E20ps_v5 by Microsoft Azure

E20ps_v5 is a EPSv5 family [ARM processor] [Premium Storage capable] 20 vCPUs server offered by Microsoft Azure with 20 vCPUs, 160 GiB of memory and 0 GB of storage. The pricing starts at 0.1201 USD per hour.
20 vCPU
160 GiB Memory
Spare SCore
36,041
(All-cores)
1,823
(Single-core)

Specifications

Server Metadata

Vendor ID
azure
Server ID
Standard_E20ps_v5
Name
E20ps_v5
API Reference
Standard_E20ps_v5
Description
EPSv5 family [ARM processor] [Premium Storage capable] 20 vCPUs
Family
EPSv5
Hw Virt
Average Time To Start
22
Status
active
Observed At
2026-10-25T01:32:09.326726

Availability

REGION / IDSPOTONDEMAND
Central India (IN) / centralindia
0.1201 USD/h 0.65 USD/h
Jio India Central (IN) / jioindiacentral
- 0.65 USD/h
Jio India West (IN) / jioindiawest
- 0.65 USD/h
East US 2 (US) / eastus2
0.1863 USD/h 1.008 USD/h
North Central US (US) / northcentralus
0.1863 USD/h 1.008 USD/h
West US 2 (US) / westus2
0.1863 USD/h 1.008 USD/h
West US 3 (US) / westus3
0.1863 USD/h 1.008 USD/h
East US (US) / eastus
0.2131 USD/h 1.008 USD/h
Sweden Central (SE) / swedencentral
0.1977 USD/h 1.07 USD/h
Canada Central (CA) / canadacentral
0.204 USD/h 1.104 USD/h
West US (US) / westus
0.207 USD/h 1.12 USD/h
North Europe (IE) / northeurope
0.2085 USD/h 1.128 USD/h
Central US (US) / centralus
0.2104 USD/h 1.139 USD/h
France Central (FR) / francecentral
0.2181 USD/h 1.18 USD/h
UK South (GB) / uksouth
0.1891 USD/h 1.184 USD/h
Australia East (AU) / australiaeast
0.2232 USD/h 1.208 USD/h
South Central US (US) / southcentralus
0.2236 USD/h 1.21 USD/h
West Central US (US) / westcentralus
0.2236 USD/h 1.21 USD/h
Germany West Central (DE) / germanywestcentral
0.2247 USD/h 1.216 USD/h
Japan East (JP) / japaneast
0.2247 USD/h 1.216 USD/h
Southeast Asia (SG) / southeastasia
0.2247 USD/h 1.216 USD/h
West Europe (NL) / westeurope
0.2247 USD/h 1.216 USD/h
Switzerland North (CH) / switzerlandnorth
0.2473 USD/h 1.338 USD/h
East Asia (HK) / eastasia
0.2476 USD/h 1.34 USD/h

Processor

vCPUs
20
Hypervisor
Microsoft Hyper-V
CPU Allocation
Dedicated
CPU Cores
20
CPU Speed
3 GHz
CPU Architecture
arm64
CPU Manufacturer
Ampere
CPU Family
ARMv8
CPU Model
Neoverse-N1
CPU L1D Cache
64 KiB
CPU L1D Cache Total
1 MiB
CPU L1I Cache
64 KiB
CPU L1I Cache Total
1 MiB
CPU L2 Cache
1 MiB
CPU L2 Cache Total
20 MiB
CPU L3 Cache
32 MiB
CPU L3 Cache Total
32 MiB
CPU Flags
fp, asimd, evtstrm, aes, pmull, sha1, sha2, crc32, atomics, fphp, asimdhp, cpuid, asimdrdm, lrcpc, dcpop, asimddp
Ecpus
19.8
Scalability
99

System Resources and Accelerators

MEMORY
Memory Amount
160 GiB
Memory Amount Actual
160 GiB
ACCELERATORS
Accelerator Count
0
Accelerator Memory Min
0 MiB
Accelerator Memory Total
0 MiB
Accelerators
    STORAGE
    Storage Size
    0 GB
    Storages
      NETWORK
      Inbound Traffic
      0 GB/month
      Outbound Traffic
      0 GB/month
      IPv4
      0

      CPU and System Topology

      Server Description

      An ARM-powered virtual machine featuring a high memory-to-core ratio and dedicated physical cores for memory-intensive and database workloads.

      Memory OptimizedGeneral Purpose

      Microsoft Azure E20ps_v5 is an ARM-based virtual machine from the EPSv5 family, configured with 20 dedicated physical cores on an Ampere Altra ARMv8 Neoverse-N1 processor running at 3.0 GHz. It features 160.0 GB of RAM, establishing a high memory-to-vCPU ratio of 8.0 GB per core. While it lacks local storage and integrated GPUs, it is Premium Storage capable. Performance benchmarks indicate average multi-core capabilities, with a stress-ng multi-core score of 36041.2 and a Redis SET throughput of 27.33 million operations per second. However, single-core performance is a tradeoff, with its Geekbench single-core score falling into the bottom 25% tier. This server type is qualitatively cost-efficient for memory-intensive, highly parallelized workloads such as database hosting, static web serving, and multi-threaded data processing that benefit from dedicated ARM cores without requiring x86 instruction sets.

      Economics

      Average Price per Region

      Prices per Zone

      Lowest Prices

      Performance

      Workload Profiles

      1.2Score
      Precomputed compound score for Cache Intensive workloads. A weighted average (geometric mean) of benchmark scores compared to their medians: score = ∏ (x_i / m_i)^(w_i / Σw). The score of 1.0 represents a synthetic baseline server with the median performance of each component benchmark; 0.5 means roughly half the performance; and 2.0 means twice the performance of that reference profile. Component weights: 50% Redis RPS (pipeline=1, SET), 20% Redis RPS (pipeline=16, SET), 10% PassMark Memory Mark (composite), 10% Memory bandwidth (read, 16 MB ~ L3), 10% PassMark single-thread CPU. Rationale for component selection: In-memory key-value store workload, mixing direct Redis performance metrics with memory speed and latency benchmarks, and single-core CPU performance profiles.
      Component Score Weight Impact
      Raw Reference Normalized Target Actual
      Redis RPS (pipeline=1, SET) 2,689,395 ops/sec 1,919,289 ops/sec 1.4 50.00% 50.00% +18.00%
      Redis RPS (pipeline=16, SET) 16,108,654 ops/sec 11,728,058 ops/sec 1.3 20.00% 20.00% +5.40%
      PassMark Memory Mark (composite) 1,900 2,416 0.79 10.00% 10.00% -2.30%
      Memory bandwidth (read, 16 MB ~ L3) 141,161 MB/sec 108,466 MB/sec 1.3 10.00% 10.00% +2.70%
      PassMark single-thread CPU 1,329 Mops/s 2,365 Mops/s 0.54 10.00% 10.00% -6.00%

      Memory Bandwidth

      Compression

      OpenSSL

      Geekbench Single-Core

      Score: 1,100

      Geekbench Multi-Core

      Score: 10,292

      Passmark CPU Scores

      BENCHMARKSCORE
      Mark
      		 13,904
      Compression
      		 112,739
      Encryption
      		 3,252
      Extended Instructions
      		 14,771
      Floating Point Maths
      		 71,303
      Integer Maths
      		 93,699
      Physics
      		 1,934
      Prime Numbers
      		 111
      Single Threaded
      		 1,329
      String Sorting
      		 66,729

      Passmark Memory Scores

      BENCHMARKSCORE
      Memory Mark
      		 1,900
      Database Operations
      		 13,196
      Memory Latency
      		 74
      Memory Read Cached
      		 11,310
      Memory Read Uncached
      		 8,626
      Memory Write
      		 14,922

      Stress-ng Stressors

      CPU integerSingleMultiScaling
      atomic
      		 97.63 43.87 0.45×
      bitops
      		 108,522 2,166,824 19.97×
      branch
      		 220,872 4,410,812 19.97×
      funccall
      		 1,337 26,731 20.00×
      funcret
      		 20,842 416,028 19.96×
      hash
      		 128,223 2,558,854 19.96×
      intmath
      		 30,388 629,503 20.72×
      nop
      		 173,586 3,464,400 19.96×
      prime
      		 336,918 6,717,054 19.94×
      rotate
      		 96,531 1,926,391 19.96×
      str
      		 514,884 10,282,060 19.97×
      wcs
      		 119,146 2,379,608 19.97×
      CryptographySingleMultiScaling
      crypt
      		 2,898 58,075 20.04×
      CompressionSingleMultiScaling
      jpeg
      		 520.69 10,366 19.91×
      zlib
      		 45.9 722.59 15.74×
      HypervisorSingleMultiScaling
      clock
      		 23,116 257,214 11.13×
      clone
      		 186.34 2,706 14.52×
      context
      		 2,611 52,195 19.99×
      cpu-sched
      		 63.82 1,287 20.16×
      eventfd
      		 51,706 599,669 11.60×
      fault
      		 13,595 178,450 13.13×
      fork
      		 3,682 42,344 11.50×
      getrandom
      		 48,954 977,275 19.96×
      hrtimers
      		 298,985 3,083,365 10.31×
      mprotect
      		 799,643 6,552,045 8.19×
      mtx
      		 5,651,847 825,655,670 146.09×
      munmap
      		 2,672 36,465 13.65×
      null
      		 468,331 9,361,037 19.99×
      pipe
      		 619,488 10,930,923 17.65×
      priv-instr
      		 756,159 11,385,290 15.06×
      resched
      		 5,290,870 3,814,307 0.72×
      switch
      		 76,128 6,873,867 90.29×
      timermix
      		 400,442 7,960,582 19.88×
      tlb-shootdown
      		 4,577 9,088 1.99×
      vdso
      		 46,821,314 934,290,409 19.95×
      vfork
      		 25,699 218,854 8.52×
      yield
      		 16,430,801 18,659,666 1.14×
      zero
      		 2,069,943 41,313,124 19.96×
      CPU float/SIMDSingleMultiScaling
      eigen
      		 1,336 26,683 19.97×
      expmath
      		 240.12 4,796 19.97×
      fma
      		 641,108 12,802,014 19.97×
      fp
      		 103.2 2,064 20.00×
      fractal
      		 6.6 128.81 19.52×
      hyperbolic
      		 244.28 4,879 19.97×
      logmath
      		 293.65 5,863 19.97×
      matrix
      		 5,344 106,787 19.98×
      matrix-3d
      		 505.06 6,040 11.96×
      monte-carlo
      		 193.68 3,872 19.99×
      mpfr
      		 641.98 12,817 19.97×
      powmath
      		 176.54 3,524 19.96×
      trig
      		 329.65 6,581 19.96×
      veccmp
      		 17,137 342,402 19.98×
      vecfp
      		 745.23 14,876 19.96×
      vecmath
      		 3,091 61,719 19.97×
      vecshuf
      		 671.71 13,418 19.98×
      vecwide
      		 18,391 367,598 19.99×
      AlgorithmsSingleMultiScaling
      bitonicsort
      		 5.4 107.99 20.00×
      bsearch
      		 258.37 5,159 19.97×
      bubblesort
      		 0.8 16 20.00×
      fibsearch
      		 273.56 5,464 19.97×
      heapsort
      		 8.2 163.99 20.00×
      hsearch
      		 4,922 98,103 19.93×
      insertionsort
      		 4.6 92 20.00×
      malloc
      		 2,655,545 29,876,535 11.25×
      mcontend
      		 167.67 40,788 243.27×
      mergesort
      		 27.8 547.96 19.71×
      qsort
      		 19.6 390.58 19.93×
      skiplist
      		 5,351 107,001 20.00×
      tree
      		 1.2 15.99 13.33×
      MemorySingleMultiScaling
      cache
      		 371,671 615,549 1.66×
      cachehammer
      		 150,195 514,804 3.43×
      cacheline
      		 46.17 925.81 20.05×
      l1cache
      		 182.38 3,633 19.92×
      memcpy
      		 443.8 8,864 19.97×
      memrate
      		 3.15 27.17 8.63×
      memthrash
      		 153.27 324.58 2.12×
      ptr-chase
      		 16,545,669 191,996,750 11.60×
      spinmem
      		 822.97 1,759 2.14×
      stream
      		 67.98 13,240 194.77×
      vm
      		 60,468 1,178,091 19.48×

      Stress-ng Raw Scores

      Stress-ng Relative Multicore Performance

      LLM Inference Speed for Prompt Processing

      LLM Inference Speed for Text Generation

      Static Web Server

      Redis

      Alternatives

      Servers of the Same Family

      INSTANCEvCPUsMEMORYAccelerators
      E2ps_v5 216 GiB0
      E4ps_v5 432 GiB0
      E8ps_v5 864 GiB0
      E16ps_v5 16128 GiB0
      E32ps_v5 32208 GiB0

      Similar Servers

      INSTANCEVENDORvCPUsMEMORYAccelerators

      E20ps_v5 FAQs