| 목차 |
|---|
| 페이지 속성 | ||||||||
|---|---|---|---|---|---|---|---|---|
| 정보 | ||||||||
| ||||||||
Requirements are based on a user story, which is is an informal, natural language description of one or more features of a software system. User stories are often written from the perspective of an end user or user of a system. Depending on the community, user stories may be written by various stakeholders including clients, users, managers or development team members. They facilitate sensemaking and communication, that is, they help software teams organize their understanding of the system and its context. Please do not confuse user story with system requirements. A user story is an informal description of a feature; a requirement is a formal description of need (See section later). User stories may follow one of several formats or templates. The most common would be: "As a <role>, I want <capability> so that <receive benefit>" "In order to <receive benefit> as a <role>, I want <goal/desire>" "As <persona>, I want <what?> so that <why?>" where a persona is a fictional stakeholder (e.g. user). A persona may include a name, picture; characteristics, behaviours, attitudes, and a goal which the product should help them achieve. Example: “As provider of the Climate gateway I want to empower researchers from academia to interact with datasets stored in the Climate Catalogue, and bring their own applications to analyse this data on remote cloud servers offered via EGI.”
|
User stories
|
Ambition
The CC will enable ELIXIR to establish an ELIXIR Compute Platform (ECP) which allows ELIXIR cloud and data providers to share cloud compute and storage capacity to replicate and share reference datasets with each other and with their users. The platform aims to enable researchers to combine technical components of the ELIXIR Compute Platform services into a seamless ecosystem, thereby creating a science ready, standardised interface to the key resources and technological capabilities that are available for life sciences. The ECP aims to leverage the EOSC Service Catalogue to enable two related yet distinct activities for ELIXIR.
User stories
No. | User stories |
|---|---|
US1 | ELIXIR wants to establish a federation of cloud sites, each providing storage and compute capacity for researchers. The federated clouds should be connected to a data replication service (Reference Data Set Distribution Service with the ELIXIR terminology - RDSDS) that enables ELIXIR to stage 'ELIXIR Core Data Resources' to the cloud sites on-demand. As a result, the cloud sites become data hosting nodes which are equipped with CPUs/GPUs and are suited for large-scale data analysis and analytics. Centrally provided and curated datasets can ensure high-quality research in any of the partner states/regions. Researchers can go to their 'local' ELIXIR cloud provider, choose an already pre-staged ELIXIR dataset or request the staging of an ELIXIR dataset, choose an application of their choice (from a VM catalogue or container catalogue), maybe upload some additional data and then perform data analysis/analytics. Different conditions of access may apply at the different cloud sites, but it is expected that the cloud compute resources would be free at the point of use for national/local researchers, while pay-for-use or other special conditions apply for foreigners. The replication of community assets to national cloud providers maximises the utilisation of national funding and lowers the total cost of access for researchers. The services in the setup should recognise users via their community/national user accounts and user attributes. (i.e. integration with community and national AAI systems may be needed.)ELIXIR identity, therefore ELIXIR AAI (Life science AAI) should be integrated with the RDSDS as well as with the national clouds. |
US2 | The cloud federation can be also equipped with an ‘application a ‘container replication and orchestration service’ that enables application providers to deploy containerised community/reference applications to any of the federated cloud sites, further improving 'reproducibility in science'and users to instantiate and use the applications on those sites. Having a centrally managed or self-managed container orchestration service will allow users who do not have access to cloud resources of their own to deploy containerised workflows co-located with existing datasets in cloud locations. The services in the setup should recognise users via their ELIXIR identity, therefore ELIXIR AAI (Life science AAI) should be integrated with the RDSDS as well as with the national clouds. |
Use cases
| 정보 | ||
|---|---|---|
| ||
A use case is a list of actions or event steps typically defining the interactions between a role (known in the Unified Modeling Language as an actor) and a system to achieve a goal. Include in this section any diagrams that could facilitate the understanding of the use cases and their relationships. |
...
Step | Description of action | Dependency on 3rd party services (EOSC-hub or other) | ||
|---|---|---|---|---|
UC1 | Joining the cloud federation with a cloud site (cloud provider perspective):
| ELIXIR AAI ↔ EOSC AAI ELIXIR Cloud Services ↔ EOSC Compute / Storage Catalogue Policy compliance between ELIXIR and EOSC | ||
UC2 | Making reference/core datasets available for replication to the federated cloud providers (data provider perspective):
| Assumption: User is allowed to perform this action EOSC-hub centrally provided data distribution service (a new requirement to EOSC-hub!) EOSC-hub centrally provided application distribution and orchestration service (a new requirement to EOSC-hub) | UC2 | Making reference/core datasets available for replication to the federated cloud providers (data provider perspective): ELIXIR cloud federation policies, protocols and interfaces. (under definition) |
UC3 | Requesting the replication of a reference/core dataset to my local cloud (researcher perspective):
| Same as above. | ||
| UC4 | Making virtualised, reference/core applications available for replication to and orchestration on the federated cloud providers (data provider perspective):
| UC5 | Requesting the replication of a reference/core application to my local cloud (researcher perspective):
|
...
| Assumption: User is allowed to perform this action ELIXIR AAI ↔ EOSC AAI Kubernetes as a Service |
Requirements for EOSC-hub providers
Technical Requirements
Requirement |
|---|
EOSC-hub service
GAP (Yes/No) + description
number | Requirement title | Link to Requirement JIRA ticket | Source Use Case |
|---|---|---|---|
RQ1 | EOSC-hub to provide an FTS data transfer service |
Yes: EOSC-hub does not provide a centrally managed FTS service
| UC1, 2, 3 | ||||||||
RQ2 |
|---|
Kubernetes as a service
EOSC-hub |
to provide Kubernetes as a |
service |
| UC1, 4, 5 |
Capacity Requirements
The cloud capacity will initially come from the ELIXIR CC members (EBI, CESNET, CSC). Others will join in a second stage.
Capacity requirements for the centrally provided EOSC-hub FTS:
- X 100 concurrent users
- Y 100 TB throughput / month
- Simultaneous staging of data to Z 4 sites
Capacity requirements for the centrally provided Kubernetes as a service:
- X 100 concurrent users
- Y 500 container throughput / month
- Simultaneous staging of containers to Z sites4 sites
Validation plan
| # | Task | Description | Expected Outcome |
|---|---|---|---|
| UC1 | Integration of Cloud Provider with EOSC | ELIXIR service provider connects their service with EOSC | End-user is able to search and discover ELIXIR Compute and Data storage locations within EOSC |
| UC2 | Data Provider: Data Replication/Synchronisation | Data provider publishes data to RDSDS | Datasets are synchronised to multiple data storage location with ELIXIR and/or EOSC |
| UC3 | User: Data Transfer | User searches and initiates transfer for data to their local cloud | Dataset is made available to the user-spcified cloud location |
| UC4 | User: Container Orchestration Service | User initiates a containerised workload deployment | Users containerised workload was successfully deployed. |