p3.16xlarge by Amazon Web Services

p3.16xlarge is a GPU accelerated Gen3 16xlarge server offered by Amazon Web Services with 64 vCPUs, 488 GiB of memory and 0 GB of storage. The last known price was 6.5858 USD per hour.
64 vCPU
488 GiB Memory
8 Accelerator
Spare SCore
29,166
(All-cores)
870
(Single-core)

Specifications

Server Metadata

Vendor ID
aws
Name
p3.16xlarge
Description
GPU accelerated Gen3 16xlarge
Family
p3
Hw Virt
-
Status
inactive
Observed At
2026-04-30T00:31:13.009488

Availability

REGION / IDSPOTONDEMAND
Northern Virgina (US) / us-east-1
11.5076 USD/h 24.48 USD/h
Ohio (US) / us-east-2
14.1719 USD/h 24.48 USD/h
Oregon (US) / us-west-2
13.117033 USD/h 24.48 USD/h
Dublin (IE) / eu-west-1
26.44 USD/h 26.44 USD/h
Quebec (CA) / ca-central-1
15.2551 USD/h 26.928 USD/h
London (GB) / eu-west-2
28.712 USD/h 28.712 USD/h
Frankfurt (DE) / eu-central-1
30.584 USD/h 30.584 USD/h
Tokyo (JP) / ap-northeast-1
33.552 USD/h 33.552 USD/h
Seoul (KR) / ap-northeast-2
33.872 USD/h 33.872 USD/h
Singapore (SG) / ap-southeast-1
33.872 USD/h 33.872 USD/h
Sydney (AU) / ap-southeast-2
33.872 USD/h 33.872 USD/h

Processor

vCPUs
64
Hypervisor
xen
CPU Allocation
Dedicated
CPU Cores
32
CPU Speed
2.7 GHz
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU Model
E5-2686 v4
CPU L1D Cache
32 KiB
CPU L1D Cache Total
1 MiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
1 MiB
CPU L2 Cache
256 KiB
CPU L2 Cache Total
8 MiB
CPU L3 Cache
45 MiB
CPU L3 Cache Total
90 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, pdpe1gb, rdtscp, lm, constant_tsc, arch_perfmon, rep_good, nopl, xtopology, nonstop_tsc, cpuid, aperfmperf, tsc_known_freq, pni, pclmulqdq, monitor, est, ssse3, fma, cx16, pcid, sse4_1, sse4_2, x2apic, movbe, popcnt, tsc_deadline_timer, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, abm, 3dnowprefetch, cpuid_fault, pti, fsgsbase, bmi1, hle, avx2, smep, bmi2, erms, invpcid, rtm, rdseed, adx, xsaveopt, ida

System Resources and Accelerators

MEMORY
Memory Amount
488 GiB
ACCELERATORS
Accelerator Count
8
Accelerator Memory Min
16 GiB
Accelerator Memory Total
128 GiB
Accelerator Manufacturer
NVIDIA
Accelerator Family
Volta
Accelerator Model
V100
Accelerators
  • NVIDIA Volta Tesla V100-SXM2-16GB (Memory amount: 16384, BIOS version: 88.00.4F.00.09, Clock rate: 1530)
  • NVIDIA Volta Tesla V100-SXM2-16GB (Memory amount: 16384, BIOS version: 88.00.4F.00.09, Clock rate: 1530)
  • NVIDIA Volta Tesla V100-SXM2-16GB (Memory amount: 16384, BIOS version: 88.00.4F.00.09, Clock rate: 1530)
  • NVIDIA Volta Tesla V100-SXM2-16GB (Memory amount: 16384, BIOS version: 88.00.4F.00.09, Clock rate: 1530)
  • NVIDIA Volta Tesla V100-SXM2-16GB (Memory amount: 16384, BIOS version: 88.00.4F.00.09, Clock rate: 1530)
  • NVIDIA Volta Tesla V100-SXM2-16GB (Memory amount: 16384, BIOS version: 88.00.4F.00.09, Clock rate: 1530)
  • NVIDIA Volta Tesla V100-SXM2-16GB (Memory amount: 16384, BIOS version: 88.00.4F.00.09, Clock rate: 1530)
  • NVIDIA Volta Tesla V100-SXM2-16GB (Memory amount: 16384, BIOS version: 88.00.4F.00.09, Clock rate: 1530)
STORAGE
Storage Size
0 GB
Storages
    NETWORK
    Network Speed Baseline
    5 Gbps
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    0 GB/month
    IPv4
    0

    CPU and System Topology

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    1.6Score
    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) 3,787,099 ops/sec 1,917,050 ops/sec 2 50.00% 50.00% +41.00%
    Redis RPS (pipeline=16, SET) 23,821,531 ops/sec 11,661,818 ops/sec 2 20.00% 20.00% +15.00%
    PassMark Memory Mark (composite) 2,181 2,417 0.92 10.00% 10.00% -0.83%
    Memory bandwidth (read, 16 MB ~ L3) 115,499 MB/sec 108,361 MB/sec 1.1 10.00% 10.00% +0.96%
    PassMark single-thread CPU 1,580 Mops/s 2,365 Mops/s 0.67 10.00% 10.00% -3.90%

    GPU Bandwidth and Latency

    Node (all GPUs)
    Node baseline directional bandwidth
    		28.5 GB/s
    Node GPU-to-host bandwidth
    		47.8 GB/s
    Node host-to-GPU bandwidth
    		44.6 GB/s
    Peer-to-peer
    GPU-to-GPU P2P duplex bandwidth
    		36.3 GB/s
    GPU-to-GPU P2P gather bandwidth
    		92.9 GB/s
    GPU-to-GPU P2P latency
    		1,185 ns
    GPU-to-GPU P2P bandwidth
    		36.3 GB/s
    Single GPU slot
    Per-GPU GPU-to-host bandwidth
    		12.4 GB/s
    Per-GPU host-to-GPU bandwidth
    		11.0 GB/s
    Per-GPU host-device latency
    		1,177 ns
    Benchmark Environment
    Driver Version
    		 535.183.01
    CUDA Runtime Version
    		 12.6
    nvbandwidth Version
    		 v0.5
    GPU Count
    		 8
    P2P Supported
    Yes
    Efficiency
    SM/CE copy efficiency
    		1.00

    Memory Bandwidth

    Compression

    OpenSSL

    Geekbench Single-Core

    Score: 884

    Geekbench Multi-Core

    Score: 8,719

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 33,705
    Compression
    		 642,085
    Encryption
    		 14,386
    Extended Instructions
    		 33,230
    Floating Point Maths
    		 93,493
    Integer Maths
    		 146,913
    Physics
    		 2,796
    Prime Numbers
    		 214
    Single Threaded
    		 1,580
    String Sorting
    		 93,756

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 2,181
    Database Operations
    		 18,580
    Memory Latency
    		 56
    Memory Read Cached
    		 18,735
    Memory Read Uncached
    		 7,746
    Memory Write
    		 6,768

    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

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYAccelerators

    p3.16xlarge FAQs