hpc7a.48xlarge by Amazon Web Services

hpc7a.48xlarge is a High performance computing [AMD processors] Gen7 48xlarge server offered by Amazon Web Services with 96 vCPUs, 768 GiB of memory and 0 GB of storage. The pricing starts at 7.2 USD per hour.
96 vCPU
768 GiB Memory
Spare SCore
294,474
(All-cores)
3,075
(Single-core)

Specifications

Server Metadata

Vendor ID
aws
Name
hpc7a.48xlarge
Description
High performance computing [AMD processors] Gen7 48xlarge
Family
hpc7a
Hw Virt
Status
active
Observed At
2026-10-03T23:20:03.547953Z

Availability

REGION / IDSPOTONDEMAND
Ohio (US) / us-east-2
- 7.2 USD/h
Stockholm (SE) / eu-north-1
- 7.7252 USD/h
Dublin (IE) / eu-west-1
- 7.7252 USD/h
Paris (FR) / eu-west-3
- 8.5553 USD/h

Processor

vCPUs
96
Hypervisor
nitro
CPU Allocation
Dedicated
CPU Cores
96
CPU Speed
3.7 GHz
CPU Architecture
x86_64
CPU Manufacturer
AMD
CPU Family
EPYC
CPU Model
9R14
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
96 MiB
CPU L3 Cache
32 MiB
CPU L3 Cache Total
768 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, 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, bmi1, avx2, smep, bmi2, invpcid, avx512f, avx512dq, rdseed, adx, smap, avx512ifma, clflushopt, clwb, avx512cd, sha_ni, avx512bw, avx512vl, xsaveopt, xsavec, xgetbv1, xsaves, avx512_bf16, clzero, xsaveerptr, rdpru, wbnoinvd, arat, avx512vbmi, pku, ospke, avx512_vbmi2, gfni, vaes, vpclmulqdq, avx512_vnni, avx512_bitalg, avx512_vpopcntdq, rdpid, flush_l1d
Ecpus
95.8
Scalability
99.79

System Resources and Accelerators

MEMORY
Memory Amount
768 GiB
Memory Amount Actual
768 GiB
Memory Generation
DDR5
Memory Speed
4800 Mhz
GPU
GPU Count
0
GPU Memory Min
0 MiB
GPU Memory Total
0 MiB
GPUs
    STORAGE
    Storage Size
    0 GB
    Storages
      NETWORK
      Network Speed Baseline
      150 Gbps
      Network Speed Max
      150 Gbps
      Network Storage Speed Baseline
      0.087 Gbps
      Network Storage Speed Max
      2.085 Gbps
      Inbound Traffic
      0 GB/month
      Outbound Traffic
      0 GB/month
      IPv4
      0

      Server Description

      A dedicated 96-core high-performance computing platform with high-speed DDR5 memory and 150 Gbps networking for demanding parallel workloads.

      Compute OptimizedMemory Optimized

      Amazon Web Services hpc7a.48xlarge is a high-performance computing server featuring 96 dedicated physical AMD EPYC 9R14 cores running at 3.7 GHz on the Nitro hypervisor. Equipped with 768.0 GB of high-speed 5 GHz DDR5 memory, this instance delivers 8.0 GB of RAM per core. It lacks local storage and GPUs but provides 150 Gbps of baseline network bandwidth. Benchmark results place the server in the top 10% for multi-core CPU performance, memory bandwidth, and multi-threaded decompression. It also demonstrates strong single-core capabilities and top-tier prompt processing for large language models. The hardware configuration is optimized for compute-intensive and memory-bound parallel workloads, such as software compilation, scientific simulations, and high-performance computing tasks, offering high resource density and qualitative cost efficiency for large-scale parallel processing.

      Economics

      Average Price per Region

      Prices per Zone

      Lowest Prices

      Performance

      Workload Profiles

      5.5Score
      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) 13,001,355 ops/sec 1,916,704 ops/sec 6.8 50.00% 50.00% +160.00%
      Redis RPS (pipeline=16, SET) 116,144,525 ops/sec 11,642,227 ops/sec 10 20.00% 20.00% +58.00%
      PassMark Memory Mark (composite) 2,908 2,417 1.2 10.00% 10.00% +1.80%
      Memory bandwidth (read, 16 MB ~ L3) 1,287,630 MB/sec 108,361 MB/sec 12 10.00% 10.00% +28.00%
      PassMark single-thread CPU 2,883 Mops/s 2,365 Mops/s 1.2 10.00% 10.00% +1.80%

      Memory Bandwidth

      Compression

      OpenSSL

      Geekbench Single-Core

      Score: 2,002

      Geekbench Multi-Core

      Score: 20,496

      Passmark CPU Scores

      BENCHMARKSCORE
      Mark
      		 117765
      Compression
      		 2523708
      Encryption
      		 125401
      Extended Instructions
      		 237643
      Floating Point Maths
      		 508791
      Integer Maths
      		 579148
      Physics
      		 32430
      Prime Numbers
      		 1871
      Single Threaded
      		 2883
      String Sorting
      		 359052

      Passmark Memory Scores

      BENCHMARKSCORE
      Memory Mark
      		 2908
      Database Operations
      		 32308
      Memory Latency
      		 76
      Memory Read Cached
      		 27836
      Memory Read Uncached
      		 27043
      Memory Write
      		 26355

      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

      INSTANCEvCPUsMEMORYGPUs
      hpc7a.12xlarge 24768 GiB0
      hpc7a.24xlarge 48768 GiB0
      hpc7a.96xlarge 192768 GiB0

      Similar Servers

      INSTANCEVENDORvCPUsMEMORYGPUs

      hpc7a.48xlarge FAQs