Archives

The MCC on R@CMon

The Monash Campus Cluster (MCC) is a heterogeneous high-performance  (HPC) and high-throughput computing (HTC) facility for conducting large-scale computationally-intensive simulations and analyses. With over 2,500 CPU cores across 230 servers of different CPU and memory configuration, the MCC is specifically designed to serve diverse computational workloads. In 2013, the MCC provided over 13 million CPU-core hours to over 300 Monash researchers.

During the past few months, we have been developing a software architecture to extend MCC’s computational resources into the NeCTAR Research Cloud. Users are presented with the MCC’s familiar batch queueing and software environment, so they can seamlessly execute compute jobs on either the legacy cluster nodes of MCC or the NeCTAR Research Cloud.

MCC-NeCTAR

Monash Campus Cluster, “bursting” into the NeCTAR Research Cloud.

We achieve this by integrating the NeCTAR virtual machines as compute bricks into the MCC batch queuing system, presently the Open Grid Scheduler. This provides users with:

    • a new nectar queue which consists of MCC on R@CMon compute bricks; and
    • the ability to pick a specific availability zone on the research cloud to run compute jobs on

Researchers not only get to burst their computational jobs into the research cloud seamlessly, they can leverage the unique properties of each of the cloud nodes (e.g., hardware and/or software capabilities).

$ qconf -sql
nectar
nectar-gaia
nectar-melbourne
nectar-monash
nectar-sa

Since December 2013, over 40,000 CPU hours worth of computational jobs have been executed into the NeCTAR queue of MCC. The resources used by MCC on R@CMon will be expanded with deployment of R@CMon Phase 2 to accommodate specialised computational workloads (e.g., high memory jobs). Monash researchers who have their own NeCTAR Allocations and want that allocation to be presented through MCC, can opt for their allocation be managed by the “MCC on R@CMon” project. This can be arranged by contacting the Monash eResearch Centre.

The CVL on R@CMon

The Characterisation Virtual Laboratory (CVL) is a powerful platform that integrates Australian imaging facilities with computational and data storage infrastructure, together with sophisticated processing and analysis toolsets. The CVL platform provides scientists working in various fields with a common analysis and collaboration environment, the CVL turns the humble remote desktop into a highly flexible Scientific Software as-a-Service delivery platform powered by the NeCTAR Research Cloud.

CVL-Desktop-01

The CVL Desktop

The current production CVL includes toolsets covering Neuroimaging, Energy Materials and Structural Biology research drivers. The project includes so-called “CVL fabric services”, which provide the necessary infrastructure to modularise popular software toolsets from any number of domains.

The R@CMon team assisted the CVL team in migrating CVL services into R@CMon. The use of persistent storage (Volumes on R@CMon) ensured consistent user home directories and software-stack repositories. The default “CVL Desktop” pool is now serving users with software-rendered CVL environments running on R@CMon. The CVL team is also a beta user of GPU flavours on R@CMon and is currently testing GPU-enabled CVL environments on the “CVL GPU node” pool (via CVL Launcher).

Screen Shot 2014-03-11 at 2.48.55 pm

The available pools on the CVL Launcher.

The following video demonstrates a GPU-enabled CVL environment launched on R@CMon. It shows the PyMOL and UCSF Chimera applications from the Structural Biology workbench, running and utilising the available GPU. The use of GPU enables seamless interaction and manipulation of datasets.

The plan is to increase the “CVL GPU node” pool to accommodate more users once GPU node capacity on R@CMon has been upgraded with deployment of R@CMon Phase 2. Watch this space for more CVL on R@CMon news. Other updates about the CVL and its sub-projects are also available on the CVL site.

Bioplatforms Australia – CSIRO Metagenomics Workshop (February 6-7, 10-11 2014)

Bioplatforms Australia and CSIRO conducted an “Introduction to Metagenomics” workshop last February 6-7, 2014 at University of New South Wales and February 10-11, 2014 at Monash University. The workshop was aimed for bench biologies with no or little experience in Bioinformatics using publicly available data resources and toolsets.

As per previous Bioplatforms Australia workshops, the Metagenomics workshop was delivered using the Monash node of the NeCTAR Research Cloud infrastructure – R@CMon. Cloud provisioning tools used in previous workshops have been reused to provide a seamless virtual desktop training platform.

The R@CMon team worked with Bioplatforms Australia, CSIRO and EMBL-EBI in producing an appropriate cloud image and toolset for the workshop. Some of the tools used in the workshop are QIIME, FastQC, and InterProScan.

Given the success and popularity of the training, R@CMon has begun work with Bioplatforms Australia and other nodes of the NeCTAR Research Cloud to scale the training environment with trainees as they progress from taking the course, to taking the training environment home with them, to preparing for production genomics facilities.

The trainees have the following to say about the 2-days workshop:

“Everyone is very helpful and the content was clear and concise. I thank everyone for getting this program up and running and I would definitely like to return for similar courses.”

“Definitely a fantastic & informative 2-days. I definitely feel that most of what I learned today can be directly applied to the molecular work I am currently engaged in.”

“Good amount of hands on to actually see what happens during the analysis of big data, most of what was taught was very clear and concise. good range of programs were used, only wish there was more time to go through more analyses.”

“I strongly recommend this as a complete course for students or beginners in NGS analyses.  i think the first metagenomic workshop is a winner.”

“Definitely this course has provided me a great basis to look at future data sets.”

“Well ran, good materials, virtual machine made life easy.”

Bioinformatics Training on R@CMon

A multidisciplinary partnership between Monash eResearch Centre and Bioplatforms Australia has provided a broadly accessible solution to delivering hands-on bioinformatics workshops with seamless access to cloud computing using the new NeCTAR Research Cloud infrastructure.

Running hands-on bioinformatics workshops in Australia has previously been hampered by the lack of specialised bioinformatics training facilities and a paucity of skilled trainers to develop and deliver these courses.

To improve the bioinformatics skills of bench scientist now faced with handling gigabyte size datasets generated by next-generation sequencing technologies, Bioplatforms Australia and CSIRO have been collaborating to advance bioinformatics expertise among Australian ‘omics researchers.  Through an international partnership with the EMBL European Bioinformatics Institute in the UK, a cutting-edge three-day Australian hands-on NGS workshop has been created. This course introduces bench scientists to quality control of NGS data, alignment, ChIPSeq, RNASeq and de novo assembly workflows and software.

Professor Paul Bonnington, Director of the Monash eResearch Centre and the R@CMon team contributed to this training initiative through the development of a cloud computing-based NGS bioinformatics training platform based on the open source bioinformatics software package CloudBioLinux. The platform allows sharing of data, tools and applications and enables trainers anywhere in the world to readily work together to develop and test new workshop material.

The Bioplatforms Australia Next Generation Sequencing workshop platform enables compute-intensive NGS training courses to be easily delivered and accessed widely around Australia and requires very little local IT expertise or need for high end computational hardware. The first hands-on workshops using the NGS workshop platforms were held in July 2012 at Monash University in Melbourne and at University of New South Wales in Sydney. To date 10 workshops have been delivered around Australia in Melbourne, Sydney, Brisbane, Adelaide, Perth and Canberra to 345 trainees.

The team of trainers from Bioplatforms Australia and CSIRO in collaboration with EBI-EMBL and Monash eResearch Centre are currently developing a metagenomics 2-day workshop using a bespoke metagenomics image built by R@CMon. This course will be run at University of New South Wales in Sydney on the 6-7 February 2014 and at Monash University on the 10-11th February 2014.

Monash University is one of the nodes of NeCTAR‚ a research cloud platform, a landmark investment that will extend the advantages of high-performance computing and high capacity networks to Australian researchers. This exciting initiative provides on-line access to scalable computational power and data storage allowing a new realm of data sharing and collaboration.

The Bioplatforms Australia Next Generation Sequencing workshop platform is now freely accessible on the NeCTAR research cloud and provides access to hundreds of bioinformatics software packages.

Further information on the Bioplatforms Australia Next Generation Sequencing workshop platform is available from Catherine Shang, Bioplatforms Australia at cshang@bioplatforms.com. Contact Prof. Paul Bonnington, MeRC Director for further details and assistance on e-research solutions.

 

Maya/mental ray Render Farm on R@CMon

Autodesk Maya is a comprehensive software suite that offers 3D computer animation, modelling, simulation, rendering and composition. Maya has been used in various industries to generate 3D assets that are used in film, television, game development and architecture. mental ray is a high performance 3D rendering software for producing highly realistic images using advanced ray tracing techniques. Used by various industry professionals, mental ray has become a standard of photorealistic renderings across many industries.

Tom Chandler, Lecturer, Faculty of Information Technology at Monash University and his team have been using both Maya and mental ray to create highly realistic images and animations for various projects. Their previous render farm was built with traditional desktop machines. As the demand for more resolution, more detail, more frames and more projects continues to increase, they realised that desktop rendering couldn’t provide the capacity and agility to meet their resource requirements. Fortunately Tom heard of a similar render farm the R@CMon team helped in porting into the NeCTAR Research Cloud and then approached us.

The R@CMon team assisted Tom’s team in setting up a Windows Server 2012 environment in the Monash node of the NeCTAR Research Cloud where both Maya and mental ray are installed, licensed and configured. Access to the environment is done using Remote Desktop Connection.

maya-remote-desktop

The Maya 3D animation software, running on a R@CMon-provisioned Windows Server 2012 virtual machine.

Tom’s team started migrating their 3D rendering projects into the Maya/mental ray render farm now running on the NeCTAR Research Cloud. One of these projects is The Visualising Angkor Project – “Visualising Angkor Project” Monash University Faculty of IT, 2013. It is a collaborative project between Monash University and University of Sydney which explores the 3D generation and animation of the Angkor landscapes during the medieval period. The project presents some challenges on how virtual Angkor is interpreted based on archaeological and historical data.

The following is the latest animation rendered on the Maya/mental ray render farm by Tom Chandler’s team for The Visualising Angkor Project – “Visualising Angkor Project” Monash University Faculty of IT, 2013It contains about 1.5 million triangles, 100 materials, 100 textures, and a total of 1275 frames.

R@CMon at Monash Technology Fair 2013

The R@CMon team, along with our colleagues from the Monash eResearch Centre, manned a booth at the recent Monash Technology Fair 2013. The event was held at Clayton Campus between the 22nd and 23rd of October. The Tech Fair is Monash’s internal “trade-show”, showcasing products and services available for Monash researchers, teachers, students and professional staff – including the NeCTAR Research Cloud!

The eResearch booths: R@CMon, Research HPC & Visualisation, Research Collaboration Tools, and Research Data Management; held prime position at the entry and attracted a good deal of interest from both old and new Monash staff. We were particularly pleased to hear how impressed our local research community is with the NeCTAR Research Cloud project, especially with Monash’s early participation in such an ambitious research-focused project.

GC9C9919

Professor Paul Bonnington, outlining MeRC capabilities.

If you picked up one of our flyers (or even if you didn’t) but need some more information or help getting started on R@CMon, then please don’t hesitate to contact us.

The GVL on R@CMon

The R@CMon team has been working closely with the GVL (Genomics Virtual Laboratory) team to bring the GVL to the Monash node of the NeCTAR Research Cloud. It’s now here!

The GVL provides an accessible infrastructure for connecting genome researchers  with their datasets using analysis tools that are executed on dynamically provisioned computational resources – Amazon AWS EC2 and the NeCTAR Research Cloud. The GVL Launcher allows users to launch CloudBioLinux, CloudMan and Galaxy computational platforms on a number of cloud computing infrastructures.

gvl-launcher-cloud

The GVL Launcher, showing available cloud infrastructures for starting GVL clusters.

Initially, the GVL was only available on the Melbourne node. Persistent volume storage is one of the main requirements of the GVL. With the initial volume service now available on the Monash node, users can now launch the GVL at the Monash node. By choosing the Monash availability zone, the GVL head node and all subsequent compute nodes will be launched at monash-01.

gvl-launcher-placement

Users can choose specific availability zones to launch their GVL instance.

The GVL team makes the latest stable GVL image available to the users.  And now, we’re ready to launch our first GVL platform in the Monash node. This will launch the head node for the platform. It will be accessible via CloudMan web interface once it has been booted up and all necessary services have been initialised. In the next parts of this blog series, the various GVL use cases will be covered, including – using a dynamically scaled GVL platform.

gvl-launch-success

GVL Launcher status page, provides information on accessing the GVL instance.