g4ad.4xlarge by Amazon Web Services

g4ad.4xlarge is a Graphics intensive [AMD processors] [Instance store volumes] Gen4 4xlarge server offered by Amazon Web Services with 16 vCPUs, 64 GiB of memory and 600 GB of storage. The pricing starts at 0.2969 USD per hour.
16 vCPU
64 GiB Memory
600 GB Storage
1 GPU
Spare SCore
6,134
(All-cores)
765
(Single-core)

Specifications

Server Metadata

Vendor ID
aws
Name
g4ad.4xlarge
Description
Graphics intensive [AMD processors] [Instance store volumes] Gen4 4xlarge
Family
g4ad
Hw Virt
Status
active
Observed At
2026-11-08T06:56:11.662873

Availability

REGION / IDSPOTONDEMAND
Quebec (CA) / ca-central-1
0.3576 USD/h -
Northern Virgina (US) / us-east-1
0.407875 USD/h -
Oregon (US) / us-west-2
0.396167 USD/h -
London (GB) / eu-west-2
0.38165 USD/h -
Ohio (US) / us-east-2
0.475233 USD/h -
Dublin (IE) / eu-west-1
0.494367 USD/h -
Frankfurt (DE) / eu-central-1
0.6007 USD/h -
Tokyo (JP) / ap-northeast-1
0.860933 USD/h -

Processor

vCPUs
16
Hypervisor
nitro
CPU Allocation
Dedicated
CPU Cores
8
CPU Speed
3 GHz
CPU Architecture
x86_64
CPU Manufacturer
AMD
CPU Family
EPYC
CPU Model
7R32
CPU L1D Cache
32 KiB
CPU L1D Cache Total
256 KiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
256 KiB
CPU L2 Cache
512 KiB
CPU L2 Cache Total
4 MiB
CPU L3 Cache
16 MiB
CPU L3 Cache Total
32 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, nonstop_tsc, cpuid, extd_apicid, aperfmperf, tsc_known_freq, pni, pclmulqdq, ssse3, fma, cx16, sse4_1, sse4_2, movbe, popcnt, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, cmp_legacy, cr8_legacy, abm, sse4a, misalignsse, 3dnowprefetch, topoext, ssbd, ibrs, ibpb, stibp, vmmcall, fsgsbase, bmi1, avx2, smep, bmi2, rdseed, adx, smap, clflushopt, clwb, sha_ni, xsaveopt, xsavec, xgetbv1, clzero, xsaveerptr, rdpru, wbnoinvd, arat, npt, nrip_save, rdpid
Ecpus
8
Scalability
100

System Resources and Accelerators

MEMORY
Memory Amount
64 GiB
Memory Amount Actual
64 GiB
Memory Generation
DDR4
Memory Speed
2933 Mhz
ACCELERATORS
Accelerator Count
1
Accelerator Memory Min
8 GiB
Accelerator Memory Total
8 GiB
Accelerator Type
GPU
Accelerator Manufacturer
AMD
Accelerator Family
Radeon Pro Navi
Accelerator Model
V520
Accelerators
  • AMD Radeon Pro V520 (Memory amount: 8192)
STORAGE
Storage Size
600 GB
Storage Type
nvme ssd
Storages
  • 600 GB nvme ssd
NETWORK
Network Speed Baseline
8.333 Gbps
Network Speed Max
10 Gbps
Network Storage Speed Baseline
1.58 Gbps
Network Storage Speed Max
3.17 Gbps
Inbound Traffic
0 GB/month
Outbound Traffic
0 GB/month
IPv4
0

CPU and System Topology

Server Description

An AMD-powered instance featuring dedicated Radeon Pro graphics and local NVMe storage for graphics-intensive workloads and media processing.

GPU AcceleratedStorage & DatabaseGeneral Purpose

Amazon Web Services g4ad.4xlarge is a graphics-optimized instance featuring an AMD EPYC 7R32 processor with 16 vCPUs, 64 GB of DDR4 memory, and a dedicated AMD Radeon Pro Navi V520 GPU with 8 GB of VRAM. Built on the Nitro hypervisor, it includes 600 GB of local NVMe SSD storage and an 8 Gbps baseline network bandwidth. Performance benchmarks indicate average multi-core CPU and cached memory bandwidth capabilities, though single-core performance and uncached large-block memory bandwidth are relatively weak. The instance is well-suited for graphics-intensive applications, virtual workstations, game streaming, and moderate media processing workloads that benefit from AMD GPU acceleration and local high-speed storage.

Economics

Average Price per Region

Prices per Zone

Lowest Prices

Performance

Workload Profiles

0.69Score
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) 1,302,162 ops/sec 1,922,940 ops/sec 0.68 50.00% 50.00% -18.00%
Redis RPS (pipeline=16, SET) 8,643,904 ops/sec 11,797,257 ops/sec 0.72 20.00% 20.00% -6.40%
PassMark Memory Mark (composite) 2,687 2,406 1.1 10.00% 10.00% +0.96%
Memory bandwidth (read, 16 MB ~ L3) 38,297 MB/sec 108,657 MB/sec 0.35 10.00% 10.00% -10.00%
PassMark single-thread CPU 2,135 Mops/s 2,365 Mops/s 0.88 10.00% 10.00% -1.30%

Memory Bandwidth

Compression

OpenSSL

Geekbench Single-Core

Score: 1,339

Geekbench Multi-Core

Score: 8,094

Passmark CPU Scores

BENCHMARKSCORE
Mark
 19,577
Compression
 239,632
Encryption
 17,000
Extended Instructions
 14,773
Floating Point Maths
 32,117
Integer Maths
 53,861
Physics
 1,802
Prime Numbers
 85
Single Threaded
 2,135
String Sorting
 32,991

Passmark Memory Scores

BENCHMARKSCORE
Memory Mark
 2,687
Database Operations
 5,626
Memory Latency
 50
Memory Read Cached
 24,692
Memory Read Uncached
 12,886
Memory Write
 13,572

Stress-ng Stressors

CPU integerSingleMultiScaling
atomic
 282.27 350.57 1.24×
bitops
 120,735 1,216,640 10.08×
branch
 152,868 2,125,983 13.91×
funccall
 502.71 5,417 10.78×
funcret
 14,101 165,535 11.74×
hash
 131,824 1,638,898 12.43×
intmath
 19,450 169,940 8.74×
nop
 236,440 1,910,069 8.08×
prime
 346,849 3,658,887 10.55×
rotate
 59,939 496,818 8.29×
str
 628,054 6,267,147 9.98×
wcs
 469,501 4,290,975 9.14×
CryptographySingleMultiScaling
crypt
 2,836 24,230 8.54×
CompressionSingleMultiScaling
jpeg
 668.81 6,921 10.35×
zlib
 52.13 702.51 13.48×
HypervisorSingleMultiScaling
clock
 19,729 166,572 8.44×
clone
 167.78 1,329 7.92×
context
 3,394 45,996 13.55×
cpu-sched
 74.96 1,077 14.37×
eventfd
 21,706 802,423 36.97×
fault
 8,312 85,933 10.34×
fork
 3,103 28,031 9.03×
getrandom
 36,458 390,121 10.70×
hrtimers
 392,804 1,238,167 3.15×
mprotect
 678,935 2,774,650 4.09×
mtx
 20,343,528 550,214,076 27.05×
munmap
 2,172 21,041 9.69×
null
 416,795 5,378,248 12.90×
pipe
 1,746,529 6,563,162 3.76×
priv-instr
 425,428 5,474,752 12.87×
resched
 3,186,690 925,476 0.29×
switch
 32,371 1,979,493 61.15×
timermix
 133,834 1,853,234 13.85×
tlb-shootdown
 9,273 35,590 3.84×
tsc
 653,337 10,143,523 15.53×
usersyscall
 292,125 3,809,109 13.04×
vdso
 79,377,800 904,975,120 11.40×
vfork
 16,236 121,397 7.48×
x86syscall
 5,283,005 68,073,285 12.89×
yield
 7,753,470 7,853,749 1.01×
zero
 2,125,701 26,634,696 12.53×
CPU float/SIMDSingleMultiScaling
eigen
 12,414 117,345 9.45×
expmath
 2,781 34,962 12.57×
fma
 740,762 9,335,662 12.60×
fp
 208.6 2,506 12.02×
fractal
 6.4 80.23 12.54×
hyperbolic
 3,642 43,704 12.00×
logmath
 4,103 47,947 11.69×
matrix
 2,755 39,634 14.39×
matrix-3d
 158.57 1,475 9.30×
monte-carlo
 183.48 2,406 13.11×
mpfr
 1,276 13,222 10.36×
powmath
 1,343 17,284 12.87×
trig
 1,563 19,037 12.18×
veccmp
 17,124 165,859 9.69×
vecfp
 1,389 17,098 12.31×
vecmath
 3,368 43,455 12.90×
vecshuf
 484.3 4,789 9.89×
vecwide
 26,550 252,587 9.51×
AlgorithmsSingleMultiScaling
bitonicsort
 6.8 68.8 10.12×
bsearch
 226.21 2,312 10.22×
bubblesort
 0.8 6.4 8.00×
fibsearch
 259.11 2,698 10.41×
heapsort
 8.6 97.8 11.37×
hsearch
 5,355 51,422 9.60×
insertionsort
 6.6 62.2 9.42×
malloc
 2,631,998 27,781,558 10.56×
mcontend
 126.75 2,603 20.54×
mergesort
 25.8 314.2 12.18×
qsort
 16.6 200.79 12.10×
skiplist
 5,519 70,349 12.75×
tree
 1.8 9.6 5.33×
MemorySingleMultiScaling
cache
 463,038 1,079,846 2.33×
cachehammer
 289,809 3,950,295 13.63×
cacheline
 49.36 844.01 17.10×
l1cache
 198.12 1,540 7.77×
memcpy
 354.81 2,960 8.34×
memrate
 1.74 6.02 3.46×
memthrash
 112.86 375.89 3.33×
ptr-chase
 29,649,147 134,045,086 4.52×
spinmem
 232.19 830.64 3.58×
stream
 173.89 5,401 31.06×
vm
 55,424 292,976 5.29×

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
g4ad.xlarge 416 GiB1
g4ad.2xlarge 832 GiB1
g4ad.8xlarge 32128 GiB2
g4ad.16xlarge 64256 GiB4

Similar Servers

INSTANCEVENDORvCPUsMEMORYAccelerators

g4ad.4xlarge FAQs