Powering next-gen
life sciences

Future of HPC in life sciences

Research suggests that high-performance computing (HPC)-grade data centers will be instrumental in future expansion of the life sciences market. HPC comprises high-density servers, storage, and networking devices used for resolving the major computational challenges and workloads faced by researchers, engineers, and businesses today.

The demand for HPC compute and storage is growing so quickly that many life sciences companies are unable to cope with the growth through traditional on-premise deployment methods. Many enterprises, therefore, are looking to transfer their pre-exascale and exascale HPC compute power to data center operators and colocation providers, or refresh their existing data centers, to be ready for the next decade.

At the same time, HPC in the cloud and cloud computing benefits, such as elasticity, flexibility, and pay-per-use, are also in the conversation. But there are inhibitors of HPC in the public cloud, and the pervasive availability of HPC capabilities in public clouds depends on how fast the following inhibitors will be overcome:

• Hardware and operating environment heterogeneity: not all life sciences applications run in the same way. Some may be more graphics processing unit (GPU) or central processing unit (CPU) intensive, while others may have different requirements in parallel computing. Thus one HPC cloud to rule them all would not be broadly applicable.
• Lack of low-latency interconnects, HPC-specialized accelerators, and noise-free operating systems to enable tightly coupled HPC applications to scale 
• Getting data in and out of the cloud in a less costly manner, preventing data lock-in
• Mapping HPC operations from a given company to a given public cloud (even with HPC specialization) without compromising security, supervision, administration, networking topology, or compliance

The critical role of data centers

It is essential that life sciences enterprises consider transforming their data center HPC infrastructure into a modernized, cloud-like environment, or expanding their operations through a colocation provider that specializes by industry or workload type.

In most cases, enterprises choose an HPC-ready data center provider in order to reduce the overhead of building, managing, and supporting their own HPC environment. This data center provider can deliver an environment that is suited to each client’s needs. While some only need floor space, power, and cooling, others may want to have management further up the HPC stack or extra security, both physical and infrastructure.

Some regional data center providers can offer the best of both worlds, namely the cost effectiveness and control of an HPC cluster and the scalability for bursting as needed. 

Ramboll can help

Ramboll is a leader in HPC-grade data center services and resources in the life sciences industry. We work with life science departments and groups to define the best strategy, either by upgrading an existing data center or using a third-party data center provider. Our services include:

• Assess space, power, and cooling density on schedule for high-performance computing (HPC) clusters, using unique methodology and tools and years of collective intellectual property
• Project capacity of current and for growth, including unpredictable spikes in use, with data center grade tools and solutions
• Identify and manage (multi)vendor selection and evaluate risks, environmental factors, and legal/compliance, with a complete risk valuation methodology