GS4-4 by Microsoft Azure

GS4-4 is a GS family [Intel processor] 16 vCPUs server offered by Microsoft Azure with 4 vCPUs, 224 GiB of memory and 459 GB of storage. The pricing starts at 0.6494 USD per hour.
4 vCPU
224 GiB Memory
459 GB Storage
Spare SCore
2,028
(All-cores)
491
(Single-core)

Specifications

Server Metadata

Vendor ID
azure
Server ID
Standard_GS4-4
Name
GS4-4
API Reference
Standard_GS4-4
Description
GS family [Intel processor] 16 vCPUs
Family
GS
Hw Virt
Average Time To Start
39
Status
planned-for-retirement
Observed At
2026-11-05T22:01:13.118097

Availability

REGION / IDSPOTONDEMAND
UK South (GB) / uksouth
0.6494 USD/h 3.514 USD/h
Canada East (CA) / canadaeast
0.7155 USD/h 3.872 USD/h
Canada Central (CA) / canadacentral
1.0671 USD/h 3.872 USD/h
West US 2 (US) / westus2
0.7248 USD/h 3.922 USD/h
East US 2 (US) / eastus2
1.5118 USD/h 4.4 USD/h
Japan East (JP) / japaneast
0.8505 USD/h 4.602 USD/h
West US (US) / westus
0.9018 USD/h 4.88 USD/h
Australia East (AU) / australiaeast
0.9437 USD/h 5.104 USD/h
Southeast Asia (SG) / southeastasia
0.9757 USD/h 5.28 USD/h
West Europe (NL) / westeurope
1.0349 USD/h 5.6 USD/h

Processor

vCPUs
4
Hypervisor
Microsoft Hyper-V
CPU Allocation
Dedicated
CPU Cores
4
CPU Speed
2 GHz
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU Model
E5-2698B v3
CPU L1D Cache
48 KiB
CPU L1D Cache Total
192 KiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
128 KiB
CPU L2 Cache
1 MiB
CPU L2 Cache Total
5 MiB
CPU L3 Cache
48 MiB
CPU L3 Cache Total
48 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, ss, ht, syscall, nx, pdpe1gb, rdtscp, lm, constant_tsc, rep_good, nopl, xtopology, cpuid, pni, pclmulqdq, ssse3, fma, cx16, pcid, sse4_1, sse4_2, movbe, popcnt, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, abm, pti, fsgsbase, bmi1, avx2, smep, bmi2, erms, invpcid, xsaveopt, md_clear
Ecpus
4.1
Scalability
102.5

System Resources and Accelerators

MEMORY
Memory Amount
224 GiB
Memory Amount Actual
224 GiB
ACCELERATORS
Accelerator Count
0
Accelerator Memory Min
0 MiB
Accelerator Memory Total
0 MiB
Accelerators
    STORAGE
    Storage Size
    459 GB
    Storage Type
    ssd
    Storages
    • 459 GB ssd (temp disk)
    NETWORK
    Inbound Traffic
    0 GB/month
    Outbound Traffic
    0 GB/month
    IPv4
    0

    CPU and System Topology

    Server Description

    An x86 virtual machine offering high memory density and local storage for memory-bound workloads with modest compute requirements.

    Memory OptimizedStorage & Database

    Microsoft Azure GS4-4 is a memory-optimized virtual machine instance running on Microsoft Hyper-V. It features an Intel Xeon E5-2698B v3 processor with 4 dedicated physical cores and 4 vCPUs running at 2.0 GHz. A key characteristic of this instance is its high memory density, providing 224.0 GB of RAM (56.0 GB per core) alongside 459 GB of local storage, with no GPU acceleration. Performance benchmarks indicate modest compute capabilities, with stress-ng and Geekbench scores falling in the bottom 10% to 25% ranges, while PassMark memory performance remains average. Given the high memory-to-vCPU ratio and older CPU architecture, the GS4-4 is qualitatively cost-effective for memory-bound, low-compute workloads. It is best suited for legacy database hosting, in-memory caching, and enterprise applications requiring a large memory footprint without intensive parallel processing requirements.

    Economics

    Average Price per Region

    Prices per Zone

    Lowest Prices

    Performance

    Workload Profiles

    0.31Score
    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) 413,371 ops/sec 1,936,298 ops/sec 0.22 50.00% 50.00% -53.00%
    Redis RPS (pipeline=16, SET) 3,059,970 ops/sec 11,797,257 ops/sec 0.26 20.00% 20.00% -24.00%
    PassMark Memory Mark (composite) 1,989 2,407 0.83 10.00% 10.00% -1.80%
    Memory bandwidth (read, 16 MB ~ L3) 36,612 MB/sec 108,490 MB/sec 0.34 10.00% 10.00% -10.00%
    PassMark single-thread CPU 1,922 Mops/s 2,365 Mops/s 0.79 10.00% 10.00% -2.30%

    Memory Bandwidth

    Compression

    OpenSSL

    Geekbench Single-Core

    Score: 1,083

    Geekbench Multi-Core

    Score: 3,239

    Passmark CPU Scores

    BENCHMARKSCORE
    Mark
    		 4,978
    Compression
    		 69,194
    Encryption
    		 1,267
    Extended Instructions
    		 4,261
    Floating Point Maths
    		 10,301
    Integer Maths
    		 13,746
    Physics
    		 709
    Prime Numbers
    		 50
    Single Threaded
    		 1,922
    String Sorting
    		 9,718

    Passmark Memory Scores

    BENCHMARKSCORE
    Memory Mark
    		 1,989
    Database Operations
    		 2,424
    Memory Latency
    		 65
    Memory Read Cached
    		 21,113
    Memory Read Uncached
    		 8,945
    Memory Write
    		 7,836

    Stress-ng Stressors

    CPU integerSingleMultiScaling
    atomic
    		 188.71 190 1.01×
    bitops
    		 93,291 371,381 3.98×
    branch
    		 98,156 390,110 3.97×
    funccall
    		 455.01 1,802 3.96×
    funcret
    		 8,679 34,334 3.96×
    hash
    		 108,717 427,345 3.93×
    intmath
    		 43,637 172,682 3.96×
    nop
    		 134,165 531,073 3.96×
    prime
    		 287,705 1,139,092 3.96×
    rotate
    		 69,895 277,383 3.97×
    str
    		 602,795 2,442,576 4.05×
    wcs
    		 445,282 1,706,078 3.83×
    CryptographySingleMultiScaling
    crypt
    		 2,033 8,072 3.97×
    CompressionSingleMultiScaling
    jpeg
    		 493.76 1,952 3.95×
    zlib
    		 33.17 141.05 4.25×
    HypervisorSingleMultiScaling
    clock
    		 17,676 64,162 3.63×
    clone
    		 1,191 3,080 2.59×
    context
    		 1,490 5,932 3.98×
    cpu-sched
    		 54.97 273.08 4.97×
    eventfd
    		 32,416 318,296 9.82×
    fault
    		 9,715 32,217 3.32×
    fork
    		 3,861 13,365 3.46×
    getrandom
    		 34,403 136,291 3.96×
    hrtimers
    		 139,004 503,026 3.62×
    mprotect
    		 686,330 781,622 1.14×
    mtx
    		 11,597,233 147,823,982 12.75×
    munmap
    		 2,787 10,182 3.65×
    null
    		 160,078 636,953 3.98×
    pipe
    		 832,209 1,899,658 2.28×
    priv-instr
    		 344,182 1,331,679 3.87×
    resched
    		 804,192 419,914 0.52×
    switch
    		 97,833 862,336 8.81×
    timermix
    		 158,129 624,564 3.95×
    tlb-shootdown
    		 7,210 10,235 1.42×
    tsc
    		 640,905 2,538,687 3.96×
    usersyscall
    		 242,550 941,211 3.88×
    vdso
    		 65,346,073 259,523,312 3.97×
    vfork
    		 22,943 87,719 3.82×
    x86syscall
    		 1,817,516 7,149,792 3.93×
    yield
    		 2,357,162 2,389,000 1.01×
    zero
    		 1,356,974 5,395,215 3.98×
    CPU float/SIMDSingleMultiScaling
    eigen
    		 10,588 41,883 3.96×
    expmath
    		 2,069 8,175 3.95×
    fma
    		 486,034 1,925,438 3.96×
    fp
    		 143.52 541.49 3.77×
    fractal
    		 3.6 14 3.89×
    hyperbolic
    		 2,206 8,736 3.96×
    logmath
    		 3,467 13,764 3.97×
    matrix
    		 2,718 10,791 3.97×
    matrix-3d
    		 569.3 1,541 2.71×
    monte-carlo
    		 85.29 340.99 4.00×
    mpfr
    		 1,010 4,001 3.96×
    powmath
    		 1,208 4,775 3.95×
    trig
    		 1,414 5,599 3.96×
    veccmp
    		 14,886 58,866 3.95×
    vecfp
    		 1,093 4,330 3.96×
    vecmath
    		 2,350 9,312 3.96×
    vecshuf
    		 476.13 1,886 3.96×
    vecwide
    		 22,707 89,581 3.94×
    AlgorithmsSingleMultiScaling
    bitonicsort
    		 5.8 23 3.97×
    bsearch
    		 177.33 705.09 3.98×
    bubblesort
    		 0.6 2.4 4.00×
    fibsearch
    		 207.71 823.03 3.96×
    heapsort
    		 6.6 26.4 4.00×
    hsearch
    		 5,036 20,224 4.02×
    insertionsort
    		 7.4 29.6 4.00×
    malloc
    		 1,759,502 8,881,705 5.05×
    mcontend
    		 107.12 514.92 4.81×
    mergesort
    		 19.2 76.2 3.97×
    qsort
    		 14.8 58.4 3.95×
    skiplist
    		 3,895 15,395 3.95×
    tree
    		 1.2 4 3.33×
    MemorySingleMultiScaling
    cache
    		 238,279 296,332 1.24×
    cachehammer
    		 198,657 792,329 3.99×
    cacheline
    		 41.11 156.2 3.80×
    l1cache
    		 658.62 2,614 3.97×
    memcpy
    		 291.84 1,158 3.97×
    memrate
    		 1.17 3.9 3.33×
    memthrash
    		 89.77 161.54 1.80×
    ptr-chase
    		 18,859,339 32,743,631 1.74×
    spinmem
    		 287.59 289.3 1.01×
    stream
    		 29.31 395.12 13.48×
    vm
    		 59,521 262,731 4.41×

    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

    GS4-4 FAQs