The Clairvoyant algorithm proposed in “A novel MPI reduction algorithm resilient to imbalances in process arrival times” was analyzed, commented and improved. The comments concern handling certain ...edge cases in the original pseudocode and description, i.e., adding another state of a process, improved cache friendliness more precise complexity estimations and some other issues improving the robustness of the algorithm implementation. The proposed improvements include skipping of idle loop rounds, simplifying generation of the ready set and management of the state array and an about 90-fold reduction in memory usage. Finally an extension enabling process arrival times (PATs) prediction was added: an additional background thread used to exchange the data with the PAT estimations. The performed tests, with a dedicated mini-benchmark executed in an HPC environment, showed correctness and improved performance of the solution, with comparison to the original or other state-of-the-art algorithms.
Abstract
Summary
The UNited RESisdue (UNRES) package for coarse-grained simulations, which has recently been optimized to treat large protein systems, has been implemented on Graphical Processor ...Units (GPUs). An over 100-time speed-up of the GPU code (run on an NVIDIA A100) with respect to the sequential code and an 8.5 speed-up with respect to the parallel Open Multi-Processing (OpenMP) code (run on 32 cores of 2 AMD EPYC 7313 Central Processor Units (CPUs)) has been achieved for large proteins (with size over 10 000 residues). Due to the averaging over the fine-grain degrees of freedom, 1 time unit of UNRES simulations is equivalent to about 1000 time units of laboratory time; therefore, millisecond time scale of large protein systems can be reached with the UNRES-GPU code.
Availability and implementation
The source code of UNRES-GPU along with the benchmarks used for tests is available at https://projects.task.gda.pl/eurohpcpl-public/unres.
The dynamics of the virus like particles (VLPs) corresponding to the GII.4 Houston, GII.2 SMV, and GI.1 Norwalk strains of human noroviruses (HuNoV) that cause gastroenteritis was investigated by ...means of long‐time (about 30 μs in the laboratory timescale) molecular dynamics simulations with the coarse‐grained UNRES force field. The main motion of VLP units turned out to be the bending at the junction between the P1 subdomain (that sits in the VLP shell) and the P2 subdomain (that protrudes outside) of the major VP1 protein, this resulting in a correlated wagging motion of the P2 subdomains with respect to the VLP surface. The fluctuations of the P2 subdomain were found to be more pronounced and the P2 domain made a greater angle with the normal to the VLP surface for the GII.2 strain, which could explain the inability of this strain to bind the histo‐blood group antigens (HBGAs).
Illustration of the dominant motion of the outer P2 domain of the VP1 protein of human norovirus (upper left), correlation between this domain motion, units clustered into two groups with intragroup‐correlated and intergroup‐anticorrelated motion (upper right), and extent of fluctuations (bottom), which is the largest for the GII.2 strain that does not bind to the histo‐blood group antigens, obtained by molecular dynamics simulations with the UNRES coarse‐grained model for three strains of human norovirus.
We report major algorithmic improvements of the UNRES package for physics‐based coarse‐grained simulations of proteins. These include (i) introduction of interaction lists to optimize computations, ...(ii) transforming the inertia matrix to a pentadiagonal form to reduce computing and memory requirements, (iii) removing explicit angles and dihedral angles from energy expressions and recoding the most time‐consuming energy/force terms to minimize the number of operations and to improve numerical stability, (iv) using OpenMP to parallelize those sections of the code for which distributed‐memory parallelization involves unfavorable computing/communication time ratio, and (v) careful memory management to minimize simultaneous access of distant memory sections. The new code enables us to run molecular dynamics simulations of protein systems with size exceeding 100,000 amino‐acid residues, reaching over 1 ns/day (1 μs/day in all‐atom timescale) with 24 cores for proteins of this size. Parallel performance of the code and comparison of its performance with that of AMBER, GROMACS and MARTINI 3 is presented.
Molecular dynamics with the optimized and efficiently parallelized implementation of the highly‐reduced physics‐based UNRES model of polypeptide chains enables us to reach, with moderate computer resources, several nanosecond/day of MD time, for protein systems with size over 100,000 residues, which translates to microsecond/day real time given a 1000 fold faster occurrence of events in the simulations with the UNRES model compared to the all‐atom representation.
In this paper a single machine time-dependent scheduling problem with total completion time criterion is considered. There are given
n jobs
J
1
,
…
,
J
n
and the processing time
p
i
of the
ith job is ...given by
p
i
=
a
+
b
i
s
i
, where
s
i
is the starting time of the
ith job
(
i
=
1
,
…
,
n
)
,
b
i
is its deterioration rate and
a is the common base processing time. If all jobs have deterioration rates different and not smaller than a certain constant
u
>
0
, then for each
ϵ
>
0
a solution with the value of the goal function that is at most
1
+
ϵ
times greater than the optimal one can be found. We give a FPTAS that finds such a solution in
O
n
1
+
6
log
1
+
u
2
1
ϵ
2
log
1
+
u
2
time. Consequently, the problem cannot be NP-hard in the strong sense.
A problem of scheduling deteriorating jobs on a single processor is considered. The processing time
of a job
is given by a function
, where
is the starting time of the job,
,
, for
Jobs are ...non-preemptive and independent and there are neither ready times nor deadlines. The goal is to minimize the total weighted completion time. We show how to employ the concept of non-dominated schedules to construct an exact algorithm for the problem. We also consider extending the algorithm to solve problems with precedence constraints and finding all Pareto-optimal solutions. Then we present how to use the concept to the problem
. We use elimination of dominated partial schedules to improve the efficiency of a branch-and-bound algorithm and present another algorithm, based solely on the elimination of dominated partial schedules.
Graphical Processor Units (GPUs) are nowadays widely used in all-atom molecular simulations because of the advantage of efficient partitioning of atom pairs between the kernels to compute the ...contributions to energy and forces, thus enabling the treatment of very large systems. Extension of time- and size-scale of computations is also sought through the development of coarse-grained (CG) models, in which atoms are merged into extended interaction sites. Implementation of CG codes on the GPUs, particularly the multiple-GPU platforms is, however, a challenge due to more complicated potentials and removing the explicit solvent, forcing developers to do interaction- rather than space-domain decomposition. In this paper, we propose a design of a multi-GPU coarse-grained simulator and report the implementation of the heavily coarse-grained physics-based UNited RESidue (UNRES) model of polypeptide chains. By moving all computations to GPUs and keeping the communication with CPUs to a minimum, we managed to achieve almost 5-fold speed-up with 8 A100 GPU accelerators for systems with over 200,000 amino-acid residues, this result making UNRES the best scalable coarse-grained software and enabling us to do laboratory-time millisecond-scale simulations of such cell components as tubulin within days of wall-clock time.
Program Title: Multi-GPU UNRES
CPC Library link to program files:https://doi.org/10.17632/hz9s4nwncf.1
Developer's repository link:https://projects.task.gda.pl/eurohpcpl-public/unres
Licensing provisions: GPLv3
Programming language: Fortran + C++/CUDA
Nature of problem: Physics-based simulations of protein systems at biologically relevant time- and size-scale are demanding and consequently require both the simplification of biomolecule representation and substantial computational resources. UNRES (from UNited RESidue) is a physics-based reduced model of polypeptide chains with which to run large-scale coarse-grained simulations of protein structure and dynamics. It enables the researchers to study protein folding, protein dynamics, and protein-protein interactions in a physically realistic manner and further unveil biological processes' mechanisms. Examples of biological applications include studies of amyloid formations, signaling mechanism, and action of molecular chaperones.
Solution method: The presented Multi-GPU UNRES relies on a highly optimized GPU implementation of non-central forces using modern CUDA constructs. Fundamentally, it is possible by proposed efficient partitioning and assignment of the interaction domain onto GPU resources. We moved as many computations as possible to the device (GPU) side. In most cases, computations are defined and scheduled as CUDA graphs. In selected cases, scheduling kernels manually yields slightly better performance. To maximize parallelism, multiple CUDA streams are used. Furthermore, the code visibly benefits from a tree-based allreduce shared-memory-based algorithm. Additionally, if present within hardware, peer memory access is enabled between all GPUs and the allreduce algorithm takes advantage of it. This feature has made the UNRES coarse-grained protein model with implicit solvent scalable for multi-GPUs so that we could achieve almost 5-fold speed-up with 8 A100 GPU accelerators for systems with over 200,000 amino-acid residues.
•UNRES coarse-grained protein model with implicit solvent made scalable for multi-GPUs.•Efficient partitioning and assignment of the interaction domain onto GPU resources.•Highly optimized GPU implementation of non-central forces using modern CUDA constructs.
In this paper a single machine time-dependent scheduling problem with total completion time criterion is considered. There are given n jobs J1,...,Jn and the processing time p(i) of the ith job is ...given by p(i)=a+b(i)s(i), where s(i) is the starting time of the ith job (i=1,...,n),b(i) is its deterioration rate and a is the common base processing time. If all jobs have deterioration rates different and not smaller than a certain constant u>0, then for each ... a solution with the value of the goal function that is at most ... times greater than the optimal one can be found. We give a FPTAS that finds such a solution in ... time. Consequently, the problem cannot be NP-hard in the strong sense. (ProQuest: ... denotes formulae/symbols omitted.)
KOALA GRAPH THEORY INTERNET SERVICE TOMASZ GOLUCH; KRZYSZTOF M. OCETKIEWICZ; KRZYSZTOF GIARO
T.A.S.K. quarterly,
10/2015, Letnik:
19, Številka:
4
Journal Article
Odprti dostop
KOALA has been created with the idea of C++ library templates, implementing a broad set of procedures in the fields of algorithmic graph theory and network problems in discrete optimization. During ...the C2 NIWA project, a library has been greatly extended, the code refactored and enclosed with the internet service available in the public repository of the project. Today it contains interconnected educational materials in the form of Wikibook, documentation and sample codes, a multifunctional web-based application for edition of graphs, a collection of over 100 web services which offers a library of selected procedures to be run on the BeesyCluster system.