Search for: All records

A burst buffer is a common method to bridge the performance gap between the I/O needs of modern supercomputing applications and the performance of the shared file system on large-scale supercomputers. However, existing I/O sharing methods require resource isolation, offline profiling, or repeated execution that significantly limit the utilization and applicability of these systems. Here we present ThemisIO, a policy-driven I/O sharing framework for a remote-shared burst buffer: a dedicated group of I/O nodes, each with a local storage device. ThemisIO preserves high utilization by implementing opportunity fairness so that it can reallocate unused I/O resources to other applications. ThemisIO accurately and efficiently allocates I/O cycles among applications, purely based on real-time I/O behavior without requiring user-supplied information or offline-profiled application characteristics. ThemisIO supports a variety of fair sharing policies, such as user-fair, size-fair, as well as composite policies, e.g., group-then-user-fair. All these features are enabled by its statistical token design. ThemisIO can alter the execution order of incoming I/O requests based on assigned tokens to precisely balance I/O cycles between applications via time slicing, thereby enforcing processing isolation. Experiments using I/O benchmarks show that ThemisIO sustains 13.5--13.7% higher I/O throughput and 19.5--40.4% lower performance variation than existing algorithms. For real applications, ThemisIO significantly reduces the slowdown by 59.1--99.8% caused by I/O interference. 

Cellulose, the main component of the plant cell wall, is synthesized by the multimeric cellulose synthase (CESA) complex (CSC). In plant cells, CSCs are assembled in the endoplasmic reticulum or Golgi and transported through the endomembrane system to the plasma membrane (PM). However, how CESA catalytic activity or conserved motifs around the catalytic core influence vesicle trafficking or protein dynamics is not well understood. Here, we used yellow fluorescent protein (YFP)-tagged AtCESA6 and created 18 mutants in key motifs of the catalytic domain to analyze how they affected seedling growth, cellulose biosynthesis, complex formation, and CSC dynamics and trafficking in Arabidopsis thaliana. Seedling growth and cellulose content were reduced by nearly all mutations. Moreover, mutations in most conserved motifs slowed CSC movement in the PM as well as delivery of CSCs to the PM. Interestingly, mutations in the DDG and QXXRW motifs affected YFP-CESA6 abundance in the Golgi. These mutations also perturbed post-Golgi trafficking of CSCs. The 18 mutations were divided into 2 groups based on their phenotypes; we propose that Group I mutations cause CSC trafficking defects, whereas Group II mutations, especially in the QXXRW motif, affect protein folding and/or CSC rosette formation. Collectively, our results demonstrate that the CESA6 catalytic domain is essential for cellulose biosynthesis as well as CSC formation, protein folding and dynamics, and vesicle trafficking.

 

Abstract Stratospheric aerosol geoengineering has been proposed as a potential solution to reduce climate change and its impacts. Here, we explore the responses of the Hadley circulation (HC) intensity and the intertropical convergence zone (ITCZ) using the strategic stratospheric aerosol geoengineering, in which sulfur dioxide was injected into the stratosphere at four different locations to maintain the global-mean surface temperature and the interhemispheric and equator-to-pole temperature gradients at present-day values (baseline). Simulations show that, relative to the baseline, strategic stratospheric aerosol geoengineering generally maintains northern winter December–January–February (DJF) HC intensity under RCP8.5, while it overcompensates for the greenhouse gas (GHG)-forced southern winter June–July–August (JJA) HC intensity increase, producing a 3.5 ± 0.4% weakening. The residual change of southern HC intensity in JJA is mainly associated with stratospheric heating and tropospheric temperature response due to enhanced stratospheric aerosol concentrations. Geoengineering overcompensates for the GHG-driven northward ITCZ shifts, producing 0.7° ± 0.1° and 0.2° ± 0.1° latitude southward migrations in JJA and DJF, respectively relative to the baseline. These migrations are affected by tropical interhemispheric temperature differences both at the surface and in the free troposphere. Further strategies for reducing the residual change of HC intensity and ITCZ shifts under stratospheric aerosol geoengineering could involve minimizing stratospheric heating and restoring and preserving the present-day tropical tropospheric interhemispheric temperature differences. 

It has been reported that the sea level falls in the tropical Southwest Indian Ocean (SWIO) from the 1960s to the early 2000s. However, a rising trend of 4.05 ± 0.56 cm/decade has occurred during the recent two decades with our analysis showing that manometric sea level contributes 41% to this sea level rise. 30% of this rise is due to steric sea level (SSL) change in the upper 2,000 m with SSL rise in the upper 300 m of secondary importance. Conversely, thermal expansion below the thermocline (300–2,000 m), likely caused by water mass spread from the Southern Ocean, induces major contribution to SSL changes. Compared to existing studies demonstrating the contribution of thermal variations above the thermocline to sea level variability in the tropical SWIO, this study emphasizes the importance of ocean mass and deeper ocean changes in a warming climate.

 

Scientific Machine Learning (SciML) is a new multidisciplinary methodology that combines the data-driven machine learning models and the principle-based computational models to improve the simulations of scientific phenomenon and uncover new scientific rules from existing measurements. This article reveals the experience of using the SciML method to discover the nonlinear dynamics that may be hard to model or be unknown in the real-world scenario. The SciML method solves the traditional principle-based differential equations by integrating a neural network to accurately model the nonlinear dynamics while respecting the scientific constraints and principles. The paper discusses the latest SciML models and apply them to the oscillator simulations and experiment. Besides better capacity to simulate, and match with the observation, the results also demonstrate a successful discovery of the hidden physics in the pendulum dynamics using SciML. 

The stability and resilience of the Earth system and human well-being are inseparably linked^1–3, yet their interdependencies are generally under-recognized; consequently, they are often treated independently^4,5. Here, we use modelling and literature assessment to quantify safe and just Earth system boundaries (ESBs) for climate, the biosphere, water and nutrient cycles, and aerosols at global and subglobal scales. We propose ESBs for maintaining the resilience and stability of the Earth system (safe ESBs) and minimizing exposure to significant harm to humans from Earth system change (a necessary but not sufficient condition for justice)⁴. The stricter of the safe or just boundaries sets the integrated safe and just ESB. Our findings show that justice considerations constrain the integrated ESBs more than safety considerations for climate and atmospheric aerosol loading. Seven of eight globally quantified safe and just ESBs and at least two regional safe and just ESBs in over half of global land area are already exceeded. We propose that our assessment provides a quantitative foundation for safeguarding the global commons for all people now and into the future.