r8a.4xlarge by Amazon Web Services

r8a.4xlarge is a Memory optimized [AMD processors] Gen8 4xlarge server offered by Amazon Web Services with 16 vCPUs, 128 GiB of memory and 0 GB of storage. The pricing starts at 0.143 USD per hour.
16 vCPU
128 GiB Memory
Spare SCore
60,224
(All-cores)
3,773
(Single-core)

Specifications

Server Metadata

Vendor ID
aws
Name
r8a.4xlarge
Description
Memory optimized [AMD processors] Gen8 4xlarge
Family
r8a
Hw Virt
Status
active
Observed At
2026-10-24T22:03:23.052524

Availability

REGION / IDSPOTONDEMAND
Ohio (US) / us-east-2
0.364733 USD/h 1.2781 USD/h
Oregon (US) / us-west-2
0.41265 USD/h 1.2781 USD/h
Northern Virgina (US) / us-east-1
0.51362 USD/h 1.2781 USD/h
Taipei (TW) / ap-east-2
0.489933 USD/h 1.3876 USD/h
Quebec (CA) / ca-central-1
0.3757 USD/h 1.3998 USD/h
Aragón (ES) / eu-south-2
0.291167 USD/h 1.4302 USD/h
Dublin (IE) / eu-west-1
0.7139 USD/h 1.4302 USD/h
Tokyo (JP) / ap-northeast-1
0.478767 USD/h 1.5418 USD/h
Frankfurt (DE) / eu-central-1
0.657667 USD/h 1.5418 USD/h

Processor

vCPUs
16
Hypervisor
nitro
CPU Allocation
Dedicated
CPU Cores
16
CPU Speed
4.5 GHz
CPU Architecture
x86_64
CPU Manufacturer
AMD
CPU Family
EPYC
CPU Model
9R45
CPU L1D Cache
48 KiB
CPU L1D Cache Total
768 KiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
512 KiB
CPU L2 Cache
1 MiB
CPU L2 Cache Total
16 MiB
CPU L3 Cache
32 MiB
CPU L3 Cache Total
64 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, xtopology, nonstop_tsc, cpuid, extd_apicid, aperfmperf, tsc_known_freq, pni, pclmulqdq, monitor, ssse3, fma, cx16, pcid, sse4_1, sse4_2, x2apic, movbe, popcnt, aes, xsave, avx, f16c, rdrand, hypervisor, lahf_lm, cmp_legacy, cr8_legacy, abm, sse4a, misalignsse, 3dnowprefetch, topoext, perfctr_core, ssbd, perfmon_v2, ibrs, ibpb, stibp, ibrs_enhanced, vmmcall, fsgsbase, tsc_adjust, bmi1, avx2, smep, bmi2, erms, invpcid, avx512f, avx512dq, rdseed, adx, smap, avx512ifma, clflushopt, clwb, avx512cd, sha_ni, avx512bw, avx512vl, xsaveopt, xsavec, xgetbv1, xsaves, avx_vnni, avx512_bf16, clzero, xsaveerptr, rdpru, wbnoinvd, arat, avx512vbmi, pku, ospke, avx512_vbmi2, gfni, vaes, vpclmulqdq, avx512_vnni, avx512_bitalg, avx512_vpopcntdq, rdpid, movdiri, movdir64b, fsrm, avx512_vp2intersect, flush_l1d
Ecpus
16
Scalability
100

System Resources and Accelerators

MEMORY
Memory Amount
128 GiB
Memory Amount Actual
128 GiB
Memory Generation
DDR5
Memory Speed
6400 Mhz
ACCELERATORS
Accelerator Count
0
Accelerator Memory Min
0 MiB
Accelerator Memory Total
0 MiB
Accelerators
    STORAGE
    Storage Size
    0 GB
    Storages
      NETWORK
      Network Speed Baseline
      7.5 Gbps
      Network Speed Max
      15 Gbps
      Network Storage Speed Baseline
      5 Gbps
      Network Storage Speed Max
      10 Gbps
      Inbound Traffic
      0 GB/month
      Outbound Traffic
      0 GB/month
      IPv4
      0

      CPU and System Topology

      Server Description

      A memory-optimized x86 server featuring high-frequency AMD EPYC processing and dense DDR5 memory for demanding single-core and caching workloads.

      Memory OptimizedGeneral Purpose

      Amazon Web Services r8a.4xlarge is a memory-optimized instance featuring 16 dedicated physical cores powered by an AMD EPYC 9R45 processor running at 4.5 GHz. It is equipped with 128.0 GB of high-speed 6 GHz DDR5 memory, providing 8.0 GB of RAM per core, but lacks local storage and dedicated GPUs. Operating on the Nitro hypervisor, the instance delivers an 8 Gbps baseline network bandwidth. Benchmarks show top-tier single-core and multi-core Geekbench performance, alongside strong memory scores and static web serving throughput. Conversely, multi-core stress-ng and database operations perform in the average tier. Given its high memory-to-vCPU ratio and strong single-threaded execution, the r8a.4xlarge is highly suited for memory-intensive workloads, database caching, and small-scale language model inference, though it requires network-attached storage for data persistence.

      Economics

      Average Price per Region

      Prices per Zone

      Lowest Prices

      Performance

      Workload Profiles

      2.4Score
      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) 5,914,438 ops/sec 1,919,289 ops/sec 3.1 50.00% 50.00% +76.00%
      Redis RPS (pipeline=16, SET) 39,532,059 ops/sec 11,728,058 ops/sec 3.3 20.00% 20.00% +27.00%
      PassMark Memory Mark (composite) 2,841 2,416 1.2 10.00% 10.00% +1.80%
      Memory bandwidth (read, 16 MB ~ L3) 125,678 MB/sec 108,466 MB/sec 1.2 10.00% 10.00% +1.80%
      PassMark single-thread CPU 3,818 Mops/s 2,365 Mops/s 1.6 10.00% 10.00% +4.80%

      Memory Bandwidth

      Compression

      OpenSSL

      Geekbench Single-Core

      Score: 2,819

      Geekbench Multi-Core

      Score: 21,849

      Passmark CPU Scores

      BENCHMARKSCORE
      Mark
      		 47,027
      Compression
      		 528,920
      Encryption
      		 25,403
      Extended Instructions
      		 49,528
      Floating Point Maths
      		 122,198
      Integer Maths
      		 121,932
      Physics
      		 6,182
      Prime Numbers
      		 438
      Single Threaded
      		 3,818
      String Sorting
      		 73,148

      Passmark Memory Scores

      BENCHMARKSCORE
      Memory Mark
      		 2,841
      Database Operations
      		 20,757
      Memory Latency
      		 82
      Memory Read Cached
      		 33,950
      Memory Read Uncached
      		 32,481
      Memory Write
      		 33,010

      Stress-ng Stressors

      CPU integerSingleMultiScaling
      atomic
      		 373.56 337.25 0.90×
      bitops
      		 232,162 3,685,035 15.87×
      branch
      		 184,891 2,939,124 15.90×
      funccall
      		 909.38 14,264 15.69×
      funcret
      		 27,509 425,338 15.46×
      hash
      		 218,496 3,469,782 15.88×
      intmath
      		 80,632 1,280,802 15.88×
      nop
      		 431,855 6,867,257 15.90×
      prime
      		 626,956 9,851,232 15.71×
      rotate
      		 142,885 2,270,538 15.89×
      str
      		 1,583,603 25,691,030 16.22×
      wcs
      		 1,078,763 15,933,511 14.77×
      CryptographySingleMultiScaling
      crypt
      		 5,464 86,308 15.80×
      CompressionSingleMultiScaling
      jpeg
      		 1,283 20,242 15.77×
      zlib
      		 65.13 1,142 17.53×
      HypervisorSingleMultiScaling
      clock
      		 51,975 351,499 6.76×
      clone
      		 364.3 2,699 7.41×
      context
      		 9,268 146,624 15.82×
      cpu-sched
      		 62.61 1,030 16.45×
      eventfd
      		 136,564 4,021,351 29.45×
      fault
      		 47,255 307,701 6.51×
      fork
      		 6,880 72,363 10.52×
      getrandom
      		 100,892 1,592,953 15.79×
      hrtimers
      		 467,042 2,318,100 4.96×
      mprotect
      		 960,940 10,296,991 10.72×
      mtx
      		 42,469,599 1,342,229,244 31.60×
      munmap
      		 5,787 55,581 9.60×
      null
      		 1,515,379 24,179,234 15.96×
      pipe
      		 2,942,849 20,680,599 7.03×
      priv-instr
      		 1,599,252 19,564,344 12.23×
      resched
      		 16,124,967 5,752,751 0.36×
      switch
      		 345,433 10,596,298 30.68×
      timermix
      		 522,575 8,274,546 15.83×
      tlb-shootdown
      		 22,119 118,587 5.36×
      tsc
      		 890,531 14,149,138 15.89×
      usersyscall
      		 1,142,950 15,186,366 13.29×
      vdso
      		 111,698,308 1,777,266,571 15.91×
      vfork
      		 40,812 303,007 7.42×
      x86syscall
      		 19,961,354 316,425,867 15.85×
      yield
      		 36,377,713 41,302,086 1.14×
      zero
      		 10,437,127 168,058,908 16.10×
      CPU float/SIMDSingleMultiScaling
      eigen
      		 17,616 278,758 15.82×
      expmath
      		 4,025 63,682 15.82×
      fma
      		 1,570,013 24,801,504 15.80×
      fp
      		 201.6 3,247 16.10×
      fractal
      		 9 139.07 15.45×
      hyperbolic
      		 5,488 87,000 15.85×
      logmath
      		 5,740 90,936 15.84×
      matrix
      		 9,348 136,882 14.64×
      matrix-3d
      		 1,354 6,019 4.44×
      monte-carlo
      		 464.13 7,380 15.90×
      mpfr
      		 2,438 38,424 15.76×
      powmath
      		 1,850 29,320 15.85×
      trig
      		 2,450 38,736 15.81×
      veccmp
      		 30,310 477,107 15.74×
      vecfp
      		 2,444 38,429 15.72×
      vecmath
      		 4,759 75,507 15.87×
      vecshuf
      		 1,006 15,872 15.77×
      vecwide
      		 40,487 636,602 15.72×
      AlgorithmsSingleMultiScaling
      bitonicsort
      		 12.8 204.4 15.97×
      bsearch
      		 381.15 5,809 15.24×
      bubblesort
      		 1.2 19.2 16.00×
      fibsearch
      		 431.59 6,843 15.86×
      heapsort
      		 14 226 16.14×
      hsearch
      		 13,767 211,991 15.40×
      insertionsort
      		 9.6 166.2 17.31×
      malloc
      		 5,471,968 44,882,994 8.20×
      mcontend
      		 160.33 3,193 19.92×
      mergesort
      		 39.6 621.99 15.71×
      qsort
      		 27 424.39 15.72×
      skiplist
      		 8,493 134,069 15.79×
      tree
      		 2.6 26.6 10.23×
      MemorySingleMultiScaling
      cache
      		 1,152,025 1,956,056 1.70×
      cachehammer
      		 270,929 4,389,661 16.20×
      cacheline
      		 65.58 1,126 17.17×
      l1cache
      		 1,377 21,705 15.76×
      memcpy
      		 808.13 12,802 15.84×
      memrate
      		 0.99 9.28 9.37×
      memthrash
      		 200.4 963.07 4.81×
      ptr-chase
      		 75,980,986 157,426,863 2.07×
      spinmem
      		 291.99 1,074 3.68×
      stream
      		 162.87 8,454 51.90×
      vm
      		 244,488 293,970 1.20×

      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
      r8a.medium 18 GiB0
      r8a.large 216 GiB0
      r8a.xlarge 432 GiB0
      r8a.2xlarge 864 GiB0
      r8a.8xlarge 32256 GiB0
      r8a.12xlarge 48384 GiB0
      r8a.16xlarge 64512 GiB0

      Similar Servers

      INSTANCEVENDORvCPUsMEMORYAccelerators

      r8a.4xlarge FAQs