NC24ads_A100_v4 by Microsoft Azure

NC24ads_A100_v4 is a NCADSA100v4 family [AMD processor] [Local Disk] [Premium Storage capable] 24 vCPUs server offered by Microsoft Azure with 24 vCPUs, 220 GiB of memory and 1.03 TB of storage. The pricing starts at 0.6788 USD per hour.
24 vCPU
220 GiB Memory
1.03 TB Storage
1 GPU
Spare SCore
61,789
(All-cores)
2,780
(Single-core)

Specifications

Server Metadata

Vendor ID
azure
Server ID
Standard_NC24ads_A100_v4
Name
NC24ads_A100_v4
API Reference
Standard_NC24ads_A100_v4
Description
NCADSA100v4 family [AMD processor] [Local Disk] [Premium Storage capable] 24 vCPUs
Family
NCADSA100v4
Hw Virt
Average Time To Start
32
Status
active
Observed At
2026-10-26T03:28:28.502231

Availability

REGION / IDSPOTONDEMAND
East US (US) / eastus
0.6788 USD/h 3.673 USD/h
East US 2 (US) / eastus2
0.6788 USD/h 3.673 USD/h
West US 2 (US) / westus2
0.6788 USD/h 3.673 USD/h
West US 3 (US) / westus3
0.6788 USD/h 3.673 USD/h
Central US (US) / centralus
0.7669 USD/h 4.15 USD/h
Canada Central (CA) / canadacentral
0.8146 USD/h 4.408 USD/h
North Central US (US) / northcentralus
0.8146 USD/h 4.408 USD/h
North Europe (IE) / northeurope
0.8146 USD/h 4.408 USD/h
South Central US (US) / southcentralus
0.8146 USD/h 4.408 USD/h
France Central (FR) / francecentral
0.8484 USD/h 4.591 USD/h
UK South (GB) / uksouth
0.8484 USD/h 4.591 USD/h
Germany West Central (DE) / germanywestcentral
0.8824 USD/h 4.775 USD/h
Italy North (IT) / italynorth
0.8824 USD/h 4.775 USD/h
Poland Central (PL) / polandcentral
0.8824 USD/h 4.775 USD/h
Southeast Asia (SG) / southeastasia
0.8824 USD/h 4.775 USD/h
Spain Central (ES) / spaincentral
0.8824 USD/h 4.775 USD/h
West Europe (NL) / westeurope
0.8824 USD/h 4.775 USD/h
West US (US) / westus
0.8824 USD/h 4.775 USD/h
Sweden Central (SE) / swedencentral
0.9373 USD/h 4.775 USD/h
Korea Central (KR) / koreacentral
0.9164 USD/h 4.959 USD/h
Central India (IN) / centralindia
0.9502 USD/h 5.142 USD/h
Jio India Central (IN) / jioindiacentral
- 5.142 USD/h
Jio India West (IN) / jioindiawest
- 5.142 USD/h
Switzerland North (CH) / switzerlandnorth
0.9706 USD/h 5.252 USD/h
Australia East (AU) / australiaeast
0.9842 USD/h 5.326 USD/h
Japan East (JP) / japaneast
0.9842 USD/h 5.326 USD/h
Brazil South (BR) / brazilsouth
1.3575 USD/h 7.346 USD/h

Processor

vCPUs
24
Hypervisor
Microsoft Hyper-V
CPU Allocation
Dedicated
CPU Cores
24
CPU Speed
3.7 GHz
CPU Architecture
x86_64
CPU Manufacturer
AMD
CPU Family
EPYC
CPU Model
7V13
CPU L1D Cache
32 KiB
CPU L1D Cache Total
768 KiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
768 KiB
CPU L2 Cache
512 KiB
CPU L2 Cache Total
12 MiB
CPU L3 Cache
32 MiB
CPU L3 Cache Total
96 MiB
CPU Flags
fpu, vme, de, pse, tsc, msr, pae, mce, cx8, apic, sep, mtrr, pge, mca, cmov, pat, pse36, clflush, mmx, fxsr, sse, sse2, ht, syscall, nx, mmxext, fxsr_opt, pdpe1gb, rdtscp, lm, constant_tsc, rep_good, nopl, tsc_reliable, nonstop_tsc, cpuid, extd_apicid, aperfmperf, pni, pclmulqdq, ssse3, fma, cx16, pcid, sse4_1, sse4_2, movbe, popcnt, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, cmp_legacy, cr8_legacy, abm, sse4a, misalignsse, 3dnowprefetch, osvw, topoext, perfctr_core, vmmcall, fsgsbase, bmi1, avx2, smep, bmi2, erms, invpcid, rdseed, adx, smap, clflushopt, clwb, sha_ni, xsaveopt, xsavec, xgetbv1, xsaves, user_shstk, clzero, xsaveerptr, rdpru, arat, umip, vaes, vpclmulqdq, rdpid, fsrm
Ecpus
22.2
Scalability
92.5

System Resources and Accelerators

MEMORY
Memory Amount
220 GiB
Memory Amount Actual
220 GiB
ACCELERATORS
Accelerator Count
1
Accelerator Memory Min
80 GiB
Accelerator Memory Total
80 GiB
Accelerator Type
GPU
Accelerator Manufacturer
NVIDIA
Accelerator Family
Ampere
Accelerator Model
A100
Accelerators
  • NVIDIA Ampere NVIDIA A100 80GB PCIe (Memory amount: 81920, Firmware version: 595.71.05, BIOS version: 92.00.68.00.01, Clock rate: 1410)
STORAGE
Storage Size
1026 GB
Storage Type
nvme ssd
Storages
  • 66 GB ssd (temp disk)
  • 960 GB nvme ssd
NETWORK
Inbound Traffic
0 GB/month
Outbound Traffic
0 GB/month
IPv4
0

CPU and System Topology

Server Description

A high-density GPU server combining dedicated AMD EPYC processors, local NVMe storage, and an NVIDIA A100 GPU for advanced AI workloads.

GPU AcceleratedMemory OptimizedStorage & Database

Microsoft Azure NC24ads_A100_v4 is a GPU-accelerated virtual machine from the NCADSA100v4 family, running on the Microsoft Hyper-V hypervisor. It features 24 dedicated AMD EPYC 7V13 physical cores clocked at 3.7 GHz, 220 GB of system memory, and a 1026 GB local NVMe SSD. Graphics and parallel computing acceleration are provided by a single NVIDIA A100 Ampere GPU with 80 GB of VRAM. Benchmark data highlights top-tier performance in LLM inference across small, medium, and large models, alongside top-tier memory performance. The server delivers strong single-core CPU execution and multi-threaded decompression. This hardware profile is highly suited for large language model inference, deep learning, and memory-intensive GPU-accelerated workloads.

Economics

Average Price per Region

Prices per Zone

Lowest Prices

Performance

Workload Profiles

1.3Score
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) 2,646,569 ops/sec 1,919,289 ops/sec 1.4 50.00% 50.00% +18.00%
Redis RPS (pipeline=16, SET) 22,625,817 ops/sec 11,728,058 ops/sec 1.9 20.00% 20.00% +14.00%
PassMark Memory Mark (composite) 3,050 2,416 1.2 10.00% 10.00% +1.80%
Memory bandwidth (read, 16 MB ~ L3) 98,313 MB/sec 108,466 MB/sec 0.89 10.00% 10.00% -1.20%
PassMark single-thread CPU 2,276 Mops/s 2,365 Mops/s 0.96 10.00% 10.00% -0.41%

GPU Bandwidth and Latency

Node (all GPUs)
Node baseline directional bandwidth
19.6 GB/s
Node GPU-to-host bandwidth
26.3 GB/s
Node host-to-GPU bandwidth
26.7 GB/s
Single GPU slot
Per-GPU GPU-to-host bandwidth
26.3 GB/s
Per-GPU host-to-GPU bandwidth
26.7 GB/s
Per-GPU host-device latency
812 ns
Benchmark Environment
Driver Version
 535.183.01
CUDA Runtime Version
 12.6
nvbandwidth Version
 v0.5
GPU Count
 1
P2P Supported
No
Efficiency
SM/CE copy efficiency
1.00

Memory Bandwidth

Compression

OpenSSL

Geekbench Single-Core

Score: 1,670

Geekbench Multi-Core

Score: 14,903

Passmark CPU Scores

BENCHMARKSCORE
Mark
 39,845
Compression
 507,919
Encryption
 25,712
Extended Instructions
 41,434
Floating Point Maths
 96,692
Integer Maths
 119,374
Physics
 5,068
Prime Numbers
 382
Single Threaded
 2,276
String Sorting
 69,939

Passmark Memory Scores

BENCHMARKSCORE
Memory Mark
 3,050
Database Operations
 18,167
Memory Latency
 56
Memory Read Cached
 23,325
Memory Read Uncached
 20,086
Memory Write
 19,758

Stress-ng Stressors

CPU integerSingleMultiScaling
atomic
 358.5 191.81 0.54×
bitops
 146,860 3,203,106 21.81×
branch
 170,501 3,704,003 21.72×
funccall
 518.62 12,052 23.24×
funcret
 16,559 385,788 23.30×
hash
 139,227 3,066,042 22.02×
intmath
 59,468 1,312,559 22.07×
nop
 289,948 6,424,494 22.16×
prime
 404,859 8,991,866 22.21×
rotate
 79,104 1,740,118 22.00×
str
 900,512 18,767,555 20.84×
wcs
 596,278 13,063,984 21.91×
CryptographySingleMultiScaling
crypt
 3,436 72,242 21.03×
CompressionSingleMultiScaling
jpeg
 764.9 16,050 20.98×
zlib
 65.49 1,235 18.86×
HypervisorSingleMultiScaling
clock
 3,787 68,205 18.01×
clone
 1,003 12,569 12.53×
context
 3,598 77,623 21.57×
cpu-sched
 61.37 1,562 25.46×
eventfd
 29,078 1,465,075 50.38×
fault
 7,375 110,651 15.00×
fork
 3,498 41,920 11.99×
getrandom
 47,514 1,029,588 21.67×
hrtimers
 68,870 1,283,104 18.63×
mprotect
 710,164 5,140,041 7.24×
mtx
 49,615,343 2,530,836,690 51.01×
munmap
 2,478 33,810 13.64×
null
 450,371 9,922,849 22.03×
pipe
 1,651,529 12,260,936 7.42×
priv-instr
 406,737 8,408,052 20.67×
resched
 6,035,027 2,078,393 0.34×
switch
 58,795 4,197,072 71.38×
timermix
 42,332 1,034,888 24.45×
tlb-shootdown
 14,130 93,108 6.59×
tsc
 706,600 15,094,883 21.36×
usersyscall
 321,145 6,658,882 20.73×
vdso
 81,095,819 1,766,784,617 21.79×
vfork
 20,060 214,732 10.70×
x86syscall
 5,678,000 124,210,917 21.88×
yield
 15,197,558 18,070,747 1.19×
zero
 2,955,082 64,508,497 21.83×
CPU float/SIMDSingleMultiScaling
eigen
 14,763 314,889 21.33×
expmath
 2,626 56,794 21.63×
fma
 760,492 17,665,729 23.23×
fp
 223.2 5,173 23.18×
fractal
 6.6 149.53 22.66×
hyperbolic
 3,442 74,076 21.52×
logmath
 4,060 87,980 21.67×
matrix
 2,804 61,043 21.77×
matrix-3d
 91.98 1,902 20.67×
monte-carlo
 214.28 4,636 21.64×
mpfr
 1,430 31,722 22.18×
powmath
 1,221 26,856 21.99×
trig
 1,549 33,277 21.49×
veccmp
 16,688 361,152 21.64×
vecfp
 1,526 33,496 21.96×
vecmath
 3,531 77,281 21.89×
vecshuf
 553.71 11,833 21.37×
vecwide
 24,224 535,136 22.09×
AlgorithmsSingleMultiScaling
bitonicsort
 9 195.99 21.78×
bsearch
 273.46 5,983 21.88×
bubblesort
 0.8 19.2 24.00×
fibsearch
 323.79 7,190 22.21×
heapsort
 9.6 210.19 21.89×
hsearch
 8,037 172,551 21.47×
insertionsort
 12 248 20.67×
malloc
 2,799,881 49,568,271 17.70×
mcontend
 108.67 3,720 34.23×
mergesort
 28.2 617.98 21.91×
qsort
 18.6 407.98 21.93×
skiplist
 6,424 138,621 21.58×
tree
 2 28.8 14.40×
MemorySingleMultiScaling
cache
 659,943 1,260,752 1.91×
cachehammer
 298,979 5,861,757 19.61×
cacheline
 23.98 707.24 29.49×
l1cache
 167.24 3,418 20.44×
memcpy
 387.08 8,108 20.95×
memrate
 3.15 9.32 2.96×
memthrash
 169.93 734.15 4.32×
ptr-chase
 63,929,436 246,053,614 3.85×
spinmem
 251.99 1,159 4.60×
stream
 115.43 7,199 62.37×
vm
 169,186 294,977 1.74×

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
NC48ads_A100_v4 48440 GiB2
NC96ads_A100_v4 96880 GiB4

Similar Servers

INSTANCEVENDORvCPUsMEMORYAccelerators

NC24ads_A100_v4 FAQs