Best multi-core performance servers with 16 vCPUs

This is a manually curated list of 16 vCPU servers with the best multi-core performance as per stress-ng's div16 CPU burning method. Note that servers using the same CPU model at the same vendor were deduplicated, and only the most general options were kept with similar memory amounts(e.g. from AWS's r6a.4xlarge, m6a.4xlarge, and c6a.4xlarge showing only m6a.4xlarge with 64 GiB of memory; similarly GCP's c2d-highmem-16, c2d-standard-16, and c2d-highcpu-16 showing only c2d-standard-16 with 64 GiB of memory).

	M7A.4XLARGE 💪	T2D-STANDARD-16 🥇	D16PS_V6	DC16S_V3	CAX41 💰	T2A-STANDARD-16	M8G.4XLARGE	CLOUDNATIVE-16XCPU-64GB
Vendor
$Core	324,355/USD Best (ondemand): 53,057/USD Best (spot): 324,355/USD	601,127/USD Best (ondemand): 96,617/USD Best (spot): 601,127/USD	484,712/USD Best (ondemand): 89,606/USD Best (spot): 484,712/USD	183,233/USD Best (ondemand): 33,865/USD Best (spot): 183,233/USD	492,688/USD Best (ondemand): 492,688/USD Best (spot): -	170,190/USD Best (ondemand): 47,051/USD Best (spot): 170,190/USD	303,936/USD Best (ondemand): 53,173/USD Best (spot): 303,936/USD	24,682/USD Best (ondemand): 24,682/USD Best (spot): -
SCore All	49205	43101	33154	30343	29118	28983	27293	11615
SCore Single	3073	2695	2073	1898	1822	1813	1707	786
Best spot price	0.1517 USD eu-south-2a	0.0717 USD us-south1-c	0.0684 USD centralindia-1	0.1656 USD italynorth-1	- -	0.1703 USD us-central1-a	0.0898 USD ap-southeast-3c	- -
Best ondemand price	0.9274 USD us-east-1d	0.4461 USD asia-south1-b	0.37 USD centralindia-1	0.896 USD italynorth-1	0.0591 USD nbg1-dc3	0.616 USD us-central1-a	0.5133 USD ap-south-1a	0.4706 USD Sydney #1
CPU
vCPUs	16	16	16	16	16	16	16	16
Hypervisor	nitro		Microsoft Hyper-V	Microsoft Hyper-V	QEMU		nitro	KVM
CPU Allocation	Dedicated	Dedicated	Dedicated	Dedicated	Shared	Dedicated	Dedicated	Shared
CPU Cores	16	16	16	16	16	16	16	16
CPU Speed	3.7 GHz		3.4 GHz	2.8 GHz	2 GHz	3 GHz	2.8 GHz	2 GHz
CPU Architecture	x86_64	x86_64	arm64	x86_64	arm64	arm64	arm64	x86_64
CPU Manufacturer	AMD	AMD	Microsoft	Intel	Ampere		AWS	AMD
CPU Family	EPYC	EPYC	ARM	Xeon	Ampere Altra	ARMv8		EPYC
CPU Model	9R14	7B13	Neoverse-N2	8370C	Neoverse-N1	Neoverse-N1	Neoverse-V2	7542
CPU L1D Cache	32 KiB	32 KiB	64 KiB	48 KiB		64 KiB	64 KiB	64 KiB
CPU L1D Cache Total	512 KiB	512 KiB	1024 KiB	768 KiB		1024 KiB	1024 KiB	1024 KiB
CPU L1I Cache	32 KiB	32 KiB	64 KiB	32 KiB		64 KiB	64 KiB	64 KiB
CPU L1I Cache Total	512 KiB	512 KiB	1024 KiB	512 KiB		1024 KiB	1024 KiB	1024 KiB
CPU L2 Cache	1024 KiB	512 KiB	1024 KiB	1280 KiB		1024 KiB	2048 KiB	512 KiB
CPU L2 Cache Total	16384 KiB	8192 KiB	16384 KiB	20480 KiB		16384 KiB	32768 KiB	8192 KiB
CPU L3 Cache	32768 KiB	32768 KiB	131072 KiB	49152 KiB		32768 KiB	36864 KiB	16384 KiB
CPU L3 Cache Total	65536 KiB	65536 KiB	131072 KiB	49152 KiB		32768 KiB	36864 KiB	16384 KiB
Memory
Memory Amount	64 GiB	64 GiB	64 GiB	128 GiB	32 GiB	64 GiB	64 GiB	64 GiB
Memory Generation	DDR5
Memory Speed	4800 Mhz
GPU
GPU Count	0	0	0	0	0	0	0	0
Storage
Storage Size	-	-	-	-	320 GB	-	-	-
Storage Type					ssd
Network
Network Speed	6.25 Gbps						7.5 Gbps
Inbound Traffic	0 GB/month	0 GB/month	0 GB/month	0 GB/month	0 GB/month	0 GB/month	0 GB/month	0 GB/month
Outbound Traffic	0 GB/month	0 GB/month	0 GB/month	0 GB/month	20480 GB/month	0 GB/month	0 GB/month	20480 GB/month
IPv4	0	0	0	0	0	0	0	0
Memory Bandwidth
bw_mem: Read bw_mem: Write bw_mem: Read/Write sc-membench: Read sc-membench: Write sc-membench: Copy (Read/Write) sc-membench: Latency
Compression

OpenSSL
AES-256-CBC ARIA-256-CBC CAMELLIA-256-CBC SM4-CBC blake2b512 sha256 sha3-256 sha3-512 sha512 shake128 shake256
Geekbench

PassMark (CPU)

String Sorting	59.77K	49.38K	82.23K	49.63K	53.75K	54.22K	74.10K	39.80K
Single Threaded	2.89K	2.38K	1.88K	2.44K	1.32K	1.32K	1.94K	2.18K
Prime Numbers	302.35	246.19	225.03	252.81	99.28	118.19	259.95	176.88
Physics	4.15K	3.88K	5.19K	3.27K	1.79K	1.94K	4.87K	2.44K
Mark	36.51K	30.45K	14.76K	30.81K	11.51K	11.60K	14.60K	25.24K
Integer Maths	96.62K	83.27K	100.91K	84.28K	75.22K	75.42K	83.46K	71.93K
Floating Point Maths	85.20K	67.28K	84.76K	69.44K	57.18K	57.26K	96.97K	50.04K
Extended Instructions	39.42K	31.10K	14.31K	29.82K	11.92K	11.90K	14.74K	21.51K
Encryption	20.85K	17.97K	3.01K	18.78K	2.59K	2.59K	3.03K	18.16K
Compression	420.87K	352.95K	109.87K	362.22K	90.07K	90.39K	106.42K	292.20K
PassMark (Memory)

Memory Write	25.79K	17.03K	25.73K	10.62K	14.69K	14.85K	21.21K	9.69K
Memory Read Uncached	27.06K	16.70K	19.22K	11.46K	8.54K	8.93K	19.93K	11.01K
Memory Read Cached	27.84K	24.31K	25.61K	24.48K	11.28K	11.29K	20.52K	24.64K
Memory Mark	2.67K	2.72K	3.24K	2.59K	1.83K	1.93K	3.21K	1.68K
Memory Latency	79.09	64.26	54.58	59.49	75.06	70.04	48.73	119.83
Database Operations	14.91K	13.22K	19.67K	14.83K	10.85K	10.12K	24.10K	8.50K
Stress-ng div16 Raw Scores per vCPU

Stress-ng Relative Multicore Performance per vCPU

LLM Inference Speed

Static Web Server

Redis

Further Benchmarks
Workload profile: Web server
Workload profile: Compute heavy
Workload profile: CI/CD build
Workload profile: Cache intensive
stress-ng CPU all
stress-ng div16 multi-core
stress-ng div16 single-core
BogoMips
View Server

Best multi-core performance servers with 16 vCPUs

CPU

Memory

GPU

Storage

Network

Memory Bandwidth

Compression

OpenSSL

Geekbench

PassMark (CPU)

PassMark (Memory)

Stress-ng div16 Raw Scores per vCPU

Stress-ng Relative Multicore Performance per vCPU

LLM Inference Speed

Static Web Server

Redis

Further Benchmarks

Embed Chart