b3-640 by OVHcloud

b3-640 is a General Purpose (160 vCPUs, 640 GiB RAM, 400 GB NVMe) server offered by OVHcloud with 160 vCPUs, 640 GiB of memory and 400 GB of storage. The pricing starts at 4.5716 USD per hour.
160 vCPU
640 GiB Memory
400 GB Storage
Spare SCore
288,171
(All-cores)
3,078
(Single-core)

Specifications

Server Metadata

Vendor ID
ovh
Name
b3-640
Description
General Purpose (160 vCPUs, 640 GiB RAM, 400 GB NVMe)
Family
General Purpose
Hw Virt
Average Time To Start
64
Status
active
Observed At
2026-11-01T09:28:26.233090

Availability

REGION / IDSPOTONDEMAND
AP-SOUTHEAST-SYD-2 (AU) / AP-SOUTHEAST-SYD-2
- 4.5716 USD/h
AP-SOUTH-MUM-1 (IN) / AP-SOUTH-MUM-1
- 4.5716 USD/h
BHS5 (CA) / BHS5
- 4.5716 USD/h
DE1 (DE) / DE1
- 4.5716 USD/h
EU-SOUTH-MIL (IT) / EU-SOUTH-MIL
- 4.5716 USD/h
EU-WEST-PAR (FR) / EU-WEST-PAR
- 4.5716 USD/h
GRA11 (FR) / GRA11
- 4.5716 USD/h
GRA7 (FR) / GRA7
- 4.5716 USD/h
GRA9 (FR) / GRA9
- 4.5716 USD/h
RBX-A (FR) / RBX-A
- 4.5716 USD/h
SBG5 (FR) / SBG5
- 4.5716 USD/h
SBG7 (FR) / SBG7
- 4.5716 USD/h
SGP1 (SG) / SGP1
- 4.5716 USD/h
SYD1 (AU) / SYD1
- 4.5716 USD/h
UK1 (GB) / UK1
- 4.5716 USD/h
WAW1 (PL) / WAW1
- 4.5716 USD/h

Processor

vCPUs
160
Hypervisor
KVM
CPU Allocation
Dedicated
CPU Cores
160
CPU Speed
2.3 GHz
CPU Architecture
x86_64
CPU Manufacturer
AMD
CPU Family
EPYC
CPU L1D Cache
32 KiB
CPU L1D Cache Total
5 MiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
5 MiB
CPU L2 Cache
512 KiB
CPU L2 Cache Total
80 MiB
CPU L3 Cache
32 MiB
CPU L3 Cache Total
5 GiB
CPU Flags
fpu, vme, de, pse, tsc, msr, pae, mce, cx8, apic, sep, mtrr, pge, mca, cmov, pat, pse36, clflush, mmx, fxsr, sse, sse2, syscall, nx, mmxext, fxsr_opt, pdpe1gb, rdtscp, lm, rep_good, nopl, cpuid, extd_apicid, tsc_known_freq, pni, pclmulqdq, ssse3, fma, cx16, pcid, sse4_1, sse4_2, x2apic, movbe, popcnt, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, svm, cr8_legacy, abm, sse4a, misalignsse, 3dnowprefetch, osvw, topoext, perfctr_core, ssbd, ibrs, ibpb, stibp, vmmcall, fsgsbase, bmi1, avx2, smep, bmi2, erms, invpcid, rdseed, adx, smap, clflushopt, clwb, sha_ni, xsaveopt, xsavec, xgetbv1, xsaves, clzero, xsaveerptr, wbnoinvd, arat, npt, nrip_save, umip, pku, ospke, rdpid, fsrm
Ecpus
93.6
Scalability
58.5

System Resources and Accelerators

MEMORY
Memory Amount
640 GiB
Memory Amount Actual
625 GiB
ACCELERATORS
Accelerator Count
0
Accelerator Memory Min
0 MiB
Accelerator Memory Total
0 MiB
Accelerators
    STORAGE
    Storage Size
    400 GB
    Storage Type
    nvme ssd
    Storages
    • 400 GB nvme ssd
    NETWORK
    Network Speed Baseline
    20 Gbps
    Network Speed Max
    20 Gbps
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    0 GB/month
    IPv4
    1

    CPU and System Topology

    Server Description

    A high-density, 160-core dedicated processor platform designed for parallel compute, software compilation, and high-throughput memory operations.

    General PurposeCompute Optimized

    OVHcloud b3-640 is a general-purpose server featuring 160 dedicated physical AMD EPYC cores on the x86_64 architecture, paired with 640 GB of RAM and a 400 GB local NVMe SSD. Operating at 2.3 GHz with a 20 Gbps network baseline, this KVM-based instance delivers top-tier multi-core and memory bandwidth performance, as evidenced by high scores in Passmark CPU and multi-threaded bzip3 decompression benchmarks. However, single-core performance and single-threaded compression tasks remain average. Lacking GPU acceleration, it struggles with large LLM text generation, but excels at highly parallelized workloads. The server is suited for software compilation, in-memory databases, and parallel data processing.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    4.1Score
    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) 10,782,099 ops/sec 1,918,343 ops/sec 5.8 50.00% 50.00% +140.00%
    Redis RPS (pipeline=16, SET) 95,392,108 ops/sec 11,704,734 ops/sec 7.9 20.00% 20.00% +51.00%
    PassMark Memory Mark (composite) 3,003 2,416 1.2 10.00% 10.00% +1.80%
    Memory bandwidth (read, 16 MB ~ L3) 276,182 MB/sec 108,466 MB/sec 2.5 10.00% 10.00% +9.60%
    PassMark single-thread CPU 2,941 Mops/s 2,365 Mops/s 1.2 10.00% 10.00% +1.80%

    Memory Bandwidth

    Compression

    OpenSSL

    Geekbench Single-Core

    Score: 1,912

    Geekbench Multi-Core

    Score: 20,311

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 117,207
    Compression
    		 2,436,969
    Encryption
    		 160,620
    Extended Instructions
    		 150,405
    Floating Point Maths
    		 453,501
    Integer Maths
    		 764,616
    Physics
    		 18,558
    Prime Numbers
    		 1,397
    Single Threaded
    		 2,941
    String Sorting
    		 340,318

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 3,003
    Database Operations
    		 30,806
    Memory Latency
    		 72
    Memory Read Cached
    		 27,856
    Memory Read Uncached
    		 26,728
    Memory Write
    		 25,964

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 968.57 197.5 0.20×
    bitops
    		 160,429 14,185,288 88.42×
    branch
    		 212,853 21,619,095 101.57×
    funccall
    		 618.21 59,217 95.79×
    funcret
    		 20,973 1,896,688 90.43×
    hash
    		 162,677 17,830,303 109.61×
    intmath
    		 66,098 7,353,968 111.26×
    nop
    		 626,204 59,432,265 94.91×
    prime
    		 495,332 42,649,466 86.10×
    rotate
    		 88,275 7,865,990 89.11×
    str
    		 1,084,523 83,134,692 76.66×
    wcs
    		 673,164 49,299,259 73.24×
    CryptographySingleMultiScaling
    crypt
    		 3,738 289,077 77.33×
    CompressionSingleMultiScaling
    jpeg
    		 905.65 32,077 35.42×
    zlib
    		 72.15 6,988 96.86×
    HypervisorSingleMultiScaling
    clock
    		 16,275 181,536 11.15×
    clone
    		 270.36 8,861 32.77×
    context
    		 2,935 374,460 127.59×
    cpu-sched
    		 59.59 10,284 172.58×
    eventfd
    		 36,103 3,146,590 87.16×
    fault
    		 10,572 108,565 10.27×
    fork
    		 2,971 64,581 21.74×
    getrandom
    		 53,137 4,770,731 89.78×
    hrtimers
    		 720,132 8,577,699 11.91×
    mprotect
    		 584,641 21,078,822 36.05×
    mtx
    		 66,354,277 7,419,903,467 111.82×
    munmap
    		 2,193 74,925 34.17×
    null
    		 354,534 42,799,895 120.72×
    pipe
    		 355,594 26,216,541 73.73×
    priv-instr
    		 421,681 19,892,528 47.17×
    resched
    		 6,769,699 20,063,147 2.96×
    switch
    		 78,242 7,345,716 93.88×
    timermix
    		 167,404 17,292,217 103.30×
    tlb-shootdown
    		 13,310 197,180 14.81×
    tsc
    		 783,469 115,281,949 147.14×
    usersyscall
    		 324,704 18,621,395 57.35×
    vdso
    		 83,927,293 9,633,664,969 114.79×
    vfork
    		 18,101 85,946 4.75×
    x86syscall
    		 3,993,195 497,624,747 124.62×
    yield
    		 76,446,402 96,958,573 1.27×
    zero
    		 2,763,892 321,169,269 116.20×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 16,442 1,341,326 81.58×
    expmath
    		 2,859 302,011 105.62×
    fma
    		 920,701 89,163,129 96.84×
    fp
    		 251.99 26,174 103.87×
    fractal
    		 7.4 739.79 99.97×
    hyperbolic
    		 3,802 391,860 103.08×
    logmath
    		 4,563 449,910 98.61×
    matrix
    		 5,025 360,795 71.80×
    matrix-3d
    		 414.41 10,088 24.34×
    monte-carlo
    		 171.41 19,415 113.27×
    mpfr
    		 1,699 142,639 83.97×
    powmath
    		 1,296 144,945 111.87×
    trig
    		 1,874 181,729 96.98×
    veccmp
    		 19,049 1,591,423 83.54×
    vecfp
    		 1,812 167,551 92.45×
    vecmath
    		 4,023 421,859 104.85×
    vecshuf
    		 638.65 57,276 89.68×
    vecwide
    		 28,656 2,164,801 75.54×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 10.4 842.14 80.97×
    bsearch
    		 305.89 27,007 88.29×
    bubblesort
    		 0.8 70.6 88.25×
    fibsearch
    		 380.16 35,178 92.54×
    heapsort
    		 11 973.37 88.49×
    hsearch
    		 11,506 831,419 72.26×
    insertionsort
    		 13.2 839.99 63.64×
    malloc
    		 3,729,609 218,416,637 58.56×
    mcontend
    		 81.49 23,577 289.32×
    mergesort
    		 34.8 3,058 87.86×
    qsort
    		 22 1,865 84.76×
    skiplist
    		 7,006 722,976 103.20×
    tree
    		 2 133.52 66.76×
    MemorySingleMultiScaling
    cache
    		 740,689 1,736,800 2.34×
    cachehammer
    		 278,481 19,269,720 69.20×
    cacheline
    		 68.76 4,964 72.19×
    l1cache
    		 257.3 20,747 80.63×
    memcpy
    		 448.88 33,498 74.63×
    memrate
    		 3.73 30.35 8.14×
    memthrash
    		 413.74 123.06 0.30×
    ptr-chase
    		 40,595,338 1,141,093,832 28.11×
    spinmem
    		 222.39 7,082 31.84×
    stream
    		 128.12 184,956 1443.62×
    vm
    		 85,451 416,903 4.88×

    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

    PostgreSQL heavy read-only throughput

    Alternatives

    Servers of the Same Family

    INSTANCEvCPUsMEMORYAccelerators
    b2-7 27 GiB0
    b3-8 28 GiB0
    b2-15 415 GiB0
    b3-16 416 GiB0
    b2-30 830 GiB0
    b3-32 832 GiB0
    b2-60 1660 GiB0

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYAccelerators

    b3-640 FAQs