Skip to content
Home » Building a Robust Genome Bioinformatics Computing Infrastructure

Building a Robust Genome Bioinformatics Computing Infrastructure

In 2022, ABI embarked on a important mission to enhance its genome bioinformatics capabilities, recognizing the need for substantial computational resources. Spearheading a fundraising campaign, ABI garnered support from various entities, including grant funding agencies, donor organizations, and individuals via the ReArmenia platform. This collective effort resulted in the acquisition of a compute node, a crucial enabling technology for bioinformatics data analysis projects. 

Notable contributors to this initiative include the Science Committee RA, FAST (Foundation for Armenian Science and Technology), Flagship Pioneering, SASTIC (Strategic Armenian Science & Technology Investment Community), and SmartClick. With over 100 organizations and individuals backing the cause (worth highlighting the strong support of the VMware community in Armenia), ABI secured funding of 60,000 USD for its new compute node, laying the foundation for a state-of-the-art genome bioinformatics computing infrastructure.

The compute node, a 2x AMD EPYC 7702 system with dual sockets, 64 cores, 2TB of memory, and 240TB of storage, was selected to meet the specific requirements of ABI’s researchers. Technical support from Narek Khachatrian’s team at Security Systems LTD facilitated the acquisition and installation process, ensuring a seamless transition to a more robust computing environment.

A dedicated system administrator, recruited through the campaign’s success, played a crucial role in setting up the node and continues to ensure its ongoing maintenance. With access granted to over thirty individuals from both ABI and IMB, the server is not only maintained to current standards but also undergoes regular updates to provide researchers with the necessary tools and software for their work.

Setting up and maintaining the node

Eduard Avetisyan, the head of the European Molecular Biology Laboratory (EMBL) IT team in Hamburg, Germany, and a member of ABI’s advisory board, played an instrumental role in establishing and maintaining the compute node, providing advice on the configuration of the system and on the setup. In the EMBL blog post We are EMBL: Eduard Avetisyan on the role of IT in structural biology, Eduard highlights the crucial role the computational infrastructure plays in the field: 

“It’s often said that IT services are like air – usually you don’t notice them, but without them, your work would suffocate”. (Eduard Avetisyan in the server room at EMBL Hamburg. Credit: Dorota Badowska/EMBL)

The implementation of the setup has been skillfully managed by ABI System administrator Antranig Vartanian and researcher Siras Hakobyan.

Antranig Vartanian, a key figure in setting up and maintaining the node, shares insights and experiences in his latest blog post. Dive into the details of the compute node’s implementation and maintenance, as Antranig sheds light on the practical aspects that contribute to ABI’s stride towards cutting-edge research infrastructure.

Over the last months, The Armenian Bioinformatics Institute purchased, set up, and deployed multiple new systems, including routers, switches, and a server powerhouse.

With the amount of computation that we do at ABI, there’s no such thing as “it’s enough” for storage or computation, that’s why our new AMD EPYC 7702 (dual-socket) system with its 2TB of memory and 240TB of storage has become the new home of our scientists.​

We (with the help of our partners) have redone the electricity system of our building, spent weeks in cable management for our labs, and updated the network equipment for the institute.​

We welcome our new 10+25G Mikrotik Router (CCR2004-1G-12S+2XS), 10+40G Mikrotik Switch (CRS326-24S+2Q+), 3kW UPS, and a tiny little Raspberry Pi 4 (Model B, 8GB RAM) to our infrastructure family.

The infrastructure team has also set up a new website with a blog and an alerting system for our users: [Infrastructure @ ABI] (https://infra.abi.am/). The new system uses rock-solid technologies such as FreeBSD, OpenZFS, Jails, Ubuntu, and SLURM which gives our scientists the ability to submit work around the clock. In the coming months, we plan to implement more features such as eduroam for our guests and a public uptime/statistics dashboard for the community. We hope these changes allow our users and community members to expand their work into more areas and handle more projects than ever.

Current Milestones and Future Expansion Plans

Celebrating the successful setup and utilization of the new compute node, ABI looks to the future with optimism. Plans are underway to expand the system administration team, and storage volumes, ensuring sustained excellence in infrastructure support. The goal is not only to maintain current standards but also to extend access beyond ABI and IMB, reaching educational and research institutes, as well as clinics.

In alignment with ABI’s commitment to fostering widespread accessibility to bioinformatics resources, this initiative stands as a testament to the collaborative spirit of the scientific community. The server, though modest in its description, represents a significant leap forward, empowering researchers and benefitting the broader scientific landscape.

Blogpost by Andranik Wartanian, Araz Chiloyan