HOW CLOUD HPC IS REDUCING TIME TO INSIGHTS IN PRECLINICAL RESEARCH
share:
system to handle the work. The resulting cloud sys-
tem distributed 10,600 spot instances across four
connected zones, giving Novartis access to approxi-
mately 87,000 compute cores.
Novartis calculated the project equated to 39 years of
computational chemistry. The cloud HPC performed
the screening in nine hours at a cost of $4,232. "It real-
ly was amazing," Litster said. "Most importantly, we
actually identified three promising compounds."
Cloud HPC's flexibility--and the cost at which it is avail-
able--has implications for how preclinical researchers
approach experiments. If Novartis had been limited to
on-premise HPC, the screening program may never
have happened and the three promising compounds
would never have been identified. This is an example
of how cloud HPC is changing the economics of drug
discovery.
"When you have the capability to bring low-cost
experimentation to the scientific discovery cycle,
we believe there is going to be a lot more science
that gets done," said Mark Johnston, director of
global business development, healthcare and life
sciences at AWS. "Companies won't be so cautious
or conservative with how they go about resourcing
experimentation because we've effectively changed
the economics and the dynamics of it."
Users of cloud HPC move at whatever pace is most
appropriate for their experiments. If a research team
needs to complete a task very quickly, for example,
because related projects cannot advance until it is
done, it scales up its cloud HPC to accelerate prog-
ress. Such scalability shortens time to insight.
"By spinning up a few hundred nodes on AWS
and getting results in less than a day, our scientific
researchers have a lot more freedom to ask questions
that weren't even possible before," Celgene's Smith
said.
The cloud also optimizes resource allocation for less
time-sensitive projects. Scaling down cloud HPC opti-
mizes the balance between cost and speed.
Future-proofed computing
The flexibility of cloud HPC has long-term implica-
tions. Researchers' computing needs will change in
the years to come. The amount of data generated in
preclinical research is growing and will continue to do
so.
Expanding on-premise HPC to manage these data
is a major undertaking. Users of cloud HPC, by con-
trast, can automatically scale up clusters to meet their
growing compute needs. The adaptability and scal-
ability of cloud HPC ensures researchers will always
have the right cluster for the job.
This promise extends to the technology used in cloud
HPC. On-premise HPC systems can be dated by the
time they are installed. Cloud HPC frees researchers
from worries about technology obsolescence. The
burden of maintaining leading-edge technology falls
on service providers, which, in the competitive cloud
sector, need to offer the most advanced computing
components to succeed.
In this regard, as in many others, cloud HPC shifts
technical pressures from researchers to service pro-
viders. The upshot is users of cloud HPC go to work
each day knowing they have what they need: scalable,
leading-edge compute capacity tailored to cutting
the time it takes to achieve their research endpoints.
These endpoints are far from abstract goals. They
are staging posts on route to therapeutics that will
address major unmet medical needs and in doing
so touch the lives of billions of people. Cloud HPC
will be with researchers at each step in the process.
Drug development will remain the toughest of tasks.
But with cloud HPC researchers can strike compute
capacity from their list of challenges. l
For 10 years, Amazon Web Services has been the world's most comprehensive
and broadly adopted cloud platform. AWS offers over 70 fully featured services
for compute, storage, databases, analytics, mobile, Internet of Things (IoT) and
enterprise applications from 38 Availability Zones (AZs) across 14 geographic
regions in the U.S., Australia, Brazil, China, Germany, Ireland, Japan, Korea,
Singapore, and India. AWS services are trusted by more than a million active customers around the world —
including the fastest growing startups, largest enterprises, and leading government agencies — to power their
infrastructure, make them more agile, and lower costs. To learn more about AWS, visit http://aws.amazon.com.
