Adding MUSCLE to Multiscale Simulations

By Joris Borgdorff, Derek Groen, and Mariusz Mamonski

December 11, 2013

Multiscale models help understand phenomena with a wider scope or an increased level of detail. These models allow us to take the best from multiple worlds, for example by combining models with a fine-grained time or space resolution with models that capture systems over a large baseline.

Classical examples of multiscale modeling include coupling atomistic models to coarse-grained models, where several atoms are represented instead as a single fused particle, or coupling fine fluid dynamics models to coarser structural mechanics models. In both cases, we have a fairly good understanding of the single scale phenomena, but still have little knowledge of the interactions between these phenomena. As understanding these interactions is key to understanding these phenomena as a whole, many researchers are now actively developing and using multiscale modeling techniques [1].

Researchers from different disciplines recognized their common need for a general multiscale computing approach, and started the European e-Infrastructure MAPPER project in 2010. The project aimed to bring their demanding multiscale applications to HPC, in an approach that uses the commonalities in multiscale modeling for applications in biomedicine, hydrology, nanomaterials, fusion, and systems biology.

muscle5The project settled on the theoretical and components-based Multiscale Modeling and Simulation Framework [2], which defines a multiscale model as a set of coupled single scale models (see Fig. 1). This approach allows code reuse, since single scale models often already exist, and defines clear opportunities for scheduling and distributing multiscale models, since the coupling is separated from the single scale code. Single scale models (or submodels) are implemented, verified, and validated in isolation, after which their interaction is added. They interact through input and output ports, sending or receiving simple parameters there, or entire datasets or geometries. A conduit transports the data from one port to another and intermediate components transform the data, implementing so-called scale-bridging techniques.

The Multiscale Coupling Library and Environment 2 (MUSCLE 2, http://apps.man.poznan.pl/trac/muscle) was created to implement and execute multiscale models with feedback loops, which we call cyclic coupling topologies. MUSCLE is truly a domain-independent approach, as it has so far been adopted, amongst others, by the abovementioned applications in MAPPER, and run on several supercomputers and clusters in Europe, as well as the Amazon cloud infrastructure. It consists of a library,  scripted coupling and a runtime environment.

Figure 2. Layered design of MUSCLE 2, separating implementation, coupling and execution.
Figure 2. Layered design of MUSCLE 2, separating implementation, coupling and execution.

By design, submodels do their computations independently, and MUSCLE 2 allows them to be implemented in different programming languages (C, C++, Fortran, Java, Scala, Python, or MATLAB) and for them to run on multiple machines. Conduits in MUSCLE 2 use shared memory communication where possible, and TCP/IP communication otherwise.  Using TCP/IP makes communicating between different computers as transparent as running on a single host. A central simulation manager acts as a white-page service for the submodels in a simulation, but after a submodel is registered there it does decentralized message passing with other submodels to prevent possible bottlenecks. By running each submodel in a separate process or thread, MUSCLE 2 has inherent multiscale parallelism.

MUSCLE 2 separates the submodel code, which just knows about input and output ports, from coupling code, which knows which ports will be coupled. This allows users to change the coupling topology without recompiling or redeploying code. Additionally, the coupling code is independent from the resources that the simulation will be eventually run on, so the same coupling can be submitted to multiple machines or be spread out over them, even when they reside in different countries. This allows us not only to use more resources but, more importantly, to take advantage of architectures that are optimal for each of the models involved. For example, some models in a simulation may greatly benefit from GPGPUs, whereas others have large memory requirements.

Within a supercomputer, MUSCLE 2 can make direct connections between processes, but almost all supercomputers have firewalls in place that prevent direct connections between worker nodes in different supercomputers. In MUSCLE 2we resolve this by providing a TCP/IP forwarding service, the MUSCLE Transport Overlay, that runs on interactive nodes of the cluster and forwards messages between MUSCLE 2 installations on different clusters.

To optimize the speed for large messages, MUSCLE 2 has the option of using the MPWide library (http://www.github.com/djgroen/MPWide), a high-performance communication library that was used to cosmological simulations in parallelized across multiple supercomputers [3]. MUSCLE 2 processes can be run directly as a supercomputer job and have few dependencies, but require the address of the site simulation manager to connect to other processes. This can be done manually or automated via a dedicated service such as provided by the QosCosGrid software stack (http://www.qoscosgrid.org/). QosCosGrid provides middleware solutions, notably on the Polish national grid (PL-Grid), that allow users to schedule and coordinate simulations running in multiple distributed HPC resources [4]. MUSCLE 2 is open source software (LGPL version 3 license) that runs on Linux and OS X and can be freely installed without administrative privileges.

In memoriam Mariusz Mamonski (1984 – 2013)

We wish to dedicate this paper to the memory of Mariusz Mamonski, whose sudden decease came as a shock to us all. The MUSCLE 2 team would like to thank Mariusz for all his professional and personal contributions to distributed multiscale computing. His dedication to end-users, his insight in software quality and his experience with infrastructures was truly impressive, and he will be sorely missed.

Biographies:

muscle3Joris Borgdorff is a PhD candidate at the Computational Science group of the University of Amsterdam, researching the formal background of multiscale and complex systems modeling and the applied aspects of distributed multiscale computing. He received a BSc in Mathematics and in Computer Science (2006) and an MSc in Applied Computing Science (2009) from Utrecht University. He is involved in the European MAPPER and Sophocles projects.

 

muscle2Derek Groen is a post-doctoral researcher at the Centre for Computational Science at University College London and a Fellow of the Software Sustainability Institute. He has expertise in high performance and distributed computing, as well as multiscale simulation. Derek has worked on a range of applications, and modelled star clusters, cosmological dark matter structures, clay-polymer nanocomposite materials, turbulence and human bloodflow using supercomputers. He obtained his PhD in Computational Astrophysics in Amsterdam in 2010.

 

muscle1Mariusz Mamonski (1984 – 2013) received his diploma in Computer Science at the Poznan University of Technology (Laboratory of Computing Systems) in 2008. He started working at the Application Department of the Poznan Supercomputing and Networking Center in 2005. He contributed to several research EU projects, in particular: GridLab, InteliGrid, BREIN and QosCosGrid, and was involved in the national and European e-Infrastracture projects PL-Grid and MAPPER. His research primarily focussed on web services, queueing systems and parallel execution and programing environments. He was an active member of the Open Grid Forum Distributed Resource Management Application API (OGF DRMAA) working group.

Acknowledgements

We would like to thank the Bartosz Bosak and Krzysztof Kurowski from the Poznan Supercomputing and Networking Center for their support and input, and we thank Alfons G. Hoekstra from the University of Amsterdam for his feedback.

References:

[1] Groen et al., Survey of Multiscale and Multiphysics Applications and Communities, IEEE Computing in Science and Engineering, http://dx.doi.org/10.1109/MCSE.2013.47.

[2] Borgdorff et al., Foundations of distributed multiscale computing: Formalization, specification, and analysis, Journal of Parallel and Distributed Computing http://dx.doi.org/10.1016/j.jpdc.2012.12.011

[3] Groen et al., A lightweight communication library for distributed computing, accepted by the Journal of Open Research Software, http://arxiv.org/abs/1312.0910.

[4] Kravtsov et al., Grid-enabling complex system applications with QosCosGrid: An architectural perspective, Proceedings of the 2008 International Conference on Grid Computing & Applications, Las Vegas, Nevada, USA, 2008, pp. 168–174.

[5] Borgdorff et al., Distributed Multiscale Computing with MUSCLE 2, the Multiscale Coupling Library and Environment, submitted to the Journal of Computational Science, http://arxiv.org/abs/1311.5740

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industy updates delivered to you every week!

Weekly Twitter Roundup (Feb. 23, 2017)

February 23, 2017

Here at HPCwire, we aim to keep the HPC community apprised of the most relevant and interesting news items that get tweeted throughout the week. Read more…

By Thomas Ayres

HPE Server Shows Low Latency on STAC-N1 Test

February 22, 2017

The performance of trade and match servers can be a critical differentiator for financial trading houses. Read more…

By John Russell

HPC Financial Update (Feb. 2017)

February 22, 2017

In this recurring feature, we’ll provide you with financial highlights from companies in the HPC industry. Check back in regularly for an updated list with the most pertinent fiscal information. Read more…

By Thomas Ayres

Rethinking HPC Platforms for ‘Second Gen’ Applications

February 22, 2017

Just what constitutes HPC and how best to support it is a keen topic currently. Read more…

By John Russell

HPE Extreme Performance Solutions

O&G Companies Create Value with High Performance Remote Visualization

Today’s oil and gas (O&G) companies are striving to process datasets that have become not only tremendously large, but extremely complex. And the larger that data becomes, the harder it is to move and analyze it – particularly with a workforce that could be distributed between drilling sites, offshore rigs, and remote offices. Read more…

HPC Technique Propels Deep Learning at Scale

February 21, 2017

Researchers from Baidu’s Silicon Valley AI Lab (SVAIL) have adapted a well-known HPC communication technique to boost the speed and scale of their neural network training and now they are sharing their implementation with the larger deep learning community. Read more…

By Tiffany Trader

IDC: Will the Real Exascale Race Please Stand Up?

February 21, 2017

So the exascale race is on. And lots of organizations are in the pack. Government announcements from the US, China, India, Japan, and the EU indicate that they are working hard to make it happen – some sooner, some later. Read more…

By Bob Sorensen, IDC

ExxonMobil, NCSA, Cray Scale Reservoir Simulation to 700,000+ Processors

February 17, 2017

In a scaling breakthrough for oil and gas discovery, ExxonMobil geoscientists report they have harnessed the power of 717,000 processors – the equivalent of 22,000 32-processor computers – to run complex oil and gas reservoir simulation models. Read more…

By Doug Black

TSUBAME3.0 Points to Future HPE Pascal-NVLink-OPA Server

February 17, 2017

Since our initial coverage of the TSUBAME3.0 supercomputer yesterday, more details have come to light on this innovative project. Of particular interest is a new board design for NVLink-equipped Pascal P100 GPUs that will create another entrant to the space currently occupied by Nvidia's DGX-1 system, IBM's "Minsky" platform and the Supermicro SuperServer (1028GQ-TXR). Read more…

By Tiffany Trader

HPC Technique Propels Deep Learning at Scale

February 21, 2017

Researchers from Baidu’s Silicon Valley AI Lab (SVAIL) have adapted a well-known HPC communication technique to boost the speed and scale of their neural network training and now they are sharing their implementation with the larger deep learning community. Read more…

By Tiffany Trader

IDC: Will the Real Exascale Race Please Stand Up?

February 21, 2017

So the exascale race is on. And lots of organizations are in the pack. Government announcements from the US, China, India, Japan, and the EU indicate that they are working hard to make it happen – some sooner, some later. Read more…

By Bob Sorensen, IDC

TSUBAME3.0 Points to Future HPE Pascal-NVLink-OPA Server

February 17, 2017

Since our initial coverage of the TSUBAME3.0 supercomputer yesterday, more details have come to light on this innovative project. Of particular interest is a new board design for NVLink-equipped Pascal P100 GPUs that will create another entrant to the space currently occupied by Nvidia's DGX-1 system, IBM's "Minsky" platform and the Supermicro SuperServer (1028GQ-TXR). Read more…

By Tiffany Trader

Tokyo Tech’s TSUBAME3.0 Will Be First HPE-SGI Super

February 16, 2017

In a press event Friday afternoon local time in Japan, Tokyo Institute of Technology (Tokyo Tech) announced its plans for the TSUBAME3.0 supercomputer, which will be Japan’s “fastest AI supercomputer,” Read more…

By Tiffany Trader

Drug Developers Use Google Cloud HPC in the Fight Against ALS

February 16, 2017

Within the haystack of a lethal disease such as ALS (amyotrophic lateral sclerosis / Lou Gehrig’s Disease) there exists, somewhere, the needle that will pierce this therapy-resistant affliction. Read more…

By Doug Black

Azure Edges AWS in Linpack Benchmark Study

February 15, 2017

The “when will clouds be ready for HPC” question has ebbed and flowed for years. Read more…

By John Russell

Is Liquid Cooling Ready to Go Mainstream?

February 13, 2017

Lost in the frenzy of SC16 was a substantial rise in the number of vendors showing server oriented liquid cooling technologies. Three decades ago liquid cooling was pretty much the exclusive realm of the Cray-2 and IBM mainframe class products. That’s changing. We are now seeing an emergence of x86 class server products with exotic plumbing technology ranging from Direct-to-Chip to servers and storage completely immersed in a dielectric fluid. Read more…

By Steve Campbell

Cray Posts Best-Ever Quarter, Visibility Still Limited

February 10, 2017

On its Wednesday earnings call, Cray announced the largest revenue quarter in the company’s history and the second-highest revenue year. Read more…

By Tiffany Trader

For IBM/OpenPOWER: Success in 2017 = (Volume) Sales

January 11, 2017

To a large degree IBM and the OpenPOWER Foundation have done what they said they would – assembling a substantial and growing ecosystem and bringing Power-based products to market, all in about three years. Read more…

By John Russell

US, China Vie for Supercomputing Supremacy

November 14, 2016

The 48th edition of the TOP500 list is fresh off the presses and while there is no new number one system, as previously teased by China, there are a number of notable entrants from the US and around the world and significant trends to report on. Read more…

By Tiffany Trader

Lighting up Aurora: Behind the Scenes at the Creation of the DOE’s Upcoming 200 Petaflops Supercomputer

December 1, 2016

In April 2015, U.S. Department of Energy Undersecretary Franklin Orr announced that Intel would be the prime contractor for Aurora: Read more…

By Jan Rowell

Enlisting Deep Learning in the War on Cancer

December 7, 2016

Sometime in Q2 2017 the first ‘results’ of the Joint Design of Advanced Computing Solutions for Cancer (JDACS4C) will become publicly available according to Rick Stevens. He leads one of three JDACS4C pilot projects pressing deep learning (DL) into service in the War on Cancer. Read more…

By John Russell

D-Wave SC16 Update: What’s Bo Ewald Saying These Days

November 18, 2016

Tucked in a back section of the SC16 exhibit hall, quantum computing pioneer D-Wave has been talking up its new 2000-qubit processor announced in September. Forget for a moment the criticism sometimes aimed at D-Wave. This small Canadian company has sold several machines including, for example, ones to Lockheed and NASA, and has worked with Google on mapping machine learning problems to quantum computing. In July Los Alamos National Laboratory took possession of a 1000-quibit D-Wave 2X system that LANL ordered a year ago around the time of SC15. Read more…

By John Russell

IBM Wants to be “Red Hat” of Deep Learning

January 26, 2017

IBM today announced the addition of TensorFlow and Chainer deep learning frameworks to its PowerAI suite of deep learning tools, which already includes popular offerings such as Caffe, Theano, and Torch. Read more…

By John Russell

HPC Startup Advances Auto-Parallelization’s Promise

January 23, 2017

The shift from single core to multicore hardware has made finding parallelism in codes more important than ever, but that hasn’t made the task of parallel programming any easier. Read more…

By Tiffany Trader

CPU Benchmarking: Haswell Versus POWER8

June 2, 2015

With OpenPOWER activity ramping up and IBM’s prominent role in the upcoming DOE machines Summit and Sierra, it’s a good time to look at how the IBM POWER CPU stacks up against the x86 Xeon Haswell CPU from Intel. Read more…

By Tiffany Trader

Leading Solution Providers

Nvidia Sees Bright Future for AI Supercomputing

November 23, 2016

Graphics chipmaker Nvidia made a strong showing at SC16 in Salt Lake City last week. Read more…

By Tiffany Trader

BioTeam’s Berman Charts 2017 HPC Trends in Life Sciences

January 4, 2017

Twenty years ago high performance computing was nearly absent from life sciences. Today it’s used throughout life sciences and biomedical research. Genomics and the data deluge from modern lab instruments are the main drivers, but so is the longer-term desire to perform predictive simulation in support of Precision Medicine (PM). There’s even a specialized life sciences supercomputer, ‘Anton’ from D.E. Shaw Research, and the Pittsburgh Supercomputing Center is standing up its second Anton 2 and actively soliciting project proposals. There’s a lot going on. Read more…

By John Russell

Tokyo Tech’s TSUBAME3.0 Will Be First HPE-SGI Super

February 16, 2017

In a press event Friday afternoon local time in Japan, Tokyo Institute of Technology (Tokyo Tech) announced its plans for the TSUBAME3.0 supercomputer, which will be Japan’s “fastest AI supercomputer,” Read more…

By Tiffany Trader

IDG to Be Bought by Chinese Investors; IDC to Spin Out HPC Group

January 19, 2017

US-based publishing and investment firm International Data Group, Inc. (IDG) will be acquired by a pair of Chinese investors, China Oceanwide Holdings Group Co., Ltd. Read more…

By Tiffany Trader

Dell Knights Landing Machine Sets New STAC Records

November 2, 2016

The Securities Technology Analysis Center, commonly known as STAC, has released a new report characterizing the performance of the Knight Landing-based Dell PowerEdge C6320p server on the STAC-A2 benchmarking suite, widely used by the financial services industry to test and evaluate computing platforms. The Dell machine has set new records for both the baseline Greeks benchmark and the large Greeks benchmark. Read more…

By Tiffany Trader

What Knights Landing Is Not

June 18, 2016

As we get ready to launch the newest member of the Intel Xeon Phi family, code named Knights Landing, it is natural that there be some questions and potentially some confusion. Read more…

By James Reinders, Intel

Is Liquid Cooling Ready to Go Mainstream?

February 13, 2017

Lost in the frenzy of SC16 was a substantial rise in the number of vendors showing server oriented liquid cooling technologies. Three decades ago liquid cooling was pretty much the exclusive realm of the Cray-2 and IBM mainframe class products. That’s changing. We are now seeing an emergence of x86 class server products with exotic plumbing technology ranging from Direct-to-Chip to servers and storage completely immersed in a dielectric fluid. Read more…

By Steve Campbell

KNUPATH Hermosa-based Commercial Boards Expected in Q1 2017

December 15, 2016

Last June tech start-up KnuEdge emerged from stealth mode to begin spreading the word about its new processor and fabric technology that’s been roughly a decade in the making. Read more…

By John Russell

  • arrow
  • Click Here for More Headlines
  • arrow
Share This