The LBNL Superfacility Project Report

Bard, Deborah; Snavely, C.; Gerhardt, L.; Lee, Jason; Totzke, Becci; Antypas, Katie; Arndt, William B.; Blaschke, Johannes; Byna, Suren; Cheema, Ravi; Cholia, Shreyas; Day, Mark L.; Enders, Bjoern; Gaur, Aditi; Greiner, Annette; Groves, Taylor; Kiran, Mariam; Koziol, Quincey; Lehman, Tom; Rowland, Kelly; Samuel, Chris; Selvarajan, Ashwin; Sim, Alex; Skinner, David B.; Stephey, L.; Thomas, R. C.; Török, Gábor

doi:10.48550/arxiv.2206.11992

Cited by 2 publications

(4 citation statements)

References 0 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Data sets are expected to grow at least 3000-fold with increasing detector resolution, beam intensity, and measurement rate. 2,3,10 Therefore, the performance profiling data we collected represents an important benchmark, allowing us to extrapolate the overall performance of this Superfacility workflow and predict future bottlenecks which would prevent scaling.…”

Section: Hpc Challengesmentioning

confidence: 99%

“…At NERSC, together with SchedMD, we have developed a reservation system by which preemptible jobs can enter a reservation. 3,11 These will then be stopped if new jobs are submitted directly to the reservation (after a warning period during which SIGINT is used to request that the job saves and quits). This technology in its early stages, and we describe our initial experiences with preemptible reservations in Reference 11.…”

Section: Stochastic and Bursty Computementioning

confidence: 99%

“…By 2025 the next generation of XFEL experiments will more than double the detector resolution, and increase the rate at which measurements are taken by a factor of over 400× compared to existing facilities. 2,3 This will require computational intensity levels to escalate from petascale to exascale, for data analysis to keep pace with data collection.…”

mentioning

confidence: 99%

See 2 more Smart Citations

Real‐time XFEL data analysis at SLAC and NERSC: A trial run of nascent exascale experimental data analysis

Blaschke,

Brewster,

Paley

et al. 2024

Concurrency and Computation

View full text Add to dashboard Cite

SummaryX‐ray scattering experiments using free electron lasers (XFELs) are a powerful tool to determine the molecular structure and function of unknown samples (such as COVID‐19 viral proteins). XFEL experiments are a challenge to computing in two ways: (i) due to the high cost of running XFELs, a fast turnaround time from data acquisition to data analysis is essential to make informed decisions on experimental protocols; (ii) data‐collection rates are growing exponentially, requiring new scalable algorithms. Here we report our experiences analyzing data from two experiments at the Linac Coherent Light Source (LCLS) during September 2020. Raw data were analyzed on NERSC's Cori XC40 system, using the Superfacility paradigm: our workflow automatically moves raw data between LCLS and NERSC, where it is analyzed using the software package CCTBX. We achieved real time data analysis with a turnaround time from data acquisition to full molecular reconstruction in as little as 10 min—sufficient time for the experiment's operators to make informed decisions. By hosting the data analysis on Cori, and by automating LCLS‐NERSC interoperability, we achieved a data analysis rate which matches the data acquisition rate. Completing data analysis within 10 min is a first for XFEL experiments and an important milestone if we are to keep up with data‐collection trends.

show abstract

Section: Hpc Challengesmentioning

confidence: 99%

Section: Stochastic and Bursty Computementioning

confidence: 99%

See 1 more Smart Citation

Real‐time XFEL data analysis at SLAC and NERSC: A trial run of nascent exascale experimental data analysis

Blaschke,

Brewster,

Paley

et al. 2024

Concurrency and Computation

View full text Add to dashboard Cite

show abstract

“…Such applications have motivated the development of specialized interfaces for remote job submission [90,91] and for managing workloads across systems [92][93][94][95]. The LBNL superfacility project has studied requirements for linking instruments with HPC [96] and proposed an OAuth-based API [97] that is similar to our action provider interface. DataFed [98] federates various scientific data stores.…”

Section: Related Workmentioning

confidence: 99%