Dirac 2.5 Operations
Lead Research Organisation:
University of Edinburgh
Department Name: Sch of Physics and Astronomy
Abstract
Physicists across the astronomy, nuclear and particle physics communities are focussed
on understanding how the Universe works at a very fundamental level. The distance scales
with which they work vary by 50 orders of magnitude from the smallest distances probed
by experiments at the Large Hadron Collider, deep within the atomic
nucleus, to the largest scale galaxy clusters discovered out in space. The Science challenges,
however, are linked through questions such as: How did the Universe begin and how is it evolving?
and What are the fundamental constituents and fabric of the Universe and how do they interact?
Progress requires new astronomical observations and experimental data but also
new theoretical insights. Theoretical understanding comes increasingly from large-scale
computations that allow us to confront the consequences of our theories very accurately
with the data or allow us to interrogate the data in detail to extract information that has
impact on our theories. These computations test the fastest computers that we have and
push the boundaries of technology in this sector. They also provide an excellent
environment for training students in state-of-the-art techniques for code optimisation and
data mining and visualisation.
The DiRAC-2.5 project builds on the success of the DiRAC HPC facility and will provide the resources needed
to support cutting edge research during 2017 in all areas of science supported by STFC.
DiRAC-2.5 will provide maintain the existing DiRAC-2 services from April 2017, and also provide and increase in computational
resources at Durham, Cambridge and Leicester.
This grant will support the operation of the Edinburgh DiRAC services, which presently comprise
98384 operational computing cores serving around 80% of DiRAC computing cycles. The system is made up
from both the original 1.26PFlop/s DiRAC BlueGene/Q system and, following a recent transfer
to Edinburgh by STFC, six racks of the Hartree BlueJoule supercomputer.
The DiRAC project also will offer a team of three research software engineers who will help DiRAC researchers to ensure their scientific codes to extract
the best possible performance from the hardware components of the DiRAC clusters. These highly skilled programmers will
increase the effective computational power of the DiRAC facility during 2017.
on understanding how the Universe works at a very fundamental level. The distance scales
with which they work vary by 50 orders of magnitude from the smallest distances probed
by experiments at the Large Hadron Collider, deep within the atomic
nucleus, to the largest scale galaxy clusters discovered out in space. The Science challenges,
however, are linked through questions such as: How did the Universe begin and how is it evolving?
and What are the fundamental constituents and fabric of the Universe and how do they interact?
Progress requires new astronomical observations and experimental data but also
new theoretical insights. Theoretical understanding comes increasingly from large-scale
computations that allow us to confront the consequences of our theories very accurately
with the data or allow us to interrogate the data in detail to extract information that has
impact on our theories. These computations test the fastest computers that we have and
push the boundaries of technology in this sector. They also provide an excellent
environment for training students in state-of-the-art techniques for code optimisation and
data mining and visualisation.
The DiRAC-2.5 project builds on the success of the DiRAC HPC facility and will provide the resources needed
to support cutting edge research during 2017 in all areas of science supported by STFC.
DiRAC-2.5 will provide maintain the existing DiRAC-2 services from April 2017, and also provide and increase in computational
resources at Durham, Cambridge and Leicester.
This grant will support the operation of the Edinburgh DiRAC services, which presently comprise
98384 operational computing cores serving around 80% of DiRAC computing cycles. The system is made up
from both the original 1.26PFlop/s DiRAC BlueGene/Q system and, following a recent transfer
to Edinburgh by STFC, six racks of the Hartree BlueJoule supercomputer.
The DiRAC project also will offer a team of three research software engineers who will help DiRAC researchers to ensure their scientific codes to extract
the best possible performance from the hardware components of the DiRAC clusters. These highly skilled programmers will
increase the effective computational power of the DiRAC facility during 2017.
Planned Impact
The expected impact of the DiRAC 2.5 HPC facility is fully described in the attached pathways to impact document and includes:
1) Disseminating best practice in High Performance Computing software engineering throughout the theoretical Particle Physics, Astronomy and Nuclear physics communities in the UK as well as to industry partners.
2) Working on co-design projects with industry partners to improve future generations of hardware and software.
3) Development of new techniques in the area of High Performance Data Analytics which will benefit industry partners and researchers in other fields such as biomedicine, biology, engineering, economics and social science, and the natural environment who can use this new technology to improve research outcomes in their areas.
4) Share best practice on the design and operation of distributed HPC facilities with UK National e-Infrastructure partners.
5) Training of the next generation of research scientists of physical scientists to tackle problems effectively on state-of-the-art of High Performance Computing facilities. Such skills are much in demand from high-tech industry.
6) Engagement with the general public to promote interest in science, and to explain how our ability to solve complex problems using the latest computer technology leads to new scientific capabilities/insights. Engagement of this kind also naturally encourages the uptake of STEM subjects in schools.
1) Disseminating best practice in High Performance Computing software engineering throughout the theoretical Particle Physics, Astronomy and Nuclear physics communities in the UK as well as to industry partners.
2) Working on co-design projects with industry partners to improve future generations of hardware and software.
3) Development of new techniques in the area of High Performance Data Analytics which will benefit industry partners and researchers in other fields such as biomedicine, biology, engineering, economics and social science, and the natural environment who can use this new technology to improve research outcomes in their areas.
4) Share best practice on the design and operation of distributed HPC facilities with UK National e-Infrastructure partners.
5) Training of the next generation of research scientists of physical scientists to tackle problems effectively on state-of-the-art of High Performance Computing facilities. Such skills are much in demand from high-tech industry.
6) Engagement with the general public to promote interest in science, and to explain how our ability to solve complex problems using the latest computer technology leads to new scientific capabilities/insights. Engagement of this kind also naturally encourages the uptake of STEM subjects in schools.
Publications
Rizzuti F
(2023)
3D stellar evolution: hydrodynamic simulations of a complete burning phase in a massive star
in Monthly Notices of the Royal Astronomical Society
Iyer K
(2020)
The diversity and variability of star formation histories in models of galaxy evolution
in Monthly Notices of the Royal Astronomical Society
Owens A
(2022)
ExoMol line lists - XLVII. Rovibronic molecular line list of the calcium monohydroxide radical (CaOH)
in Monthly Notices of the Royal Astronomical Society
Wilkins S
(2022)
First Light and Reionisation Epoch Simulations (FLARES) - VI. The colour evolution of galaxies z = 5-15
in Monthly Notices of the Royal Astronomical Society
Grisdale K
(2019)
On the observed diversity of star formation efficiencies in Giant Molecular Clouds
in Monthly Notices of the Royal Astronomical Society
Sorini D
(2020)
simba: the average properties of the circumgalactic medium of 2 = z = 3 quasars are determined primarily by stellar feedback
in Monthly Notices of the Royal Astronomical Society
Katz H
(2022)
RAMSES-RTZ: non-equilibrium metal chemistry and cooling coupled to on-the-fly radiation hydrodynamics
in Monthly Notices of the Royal Astronomical Society
Ghosh S
(2022)
Age dissection of the vertical breathing motions in Gaia DR2: evidence for spiral driving
in Monthly Notices of the Royal Astronomical Society
Sawala T
(2021)
Setting the stage: structures from Gaussian random fields
in Monthly Notices of the Royal Astronomical Society
Altamura E
(2023)
EAGLE-like simulation models do not solve the entropy core problem in groups and clusters of galaxies
in Monthly Notices of the Royal Astronomical Society
Owens A
(2024)
ExoMol line lists - LI. Molecular line lists for lithium hydroxide (LiOH)
in Monthly Notices of the Royal Astronomical Society
Astoul A
(2022)
The effects of non-linearities on tidal flows in the convective envelopes of rotating stars and planets in exoplanetary systems
in Monthly Notices of the Royal Astronomical Society
Young A
(2021)
Chemical signatures of a warped protoplanetary disc
in Monthly Notices of the Royal Astronomical Society
Pfeifer S
(2020)
The bahamas project: effects of a running scalar spectral index on large-scale structure
in Monthly Notices of the Royal Astronomical Society
Benitez-Llambay A
(2020)
The detailed structure and the onset of galaxy formation in low-mass gaseous dark matter haloes
in Monthly Notices of the Royal Astronomical Society
Garzilli A
(2021)
How to constrain warm dark matter with the Lyman-a forest
in Monthly Notices of the Royal Astronomical Society
Bartlett D
(2021)
Spatially offset black holes in the Horizon-AGN simulation and comparison to observations
in Monthly Notices of the Royal Astronomical Society
Chachan Y
(2019)
Dust accretion in binary systems: implications for planets and transition discs
in Monthly Notices of the Royal Astronomical Society
Baugh C
(2020)
Sensitivity analysis of a galaxy formation model
in Monthly Notices of the Royal Astronomical Society
Borukhovetskaya A
(2022)
Galactic tides and the Crater II dwarf spheroidal: a challenge to LCDM?
in Monthly Notices of the Royal Astronomical Society
Brown S
(2022)
Towards a universal model for the density profiles of dark matter haloes
in Monthly Notices of the Royal Astronomical Society
Nobels F
(2022)
The interplay between AGN feedback and precipitation of the intracluster medium in simulations of galaxy groups and clusters
in Monthly Notices of the Royal Astronomical Society
Katsianis A
(2020)
The high-redshift SFR-M* relation is sensitive to the employed star formation rate and stellar mass indicators: towards addressing the tension between observations and simulations
in Monthly Notices of the Royal Astronomical Society
Barker A
(2019)
Angular momentum transport by the GSF instability: non-linear simulations at the equator
in Monthly Notices of the Royal Astronomical Society
Bosman S
(2022)
Hydrogen reionization ends by z = 5.3: Lyman-a optical depth measured by the XQR-30 sample
in Monthly Notices of the Royal Astronomical Society
Stiskalek R
(2022)
The scatter in the galaxy-halo connection: a machine learning analysis
in Monthly Notices of the Royal Astronomical Society
Theuns T
(2021)
Connecting cosmological accretion to strong Ly a absorbers
in Monthly Notices of the Royal Astronomical Society
Regan J
(2019)
Super-Eddington accretion and feedback from the first massive seed black holes
in Monthly Notices of the Royal Astronomical Society
Buzzo M
(2021)
Recovering the origins of the lenticular galaxy NGC 3115 using multiband imaging
in Monthly Notices of the Royal Astronomical Society
Stamatellos D
(2019)
ALMA reveals a pseudo-disc in a proto-brown dwarf
in Monthly Notices of the Royal Astronomical Society
Richardson M
(2020)
Simulating gas kinematic studies of high-redshift galaxies with the HARMONI integral field spectrograph
in Monthly Notices of the Royal Astronomical Society
Nightingale J
(2019)
Galaxy structure with strong gravitational lensing: decomposing the internal mass distribution of massive elliptical galaxies
in Monthly Notices of the Royal Astronomical Society
Young A
(2023)
On the conditions for warping and breaking protoplanetary discs
in Monthly Notices of the Royal Astronomical Society
Bose S
(2023)
The MillenniumTNG Project: the large-scale clustering of galaxies
in Monthly Notices of the Royal Astronomical Society
Wurster J
(2020)
Non-ideal magnetohydrodynamics versus turbulence II: Which is the dominant process in stellar core formation?
in Monthly Notices of the Royal Astronomical Society
Borukhovetskaya A
(2022)
The tidal evolution of the Fornax dwarf spheroidal and its globular clusters
in Monthly Notices of the Royal Astronomical Society
Kukstas E
(2020)
Environment from cross-correlations: connecting hot gas and the quenching of galaxies
in Monthly Notices of the Royal Astronomical Society
Appleby S
(2023)
Mapping circumgalactic medium observations to theory using machine learning
in Monthly Notices of the Royal Astronomical Society
Betts J
(2023)
Machine learning and structure formation in modified gravity
in Monthly Notices of the Royal Astronomical Society
Smith A
(2022)
A light-cone catalogue from the Millennium-XXL simulation: improved spatial interpolation and colour distributions for the DESI BGS
in Monthly Notices of the Royal Astronomical Society
McCarthy I
(2023)
The FLAMINGO project: revisiting the S 8 tension and the role of baryonic physics
in Monthly Notices of the Royal Astronomical Society
Kannan R
(2023)
The MillenniumTNG project: the galaxy population at z = 8
in Monthly Notices of the Royal Astronomical Society
Elsender D
(2023)
On the frequencies of circumbinary discs in protostellar systems
in Monthly Notices of the Royal Astronomical Society
Owens A
(2021)
ExoMol line lists - XLI. High-temperature molecular line lists for the alkali metal hydroxides KOH and NaOH
in Monthly Notices of the Royal Astronomical Society
Deason A
(2022)
Dwarf stellar haloes: a powerful probe of small-scale galaxy formation and the nature of dark matter
in Monthly Notices of the Royal Astronomical Society
Harvey D
(2019)
Observable tests of self-interacting dark matter in galaxy clusters: BCG wobbles in a constant density core
in Monthly Notices of the Royal Astronomical Society
Komissarov S
(2019)
Magnetic inhibition of centrifugal instability
in Monthly Notices of the Royal Astronomical Society
Zhang H
(2022)
Spherical accretion of collisional gas in modified gravity I: self-similar solutions and a new cosmological hydrodynamical code
in Monthly Notices of the Royal Astronomical Society
Anderson S
(2022)
The secular growth of bars revealed by flat (peak + shoulders) density profiles
in Monthly Notices of the Royal Astronomical Society
Thob A
(2019)
The relationship between the morphology and kinematics of galaxies and its dependence on dark matter halo structure in EAGLE
in Monthly Notices of the Royal Astronomical Society
Description | DiRAC 2.5 is a facility to support leading-edge computational astronomy and particle physics in the UK. This has resulted in over 500 peer-reviewed publications. |
Exploitation Route | Build on the scientific knowledge and computational techniques developed. |
Sectors | Digital/Communication/Information Technologies (including Software),Education |
Description | Intel IPAG QCD codesign project |
Organisation | Intel Corporation |
Department | Intel Corporation (Jones Farm) |
Country | United States |
Sector | Private |
PI Contribution | We have collaborated with Intel corporation since 2014 with $720k of total direct funding, starting initially as an Intel parallel computing centre, and expanding to direct close collaboration with Intel Pathfinding and Architecture Group. |
Collaborator Contribution | We have performed detailed optimisation of QCD codes (Wilson, Domain Wall, Staggered) on Intel many core architectures. We have investigated the memory system and interconnect performance, particularly on Intel's latest interconnect hardware called Omnipath. We found serious performance issues and worked with Intel to plan a solution and this has been verified and is available as beta software. It will reach general availability in the Intel MPI 2019 release, and allow threaded concurrent communications in MPI for the first time. A joint paper on the resolution to this was written with the Intel MPI team, and the application of the same QCD programming techniques to machine learning gradient reduction was applied in the paper to the Baidu Research all reduce library, demonstrating a 10x gain for this critical step in machine learning in clustered environments. We are also working with Intel verifying future architectures that will deliver the exascale performance in 2021. |
Impact | We have performed detailed optimisation of QCD codes (Wilson, Domain Wall, Staggered) on Intel many core architectures. We have investigated the memory system and interconnect performance, particularly on Intel's latest interconnect hardware called Omnipath. We found serious performance issues and worked with Intel to plan a solution and this has been verified and is available as beta software. It will reach general availability in the Intel MPI 2019 release, and allow threaded concurrent communications in MPI for the first time. A joint paper on the resolution to this was written with the Intel MPI team, and the application of the same QCD programming techniques to machine learning gradient reduction was applied in the paper to the Baidu Research all reduce library, demonstrating a 10x gain for this critical step in machine learning in clustered environments. This collaboration has been renewed annually in 2018, 2019, 2020. Two DiRAC RSE's were hired by Intel to work on the Turing collaboration. |
Start Year | 2016 |
Title | FP16-S7E8 MIXED PRECISION FOR DEEP LEARNING AND OTHER ALGORITHMS |
Description | We demonstrated that a new non-IEEE 16 bit floating point format is the optimal choice for machine learning training and proposed instructions. |
IP Reference | US20190042544 |
Protection | Patent application published |
Year Protection Granted | 2019 |
Licensed | Yes |
Impact | We demonstrated that a new non-IEEE 16 bit floating point format is the optimal choice for machine learning training and proposed instructions. Intel filed this with US patent office. This IP is owned by Intel under the terms of the Intel Turing strategic partnership contract. As a co-inventor I have been named on the patent application. The proposed format has been announced as planned for use in future Intel architectures. This collaboration with Turing emerged out of an investment in Edinburgh by Intel Pathfinding and Architecture Group in codesign with lattice gauge theory simulations. Intel hired DiRAC RSE's Kashyap and Lepper and placed them in Edinburgh to work with me on Machine Learning codesign through the Turing programme. |