The Australian SKA Regional Centre will enable science with SKA data in Australia.
The Australian-hosted SKA-low will produce around 300 PB per year of data that science teams around the globe will need to readily access.
SKA member states are forming a collaborative network of SKA Regional Centres (SRCs) to design, build, deliver, and operate end-to-end support for science data products, archives, and associated services.
SRCs will:
- Provide data flow and data dissemination solutions from the SKA to users;
- Store, publish and curate SKA data;
- Post-process and analyse SKA data products;
- Provide SKA data user support.
Once established, the SRCs will form a Regional Centre Alliance (RCA).
In 2019, the Department of Industry, Science, Energy and Resources (DISER), in coordination with the Australian and New Zealand SKA Coordination Committee (ANZSCC), granted $2 million to the University of Western Australia to organize a community based design study project for an Australian SKA Regional Centre. In partnership with CSIRO (providing additional funding of $1.8m) and Curtin University, with in kind support from ICRAR and the Pawsey Centre, the AusSRC design study project was launched in 2019 with the aims of
- Developing the requirements and design for an operational AusSRC
- Working and coordinating with regional and international SRC partners
- Supporting science projects on ASKAP and MWA to refine requirements and designs
- Submitting a proposal to the Commonwealth for a full scale AusSRC
In February 2021, the AusSRC team (headed by the AusSRC Director Dr. Karen Lee-Waddell) submitted a proposal for AusSRC funding as part of the overall DISER Federal budget submission covering Australia’s contribution to the SKA project. This proposal was successful and will deliver $63 million to build, operate and maintain an Australian SRC in the 10 year period to 2030/31.
The overall mission of the AusSRC is to deliver the ability to conduct science with current and next generation radio telescopes. This intent can be summarised by three key goals:
- Provide the Australian SKA community with the hardware, software, networking, and human resources necessary to access SKA and precursor data products
- Enable the production of advanced SKA data products and the development of analysis tools for all SKA users, and
- Build and play a leading role in a global SKA Science Archive Facility enabled by an international network of SKA Regional Centres (SRCs).
An Australian SRC will be developed over the next 10 years in a manner consistent with the SRC White Paper and in coordination with the international SRC Network effort. The AusSRC proposal identified resources that would constitute approximately 14% of the global SRC Network capacity needed for the anticipated data flow of SKA1. This is consistent with Australia’s 14% share of the SKA1 construction and operations budget. The AusSRC will also provide for resources and personnel to contribute to
- Expanding the ability of the Australian SKA community to define forefront and large-scale science programs that are not possible within the nominal 100% SRC Network. These large-scale programs will enhance our ability to lead and innovate in radio astronomy.
- Enabling engagement with science and research communities outside of the SKA (e.g. multi-wavelength and multi-messenger astronomy, fundamental physics, data-intensive sensor networks, and resource exploration), providing the AusSRC infrastructure to new communities for a broader range of research returns
- Translation and impact engagement with industry to co-develop and deliver data and processing intensive innovations enabled by exposure to exascale data sets
- Education and outreach, training and community engagement surrounding big data, machine learning, and data driven discovery
The AusSRC will be built and operated by a partnership of Australian organizations growing from the relationships put in place by the AusSRC design study. It will be a new operational organization that consists of dedicated staff and facilities distributed across a number of sites. The Federal funding agreement, governance and partnership agreements, detailed implementation plan and development timeline (in coordination with international efforts) will be developed over the next 18 months with the aim of having a new AusSRC organization in place by the end of 2022 that will grow to full SKA1 capacity by 2030/31.
AusSRC Design Study Program
The AusSRC Design Study Program is $4m program, funded by the Australian Government and CSIRO. The program is lead by ICRAR and CSIRO in collaboration with the Pawsey Centre, the MWA, the Shanghai Astronomical Observatory (ERIDANUS Project), AAL, and the international SRC Steering Committee. The program will define an AusSRC design and costing based on requirements and experiences gathered from the Australian and regional communities and the SKA precursors. The program will initially run from 2019 to 2022 and produce a submission to the Federal Cabinet for Stage 1 operational funding. Stage 1 (2022-2027) will focus on the operational support for science with SKA precursors. Stage 1 will also engage in SKA-LOW commissioning and early science from SKA-LOW. A Stage 2 proposal and deployment of AusSRC would take place in the 2028-30 period.
The program will utilise a top-down analysis of SRC requirements in global collaboration with other SRCs and the SKAO, as well as a bottom-up approach solving practical computational and data problems within the SKA precursor projects, leading to the design and prototyping of the architecture of the future Australian SRC.
A significant body of work is required to develop the SRC concept. The program will use a top-down analysis of SRC requirements in global collaboration with other SRCs and the SKAO, and a bottom-up approach solving practical computational and data problems within the SKA precursor projects, leading to the design and prototyping of the architecture of the future Australian SRC.
The AusSRC Design Study Program will identify, assess, and test potential solutions for providing the tools, services, and people needed for the Australian (and regional partner) research communities to utilise SKA data outputs. When completed, this is expected to bring together a range of resources, potentially including Australia’s HPC network (including the Pawsey Centre); other HPC capabilities; commercial cloud services; AARNet; and overseas facilities.
Program Key Activities
- Develop AusSRC requirements based on inputs from the SKA, international SRC development, and Australian SKA Precursors;
- Develop a technical architecture of the AusSRC, prototype its key elements, and develop an implementation plan;
- Execute a range of projects focused on supporting and improving the science outputs of ASKAP and MWA in order to enable the development of precursor driven requirements, and assist the precursors with their specific needs;
- Participate in development and verification of requirements for international SRCs, particularly through prototyping and deploying elements of infrastructure between AusSRC and ChinaSRC;
- Participate in the SKA and SRC data challenges;
- Develop and prototype the architecture of the SKA Science Archive in AusSRC and its API for post-processing;
- Prototype solutions for visual data analysis of SKA-scale data and test within the ASKAP and MWA science projects;
- Develop and present the Australian Government with the AusSRC Business Case/Plan;
- Establish links and collaborations with the international effort to develop the SKA Regional Centres, participate in established SRCSC working groups;
- Establish links, develop strategies and collaborative projects with Australian ICT companies that can assist with achieving the above goals;
- Establish links to other relevant projects such as AENEAS, OzGrav, ASVO, WLCG (computing for LHC) etc.
Program Components
SRC Architecture and Prototyping
AusSRC will be a significant investment. The AusSRC Design Program will collect in depth requirements working with ASKAP and MWA, as well as the SKA and other SRCs. These requirements will then drive the design of the System Architecture.
The three system domains—Science Archive, Post-processing, and Visualisation—will inform the System Architecture design to enable interactions between the components via various interfaces and communication protocols.
The System Architecture will drive the requirements on the Infrastructure domain, which will be prototyped considering forthcoming and prospective technologies.
The Applications and Algorithms domain will use external interfaces and protocols to interact with the system components to do specific user defined tasks.
Finally, the Operations and User Support domain will define how AusSRC systems are to be maintained and operated, and what support to provide users.
SKA Precursor Projects
While the key objectives of the program are focused on enabling science with SKA data in Australia, the proposed approach is to use the most challenging (from the point of view of data and computing) ASKAP and MWA science cases to inform AusSRC design. This approach is also targeted at creating tangible outcomes for ASKAP and MWA through addressing the challenges in post-processing, storing, visualising, and analysing Precursor data. Participation in the ASKAP and MWA projects will allow the AusSRC Design Program (DSP) team to develop transferable skills and expertise for the future AusSRC.
AusSRC DSP personnel employed through partner organisations are deployed by the Program within the ASKAP and MWA teams to assist with specific. The Program Leader deploys personnel on tasks determined by the project/task skill-set required.
EIRDANUS Project
The ERIDANUS Project is a three year design study between (and within) Australia and China from April 2017. ERIDANUS is aimed at deploying prototype data intensive research infrastructure and middleware capable of addressing SKA-class data and processing challenges. Read more at ERIDANUS.
SKA Regional Centres
Planning for a network of SRCs
The resources needed to fully process, distribute, curate and utilize data flowing from the SKA are currently beyond the scope of the SKA1 construction and operations budget. As previously experienced by the Large Hadron Collider (LHC) project, the SKAO and the international SKA science community will need to work collaboratively to shape and establish a shared, distributed data, computing and networking capability that draws on international cooperation and supports the broad spectrum of SKA science. This distributed and shared capability needs to be persistent and needs to utilize a variety of sources of funding from multiple governments and stakeholders. It needs to also be coherent and logistically centralised in terms of the supported services and shared resources that enable a coordinated functionality in support of both the SKA science community and the SKAO.
The first meeting of the SRCSC occurred in May 2019.
Peter Quinn (Chair, Australia), Tao An (China), Domingos Barbosa (Portugal), Rosie Bolton (SKAO), Antonio Chrysostomou (SKAO), John Conway (Sweden), Séverin Gaudet (Canada), Michiel van Haarlem (Deputy Chair, Netherlands), Hans-Rainer Klockner (Germany), Andrea Possenti (Italy), Bradley Frank (RSA), Jeremy Yates (UK), Lourdes Verdes-Montenegro (Spain), Jean-Pierre Vilotte (France), Yogesh Wadadekar (India), Emma Tolley (Switzedland), Takuya Akahori (Japan, observer)).
In May 2020 the SRCSC released an SRC White Paper as the starting point for the formation of an operational SRC partnership and collective set of SRC services. The collection of both services and infrastructure that constitute a global SRC capability is referred to as the “SRC Network”.
Recognising this reality, the SKAO Data Flow Advisory Panel recommended in March 2016 that the SKAO Board encourage SKA member states to form “a collaborative network of SKA Regional Centres (SRCs) to provide the essential functions that are not presently provided within the scope of the SKA1 project”. Following the endorsement of this recommendation by the Board, the SKAO formed the SRC Coordination Group (SRCCG – first meeting 23/9/16), with representatives from SKA member states (represented regionally) with advisors from the Vera C. Rubin Observatory and LHC projects. The SRCCG began the process of defining some of the basic requirements and challenges for a system of SRCs and worked with national communities as they sought to create proto-SRC projects. By the end of 2018, proto-SRC design and development projects were in advanced stages of planning and initiation across 13 SKA member states. This evolution from conceptualisation to implementation required a parallel evolution of the SRCCG to a new body that could take this work forward on a global basis. In November 2018, the SKA Board approved the formation of the SRC Steering Committee (SRCSC). The mission of the SRCSC is to:
“Guide the definition and creation of a long-term operational partnership between the SKA Observatory and an ensemble of independently-resourced SKA Regional Centres”
The SRCSC has created seven working groups, involving proto-SRC project teams and community members from across the SKA member states.
- WG0: SRC Network Architecture
- WG1: Data Logistics
- WG2: Operations
- WG3: Software: Federated Computing and Data Services
- WG4: Software: Science Archive, VO, FAIR
- WG5: Compute infrastructure
- WG6: Science User Engagement
By early 2023, through the activities of the seven working groups, the SRCSC plans to have developed an SRC Network architecture and implementation plan that will lead to a coordinated development and roll out of the full SRC Network by 2030. The ambition is to have 80% of the functionality and 10% of the full SKA1 capacity in place by 2025 (Phase 1) with full operational capacity and functionality by 2030 (Phase 2). The goal is to achieve the Phase 1 target by addressing the science cases and community needs surrounding the SKA precursors and pathfinders.