m2-ultramem-208 by Google Cloud Platform

m2-ultramem-208 is a 208 vCPUs, 5.75 TB RAM server offered by Google Cloud Platform with 208 vCPUs, 5888 GiB of memory and 0 GB of storage. The pricing starts at 42.1119 USD per hour.
208 vCPU
5888 GiB Memory
Spare SCore
183,860
(All-cores)
1,344
(Single-core)

Specifications

Server Metadata

Vendor ID
gcp
Server ID
804208
Name
m2-ultramem-208
Description
208 vCPUs, 5.75 TB RAM
Family
m2
Hw Virt
Average Time To Start
26
Status
active
Observed At
2026-10-27T07:46:20.433202

Availability

REGION / IDSPOTONDEMAND
Council Bluffs (US) / us-central1
- 42.1119 USD/h
Montréal (CA) / northamerica-northeast1
- 44.2752 USD/h
Eemshaven (NL) / europe-west4
- 44.2779 USD/h
Toronto (CA) / northamerica-northeast2
- 46.3231 USD/h
St. Ghislain (BE) / europe-west1
- 46.4361 USD/h
Ashburn (US) / us-east4
- 47.5388 USD/h
Las Vegas (US) / us-west4
- 47.5396 USD/h
Delhi (IN) / asia-south2
- 48.2925 USD/h
Mumbai (IN) / asia-south1
- 48.2993 USD/h
Milan (IT) / europe-west8
- 48.8498 USD/h
Jurong West (SG) / asia-southeast1
- 49.6009 USD/h
Madrid (ES) / europe-southwest1
- 49.6921 USD/h
Tokyo (JP) / asia-northeast1
- 51.4785 USD/h
Osaka (JP) / asia-northeast2
- 51.4798 USD/h
Seoul (KR) / asia-northeast3
- 51.4798 USD/h
London (GB) / europe-west2
- 51.7658 USD/h
Frankfurt (DE) / europe-west3
- 51.7658 USD/h
Sydney (AU) / australia-southeast1
- 57.0613 USD/h
Santiago (CL) / southamerica-west1
- 60.2201 USD/h
Osasco (BR) / southamerica-east1
- 63.6181 USD/h
Dammam (SA) / me-central2
- 67.3791 USD/h

Processor

vCPUs
208
CPU Allocation
Dedicated
CPU Cores
104
CPU Speed
2.7 GHz
CPU Architecture
x86_64
CPU Manufacturer
Intel
CPU Family
Xeon
CPU L1D Cache
32 KiB
CPU L1D Cache Total
3 MiB
CPU L1I Cache
32 KiB
CPU L1I Cache Total
3 MiB
CPU L2 Cache
1 MiB
CPU L2 Cache Total
104 MiB
CPU L3 Cache
39 MiB
CPU L3 Cache Total
154 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, nonstop_tsc, cpuid, 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, abm, 3dnowprefetch, ssbd, ibrs, ibpb, stibp, ibrs_enhanced, fsgsbase, tsc_adjust, bmi1, hle, avx2, smep, bmi2, erms, invpcid, rtm, mpx, avx512f, avx512dq, rdseed, adx, smap, clflushopt, clwb, avx512cd, avx512bw, avx512vl, xsaveopt, xsavec, xgetbv1, xsaves, arat, avx512_vnni, md_clear, arch_capabilities
Ecpus
136.8
Scalability
65.77

System Resources and Accelerators

MEMORY
Memory Amount
6 TB
Memory Amount Actual
6 TB
ACCELERATORS
Accelerator Count
0
Accelerator Memory Min
0 MiB
Accelerator Memory Total
0 MiB
Accelerators
    STORAGE
    Storage Size
    0 GB
    Storages
      NETWORK
      Inbound Traffic
      0 GB/month
      Outbound Traffic
      0 GB/month
      IPv4
      0

      CPU and System Topology

      Server Description

      A high-density memory-optimized instance featuring massive RAM capacity and strong multi-core performance for enterprise in-memory databases and caching.

      Memory Optimized

      Google Cloud Platform m2-ultramem-208 is an x86_64 memory-optimized server instance configured with 208 dedicated Intel Xeon vCPUs operating at 2.7 GHz and an extensive 5888.0 GB of RAM. This instance does not include bundled local storage or GPUs. Benchmark data highlights a significant contrast in performance: while single-core metrics are weak, multi-core CPU performance and cached memory bandwidth rank in the top 10%. Large-block memory read and write speeds and PassMark CPU scores are also strong. The primary tradeoff is lower single-threaded efficiency, making it less suitable for sequential processing. Qualitatively, the m2-ultramem-208 offers high resource density for memory-bound architectures. It is best suited for large-scale in-memory databases, massive Redis caching layers, and enterprise-grade data processing workloads requiring substantial memory capacity.

      Economics

      Average Price per Region

      Prices per Zone

      Lowest Prices

      Performance

      Workload Profiles

      4.4Score
      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) 16,245,235 ops/sec 1,918,343 ops/sec 8.4 50.00% 50.00% +190.00%
      Redis RPS (pipeline=16, SET) 86,887,990 ops/sec 11,704,734 ops/sec 7.2 20.00% 20.00% +48.00%
      PassMark Memory Mark (composite) 1,293 2,416 0.54 10.00% 10.00% -6.00%
      Memory bandwidth (read, 16 MB ~ L3) 333,965 MB/sec 108,420 MB/sec 3 10.00% 10.00% +12.00%
      PassMark single-thread CPU 1,894 Mops/s 2,365 Mops/s 0.79 10.00% 10.00% -2.30%

      Memory Bandwidth

      Compression

      OpenSSL

      Geekbench Single-Core

      Score: 961

      Geekbench Multi-Core

      Score: 8,695

      Passmark CPU Scores

      BENCHMARKSCORE
      Mark
      		 79,943
      Compression
      		 2,547,461
      Encryption
      		 78,641
      Extended Instructions
      		 164,000
      Floating Point Maths
      		 326,956
      Integer Maths
      		 609,568
      Physics
      		 8,284
      Prime Numbers
      		 543
      Single Threaded
      		 1,894
      String Sorting
      		 278,021

      Passmark Memory Scores

      BENCHMARKSCORE
      Memory Mark
      		 1,293
      Database Operations
      		 15,082
      Memory Latency
      		 74
      Memory Read Cached
      		 22,249
      Memory Read Uncached
      		 2,836
      Memory Write
      		 2,962

      Stress-ng Stressors

      CPU integerSingleMultiScaling
      atomic
      		 217.44 86.44 0.40×
      bitops
      		 104,283 12,432,009 119.21×
      branch
      		 148,490 22,243,652 149.80×
      funccall
      		 523.51 56,107 107.17×
      funcret
      		 10,926 1,567,229 143.44×
      hash
      		 121,563 17,406,740 143.19×
      intmath
      		 21,589 3,259,889 151.00×
      nop
      		 184,221 16,550,847 89.84×
      prime
      		 303,872 34,664,061 114.07×
      rotate
      		 93,248 13,219,907 141.77×
      str
      		 695,720 77,237,536 111.02×
      wcs
      		 453,581 49,104,836 108.26×
      CryptographySingleMultiScaling
      crypt
      		 2,351 236,963 100.78×
      CompressionSingleMultiScaling
      jpeg
      		 567.62 65,217 114.90×
      zlib
      		 46.41 7,876 169.71×
      HypervisorSingleMultiScaling
      clock
      		 17,627 79,943 4.54×
      clone
      		 313.39 9,745 31.09×
      context
      		 2,477 277,333 111.96×
      cpu-sched
      		 62.34 12,246 196.44×
      eventfd
      		 96,851 2,059,811 21.27×
      fault
      		 14,435 71,613 4.96×
      fork
      		 3,691 18,854 5.11×
      getrandom
      		 43,130 5,167,623 119.82×
      hrtimers
      		 356,917 10,655,350 29.85×
      mprotect
      		 401,616 13,917,988 34.66×
      mtx
      		 11,858,000 4,500,244,124 379.51×
      munmap
      		 2,727 25,205 9.24×
      null
      		 280,415 26,543,089 94.66×
      pipe
      		 813,817 37,691,320 46.31×
      priv-instr
      		 544,958 8,481,564 15.56×
      resched
      		 2,623,606 15,882,978 6.05×
      switch
      		 157,154 34,583,262 220.06×
      timermix
      		 168,271 24,229,280 143.99×
      tlb-shootdown
      		 4,253 41,422 9.74×
      tsc
      		 961,507 137,426,773 142.93×
      usersyscall
      		 371,103 7,430,296 20.02×
      vdso
      		 79,928,686 11,576,699,239 144.84×
      vfork
      		 22,475 29,382 1.31×
      x86syscall
      		 3,204,056 289,338,665 90.30×
      yield
      		 82,191,839 136,475,319 1.66×
      zero
      		 1,893,786 213,150,409 112.55×
      CPU float/SIMDSingleMultiScaling
      eigen
      		 15,565 1,851,464 118.95×
      expmath
      		 2,582 405,102 156.87×
      fma
      		 543,241 91,803,523 168.99×
      fp
      		 115.2 15,318 132.97×
      fractal
      		 4.8 720.3 150.06×
      hyperbolic
      		 2,682 410,871 153.22×
      logmath
      		 4,370 664,986 152.18×
      matrix
      		 2,905 473,641 163.06×
      matrix-3d
      		 598.12 10,210 17.07×
      monte-carlo
      		 129.52 15,416 119.02×
      mpfr
      		 1,166 129,419 111.03×
      powmath
      		 1,531 236,201 154.31×
      trig
      		 1,651 220,107 133.31×
      veccmp
      		 15,006 1,614,399 107.59×
      vecfp
      		 1,268 169,953 134.01×
      vecmath
      		 2,106 341,322 162.10×
      vecshuf
      		 543.83 60,785 111.77×
      vecwide
      		 22,487 2,639,139 117.36×
      AlgorithmsSingleMultiScaling
      bitonicsort
      		 7 787.59 112.51×
      bsearch
      		 232.27 27,484 118.33×
      bubblesort
      		 0.6 67.6 112.67×
      fibsearch
      		 234.73 27,877 118.76×
      heapsort
      		 8.4 1,033 122.97×
      hsearch
      		 4,735 605,433 127.86×
      insertionsort
      		 7.8 811.99 104.10×
      malloc
      		 1,876,087 69,633,738 37.12×
      mcontend
      		 48.42 33,939 700.94×
      mergesort
      		 22.4 2,876 128.41×
      qsort
      		 14.4 1,682 116.83×
      skiplist
      		 4,701 682,606 145.19×
      tree
      		 1.4 81.18 57.99×
      MemorySingleMultiScaling
      cache
      		 90,394 3,932,269 43.50×
      cachehammer
      		 221,662 7,105,155 32.05×
      cacheline
      		 6.95 29.86 4.30×
      l1cache
      		 212.65 1,814 8.53×
      memcpy
      		 335.06 37,235 111.13×
      memrate
      		 0.58 27.17 46.84×
      memthrash
      		 263.45 703.01 2.67×
      ptr-chase
      		 33,677,187 1,325,354,816 39.35×
      spinmem
      		 207.79 7,805 37.56×
      stream
      		 2.54 20,820 8196.82×
      vm
      		 30,296 - -

      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
      m2-megamem-416 4165888 GiB0
      m2-hypermem-416 4168832 GiB0
      m2-ultramem-416 41611776 GiB0

      Similar Servers

      INSTANCEVENDORvCPUsMEMORYAccelerators

      m2-ultramem-208 FAQs