Elevating science with high-performance data analytics.
NIRD Data Peak is a high-performance storage service tailored for data-intensive workloads and scientific investigations. It accelerates data-driven discoveries by providing a platform optimised for pre/post-processing, intensive I/O operations, supporting high-performance computing, and accommodating AI workloads.
Data integrity is safeguarded by daily snapshots, built-in redundancy, error-correcting mechanisms, and user-controlled backups to protect valuable datasets from human errors and disruptions.
Service descripton
NIRD Data peak is specifically designed for data used actively. Please refer to our Data Policy for more information on data classification.
The technology of the underlying storage solution is designed with both high-performance and cost-effectiveness in mind, to allow intensive I/O operations and to support high-performance computing and AI workloads. This is especially important if you want to:
- Pre- / post-process data
- Analyse and visualise data in place or with the computing capacity of the national HPC systems
- Train AI models
NIRD Data Peak is specifically suited for you who need:
- High-performance storage for any type of active research data
- High-performance storage for I/O intensive computing such as HPC and AI workloads
- Storage of valuable research data requiring a secondary copy
- Storage for structured data of any volume larger than 1 TiB
Features
NIRD Data Peak supports multitenancy and offers file storage which can be scaled on demand up to petabyte-volume sizes. Therefore, you can invite your collaborators to store, and work in the same shared area, and consume the data with the services of your choice.
Additionally, you can deploy your applications and services on the NIRD Service Platform. This enables you to not only access your data on NIRD through your own services but also to grant access to other users.
Intefaces
Datasets stored on NIRD Data Peak can be accessed simultaneously from a terminal and over network file-share (NFS). Data is accessible directly from the NIRD login nodes and Sigma2´s national HPC systems.
Data integrity
Data stored on the NIRD Data Peak have daily snapshots and the in-built redundancy and error correcting mechanism ensures the integrity of the data and protection from human mistakes.
Valuable data stored on NIRD Data Peak can also have a backup to secure an extra layer of protection against data corruption, human mistakes, or hardware failures. Projects using the NIRD Data Peak service can enable this by opting in for NIRD Backup.
How to get access?
The most convenient way to access the resources is by applying through our regular calls, which are published twice a year.
However, it is also possible to apply outside calls at any point in time through the year.
Your individual application is evaluated based on scientific merit by a resource allocation committee, who grants you project access to resources if the application is accepted.
Alternatively, you can reserve resources by procurement for your group and your institution. You are welcome to get in touch with us if you want to explore this model.
Research activities
Below you will find a selection of research projects that have utilised the NIRD Data Peak service.
You might also need
We offer many services that you may need in addition to NIRD Data Peak. In the National Infrastructure for Research Data (NIRD) ecosystem we offer a range of storage services designed to support scientific research in every step of the research data lifecycle. Below, you can see a few selected ones, and if you visit our services overview, you will find all we have on offer.
The NIRD Data Peak service going forward
Take a look at the roadmap for the development of the NIRD Data Peak Service to follow the progress of the service: