L48as_v3 by Microsoft Azure

L48as_v3 is a LASv3 family [AMD processor] [Premium Storage capable] 48 vCPUs server offered by Microsoft Azure with 48 vCPUs, 384 GiB of memory and 12 TB of storage. The pricing starts at 0.4488 USD per hour.
48 vCPU
384 GiB Memory
12 TB Storage
Spare SCore
64,763
(All-cores)
2,890
(Single-core)

Specifications

Server Metadata

Vendor ID
azure
Server ID
Standard_L48as_v3
Name
L48as_v3
API Reference
Standard_L48as_v3
Description
LASv3 family [AMD processor] [Premium Storage capable] 48 vCPUs
Family
LASv3
Hw Virt
Average Time To Start
101
Status
active
Observed At
2026-10-25T01:32:09.347923

Availability

REGION / IDSPOTONDEMAND
East US 2 (US) / eastus2
0.6919 USD/h 3.744 USD/h
North Central US (US) / northcentralus
0.6919 USD/h 3.744 USD/h
West US 2 (US) / westus2
0.6919 USD/h 3.744 USD/h
West US 3 (US) / westus3
0.6923 USD/h 3.744 USD/h
East US (US) / eastus
0.7885 USD/h 3.744 USD/h
Sweden Central (SE) / swedencentral
0.7229 USD/h 3.912 USD/h
Indonesia Central (ID) / indonesiacentral
0.73 USD/h 3.95 USD/h
Malaysia West (MY) / malaysiawest
0.73 USD/h 3.95 USD/h
Mexico Central (ZA) / mexicocentral
0.761 USD/h 4.118 USD/h
Canada Central (CA) / canadacentral
0.7629 USD/h 4.128 USD/h
Canada East (CA) / canadaeast
0.7629 USD/h 4.128 USD/h
North Europe (IE) / northeurope
0.7629 USD/h 4.128 USD/h
West US (US) / westus
0.7629 USD/h 4.128 USD/h
Central US (US) / centralus
0.7815 USD/h 4.231 USD/h
Central India (IN) / centralindia
0.785 USD/h 4.248 USD/h
Jio India Central (IN) / jioindiacentral
- 4.248 USD/h
Jio India West (IN) / jioindiawest
- 4.248 USD/h
Korea South (KR) / koreasouth
0.7957 USD/h 4.306 USD/h
France Central (FR) / francecentral
0.8028 USD/h 4.344 USD/h
Italy North (IT) / italynorth
0.8028 USD/h 4.344 USD/h
UK South (GB) / uksouth
0.8028 USD/h 4.344 USD/h
Japan East (JP) / japaneast
0.8116 USD/h 4.392 USD/h
Japan West (JP) / japanwest
0.8116 USD/h 4.392 USD/h
Korea Central (KR) / koreacentral
0.8116 USD/h 4.392 USD/h
Belgium Central (BE) / belgiumcentral
0.8249 USD/h 4.464 USD/h
Poland Central (PL) / polandcentral
0.8249 USD/h 4.464 USD/h
West Europe (NL) / westeurope
0.8249 USD/h 4.464 USD/h
Austria East (AT) / austriaeast
0.9075 USD/h 4.464 USD/h
Australia Central 2 (AU) / australiacentral2
0.4488 USD/h 4.488 USD/h
Australia Central (AU) / australiacentral
0.8294 USD/h 4.488 USD/h
Australia East (AU) / australiaeast
0.8294 USD/h 4.488 USD/h
Southeast Asia (SG) / southeastasia
0.8294 USD/h 4.488 USD/h
South Central US (US) / southcentralus
0.8303 USD/h 4.493 USD/h
UK West (GB) / ukwest
0.8371 USD/h 4.53 USD/h
Qatar Central (QA) / qatarcentral
0.8383 USD/h 4.536 USD/h
UAE North (AE) / uaenorth
0.8383 USD/h 4.536 USD/h
West India (IN) / westindia
0.8856 USD/h 4.792 USD/h
Australia Southeast (AU) / australiasoutheast
0.8994 USD/h 4.867 USD/h
Switzerland North (CH) / switzerlandnorth
0.907 USD/h 4.908 USD/h
South Africa North (ZA) / southafricanorth
0.9092 USD/h 4.92 USD/h
East Asia (HK) / eastasia
0.9137 USD/h 4.944 USD/h
Chile Central (CL) / chilecentral
0.9687 USD/h 5.242 USD/h
South India (IN) / southindia
0.9756 USD/h 5.279 USD/h
Norway East (NO) / norwayeast
0.9894 USD/h 5.354 USD/h
France South (FR) / francesouth
1.0436 USD/h 5.647 USD/h
Germany North (DE) / germanynorth
1.0724 USD/h 5.803 USD/h
UAE Central (AE) / uaecentral
1.0898 USD/h 5.897 USD/h
Brazil South (BR) / brazilsouth
1.1044 USD/h 5.976 USD/h
Switzerland West (CH) / switzerlandwest
1.179 USD/h 6.38 USD/h
South Africa West (ZA) / southafricawest
1.182 USD/h 6.396 USD/h
Norway West (NO) / norwaywest
1.2862 USD/h 6.96 USD/h
Brazil Southeast (US) / brazilsoutheast
1.4357 USD/h 7.769 USD/h

Processor

vCPUs
48
Hypervisor
Microsoft Hyper-V
CPU Allocation
Dedicated
CPU Cores
24
CPU Speed
3.525 GHz
CPU Architecture
x86_64
CPU Manufacturer
AMD
CPU Family
EPYC
CPU Model
7763
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, svm, cr8_legacy, abm, sse4a, misalignsse, 3dnowprefetch, osvw, topoext, 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, npt, nrip_save, tsc_scale, vmcb_clean, flushbyasid, decodeassists, pausefilter, pfthreshold, v_vmsave_vmload, umip, vaes, vpclmulqdq, rdpid, fsrm
Ecpus
22.4
Scalability
46.67

System Resources and Accelerators

MEMORY
Memory Amount
384 GiB
Memory Amount Actual
384 GiB
ACCELERATORS
Accelerator Count
0
Accelerator Memory Min
0 MiB
Accelerator Memory Total
0 MiB
Accelerators
    STORAGE
    Storage Size
    12012 GB
    Storage Type
    nvme ssd
    Storages
    • 492 GB ssd (temp disk)
    • 11520 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 storage and memory-optimized platform featuring dedicated AMD EPYC processors and massive local NVMe SSD capacity.

    Storage & DatabaseMemory Optimized

    The Microsoft Azure L48as_v3 is an x86_64 server from the LASv3 family, powered by a dedicated AMD EPYC 7763 processor. It delivers 48 vCPUs across 24 physical cores at 3.525 GHz, paired with 384 GB of RAM and an extensive 12,012 GB local NVMe SSD. This hardware configuration yields strong multi-core CPU and memory benchmark results, placing in the top 25% for multi-threaded processing, cryptography, and image manipulation. While memory bandwidth and single-core performance remain in the average tier, the massive local storage capacity and high memory-to-core ratio make this instance highly cost-efficient for storage-heavy and memory-intensive workloads. It is best suited for large-scale relational databases, NoSQL platforms, data warehousing, and high-throughput transactional applications.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    1.7Score
    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) 3,827,665 ops/sec 1,919,289 ops/sec 2 50.00% 50.00% +41.00%
    Redis RPS (pipeline=16, SET) 27,702,349 ops/sec 11,728,058 ops/sec 2.3 20.00% 20.00% +18.00%
    PassMark Memory Mark (composite) 2,912 2,416 1.2 10.00% 10.00% +1.80%
    Memory bandwidth (read, 16 MB ~ L3) 112,802 MB/sec 108,466 MB/sec 1 10.00% 10.00% 0.00%
    PassMark single-thread CPU 2,395 Mops/s 2,365 Mops/s 1 10.00% 10.00% 0.00%

    Memory Bandwidth

    Compression

    OpenSSL

    Geekbench Single-Core

    Score: 1,713

    Geekbench Multi-Core

    Score: 16,236

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 51,294
    Compression
    		 755,442
    Encryption
    		 51,422
    Extended Instructions
    		 44,994
    Floating Point Maths
    		 109,579
    Integer Maths
    		 199,366
    Physics
    		 5,139
    Prime Numbers
    		 278
    Single Threaded
    		 2,395
    String Sorting
    		 102,982

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 2,912
    Database Operations
    		 19,299
    Memory Latency
    		 63
    Memory Read Cached
    		 24,461
    Memory Read Uncached
    		 19,193
    Memory Write
    		 19,340

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 414.78 156.13 0.38×
    bitops
    		 141,692 3,882,797 27.40×
    branch
    		 164,444 7,143,596 43.44×
    funccall
    		 539.91 16,638 30.82×
    funcret
    		 16,211 517,873 31.95×
    hash
    		 136,405 4,947,008 36.27×
    intmath
    		 58,447 1,872,467 32.04×
    nop
    		 283,223 6,778,413 23.93×
    prime
    		 394,211 11,501,019 29.17×
    rotate
    		 77,134 1,838,391 23.83×
    str
    		 869,592 24,224,210 27.86×
    wcs
    		 568,659 15,774,480 27.74×
    CryptographySingleMultiScaling
    crypt
    		 3,236 74,906 23.15×
    CompressionSingleMultiScaling
    jpeg
    		 740.47 23,536 31.78×
    zlib
    		 55.1 2,270 41.19×
    HypervisorSingleMultiScaling
    clock
    		 3,720 91,835 24.68×
    clone
    		 1,017 19,586 19.25×
    context
    		 3,415 137,805 40.36×
    cpu-sched
    		 54.83 3,091 56.38×
    eventfd
    		 24,735 1,614,597 65.28×
    fault
    		 7,318 155,191 21.21×
    fork
    		 3,554 49,728 13.99×
    getrandom
    		 45,202 1,376,788 30.46×
    hrtimers
    		 69,247 1,811,681 26.16×
    mprotect
    		 681,688 8,317,487 12.20×
    mtx
    		 51,010,840 2,516,202,401 49.33×
    munmap
    		 2,510 44,555 17.75×
    null
    		 437,311 16,822,136 38.47×
    pipe
    		 1,733,504 20,410,945 11.77×
    priv-instr
    		 387,937 13,964,401 36.00×
    resched
    		 6,238,223 2,321,267 0.37×
    switch
    		 49,141 7,005,669 142.56×
    timermix
    		 41,681 1,992,005 47.79×
    tlb-shootdown
    		 10,421 109,828 10.54×
    tsc
    		 668,408 30,381,957 45.45×
    usersyscall
    		 309,019 11,206,184 36.26×
    vdso
    		 78,249,428 3,037,054,442 38.81×
    vfork
    		 18,517 168,272 9.09×
    x86syscall
    		 5,409,961 209,253,086 38.68×
    yield
    		 28,490,394 31,205,946 1.10×
    zero
    		 2,897,448 104,062,548 35.92×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 13,693 343,421 25.08×
    expmath
    		 2,426 91,902 37.88×
    fma
    		 779,650 29,085,757 37.31×
    fp
    		 227.18 7,879 34.68×
    fractal
    		 6.8 264.32 38.87×
    hyperbolic
    		 3,292 118,867 36.11×
    logmath
    		 3,882 132,074 34.02×
    matrix
    		 2,704 95,386 35.28×
    matrix-3d
    		 88.89 2,904 32.67×
    monte-carlo
    		 204.09 7,708 37.77×
    mpfr
    		 1,399 40,398 28.88×
    powmath
    		 1,175 45,961 39.12×
    trig
    		 1,493 54,871 36.76×
    veccmp
    		 16,056 496,961 30.95×
    vecfp
    		 1,477 49,578 33.57×
    vecmath
    		 3,434 131,282 38.23×
    vecshuf
    		 534.1 14,779 27.67×
    vecwide
    		 23,663 716,739 30.29×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 8.8 218.59 24.84×
    bsearch
    		 263.83 8,211 31.12×
    bubblesort
    		 0.8 26 32.50×
    fibsearch
    		 320.8 9,162 28.56×
    heapsort
    		 9.4 273.19 29.06×
    hsearch
    		 7,842 206,114 26.28×
    insertionsort
    		 11.6 267.6 23.07×
    malloc
    		 2,865,613 66,295,341 23.13×
    mcontend
    		 109.7 6,601 60.17×
    mergesort
    		 27.2 948.98 34.89×
    qsort
    		 14.8 618.58 41.80×
    skiplist
    		 6,206 224,818 36.23×
    tree
    		 1.8 33.59 18.66×
    MemorySingleMultiScaling
    cache
    		 603,912 1,395,590 2.31×
    cachehammer
    		 238,413 8,525,690 35.76×
    cacheline
    		 27.99 2,041 72.90×
    l1cache
    		 152.48 3,701 24.27×
    memcpy
    		 365.93 8,220 22.46×
    memrate
    		 2.75 9.5 3.45×
    memthrash
    		 163.38 996.14 6.10×
    ptr-chase
    		 65,372,876 371,115,900 5.68×
    spinmem
    		 253.59 1,746 6.88×
    stream
    		 85.97 19,938 231.92×
    vm
    		 153,488 297,827 1.94×

    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
    L8as_v3 864 GiB0
    L16as_v3 16128 GiB0
    L32as_v3 32256 GiB0
    L64as_v3 64512 GiB0
    L80as_v3 80640 GiB0

    Similar Servers

    INSTANCEVENDORvCPUsMEMORYAccelerators

    L48as_v3 FAQs