k | 4,300 | 4,700 | 2,230.0 | 2,420.0 |
| 1m | 2,000 | 2,300 | 2,130.0 | 2,380.0 |
Related Resources
Browse More Benchmarks
Explore comprehensive VPS benchmarks across providers and locations.
Compare Providers
View aggregated performance metrics across all VPS providers.
Performance Dashboard
Access real-time statistics, top performers, and latest results.
Testing Methodology
Learn about our comprehensive testing procedures and tools.
Understanding VPS Benchmark Results
This comprehensive benchmark report provides detailed performance analysis for Hetzner ARM64 4 Core 8 GB Ram DE VPS in Germany. Our standardized testing methodology measures CPU performance, disk I/O operations, network bandwidth, and real-world application performance to give you complete insights into VPS capabilities. Browse more VPS benchmarks to compare performance across providers.
Each benchmark metric serves a specific purpose in evaluating VPS performance. CPU benchmarks reveal processing power for applications, disk I/O tests measure storage performance for databases and file operations, network tests show bandwidth capacity for content delivery, and application benchmarks demonstrate real-world hosting performance. Together, these metrics provide a complete performance profile. Use our VPS comparison tool to analyze multiple plans side-by-side.
Benchmark Metrics & Features
Performance Analysis Use Cases
- Evaluating VPS CPU performance for compute-intensive applications
- Testing disk I/O performance for database workloads
- Comparing VPS performance across different providers
- Analyzing WordPress site performance on specific VPS plans
- Assessing value-for-money based on performance and pricing
- Checking real-world application performance metrics
- Understanding geographic performance variations by location
- Reviewing detailed benchmark metrics before VPS purchase
- Comparing single-core vs multi-core CPU performance
- Validating provider performance claims with actual data
- Making informed hosting decisions based on comprehensive benchmarks
How to Interpret Benchmark Results
1. CPU Performance (Geekbench)
Geekbench measures CPU performance through single-core and multi-core scores. Single-core scores (typically 800-2000) indicate performance for single-threaded applications and general responsiveness. Multi-core scores (1500-8000) show performance when all CPU cores are utilized for parallel processing. Higher scores mean better performance. Visit our testing methodology to learn more about CPU benchmarks.
2. Disk I/O Performance (FIO)
FIO benchmarks measure disk performance through IOPS (Input/Output Operations Per Second) and bandwidth (MB/s). High IOPS (10,000+ for NVMe, 5,000-10,000 for SSD) is crucial for database workloads with many small operations. High bandwidth is important for large file transfers and media applications. Compare disk performance across different providers in our database.
3. Application Performance
Real-world application benchmarks (WordPress, UnixBench) show how the VPS handles typical web workloads including PHP processing, database queries, and static file serving. These results measured in requests per second and page load times directly apply to web hosting scenarios. Need help choosing the right VPS? Visit about us to learn more about VPSMetrics.
Frequently Asked Questions
How do I interpret the Geekbench single-core and multi-core scores?
Geekbench single-core scores indicate CPU performance for single-threaded applications, important for general responsiveness and web applications. Multi-core scores show performance when all CPU cores are utilized, crucial for parallel processing and multi-threaded applications. Higher scores indicate better performance. Modern VPS typically score 800-2000 (single) and 1500-8000 (multi) depending on the plan.
What do IOPS numbers mean in disk I/O benchmarks?
IOPS (Input/Output Operations Per Second) measures how many read/write operations your disk can handle per second. Higher IOPS means better performance for database operations and applications with many small file operations. NVMe SSD VPS typically delivers 10,000+ IOPS, while traditional SSD offers 5,000-10,000 IOPS.
What is UnixBench and why does it matter?
UnixBench is a comprehensive system-level benchmark suite that tests various aspects of Unix-like system performance including process creation, file operations, and system calls. It provides a holistic view of overall system performance beyond just CPU or disk metrics. Higher UnixBench scores indicate better overall system efficiency.
How reliable are WordPress benchmark results?
WordPress benchmarks provide real-world application performance metrics by testing actual WordPress installations. These results show how the VPS handles typical web application workloads including PHP processing, database queries, and static file serving. Results are measured in requests per second and page load times, directly applicable to web hosting scenarios.
Can I trust these benchmark results for my use case?
Our benchmarks provide standardized, reproducible test results using industry-standard tools. While they offer excellent performance indicators, your actual experience may vary based on your specific workload, application optimization, and usage patterns. Use these benchmarks as a reliable starting point for comparison between providers.
Why do benchmark results vary by location?
Different data center locations may have different hardware configurations, CPU generations, storage types, and network infrastructure. Providers often upgrade facilities at different rates across regions. Location-specific benchmarks help you choose a VPS in the right region with the best performance for your target audience.
Making Informed VPS Decisions
Use these comprehensive benchmark results to make data-driven hosting decisions. Compare multiple metrics together rather than focusing on a single aspect - a balanced VPS performs well across CPU, disk, and network benchmarks. Consider your specific workload requirements: CPU-intensive applications need high Geekbench scores, database-heavy applications require high IOPS, and content delivery services benefit from high network bandwidth.
Our benchmarks provide standardized, reproducible test results using industry-standard tools. While they offer excellent performance indicators, your actual experience may vary based on specific workload patterns and application optimization. Use these benchmarks as a reliable starting point for comparison between providers and configurations.
Compare More VPS Options
Explore additional benchmarks, compare multiple VPS plans side-by-side, and discover the best performing options for your specific requirements and budget.
