Stay organized with collections
Save and categorize content based on your preferences.
Parallelstore is available by invitation only. If you'd like to request
access to Parallelstore in your Google Cloud project, contact your
sales representative.
Parallelstore is a fully managed, low-latency distributed file system
designed to meet the demands of high performance computing (HPC) and
data-intensive applications.
Parallelstore is ideal for use cases where multiple clients need concurrent
access to shared files with data integrity.
Parallelstore supports the POSIX standard, ensuring
compatibility with a wide range of existing applications and tools,
simplifying migration and integration.
Parallelstore instances can be mounted to Compute Engine VMs or
Google Kubernetes Engine clusters. The Parallelstore CSI driver enables
customers to use Kubernetes APIs to access the file system as volumes for
their stateful workloads.
Batch data transfers into and out of
Cloud Storage are available from the command line and the REST API.
Specifications
Parallelstore is a "scratch" file system: it's backed by local SSD with
2+1 erasure coding, with a mean time to data loss (MTTDL) from 2 to 16
months, depending on instance capacity. See the Performance
table for details.
Usable capacity can be configured from 12TiB to 100TiB.
These numbers are measured using 256 client connections to a single
instance. Latency is measured from a single client. Directory and file
striping settings are optimized for each metric.
Use Cases
High-performance computing: Parallelstore excels in HPC environments where
multiple compute nodes need fast and consistent access to shared data for
simulations, modeling, and analysis.
Machine learning: Parallelstore can handle the large datasets and high
throughput requirements of machine learning workloads, enabling efficient
training and inference.
[[["Easy to understand","easyToUnderstand","thumb-up"],["Solved my problem","solvedMyProblem","thumb-up"],["Other","otherUp","thumb-up"]],[["Hard to understand","hardToUnderstand","thumb-down"],["Incorrect information or sample code","incorrectInformationOrSampleCode","thumb-down"],["Missing the information/samples I need","missingTheInformationSamplesINeed","thumb-down"],["Other","otherDown","thumb-down"]],["Last updated 2025-08-25 UTC."],[],[],null,["# Parallelstore overview\n\nParallelstore is available by invitation only. If you'd like to request access to Parallelstore in your Google Cloud project, contact your sales representative.\n\nParallelstore is a fully managed, low-latency distributed file system\ndesigned to meet the demands of high performance computing (HPC) and\ndata-intensive applications.\n\nParallelstore is ideal for use cases where multiple clients need concurrent\naccess to shared files with data integrity.\n\nParallelstore supports the POSIX standard, ensuring\ncompatibility with a wide range of existing applications and tools,\nsimplifying migration and integration.\n\nParallelstore instances can be mounted to Compute Engine VMs or\nGoogle Kubernetes Engine clusters. The [Parallelstore CSI driver](/parallelstore/docs/csi-driver-overview) enables\ncustomers to use Kubernetes APIs to access the file system as volumes for\ntheir stateful workloads.\n\n[Batch data transfers](/parallelstore/docs/transfer-data) into and out of\nCloud Storage are available from the command line and the REST API.\n\nSpecifications\n--------------\n\n- Parallelstore is a \"scratch\" file system: it's backed by local SSD with\n 2+1 erasure coding, with a mean time to data loss (MTTDL) from 2 to 16\n months, depending on instance capacity. See the [Performance](#performance)\n table for details.\n\n- Usable capacity can be configured from 12TiB to 100TiB.\n\n- Supported in [multiple regions](/parallelstore/docs/locations).\n\nPerformance\n-----------\n\nExpected performance from Parallelstore is shown in the following table.\n\nThese numbers are measured using 256 client connections to a single\ninstance. Latency is measured from a single client. Directory and file\nstriping settings are optimized for each metric.\n\nUse Cases\n---------\n\n- **High-performance computing**: Parallelstore excels in HPC environments where\n multiple compute nodes need fast and consistent access to shared data for\n simulations, modeling, and analysis.\n\n- **Machine learning**: Parallelstore can handle the large datasets and high\n throughput requirements of machine learning workloads, enabling efficient\n training and inference.\n\nPricing\n-------\n\nSee the [Pricing](/parallelstore/pricing) page for details."]]
