LABORATORY INFORMATICS GUIDE 2022


analytics and AI and ML, the requirements to support these systems continues to increase. This is leading more and more companies in biotech and drug discovery to adopt managed services, which free the organisation to focus on its scientific output. ‘Typically, we would meet with


a customer, they would outline the requirements for a project or a set of projects. We would deliver with them a proposal in the form of a statement of work that outlines their requirements, our deliverables, pricing and finite times scales,’ said Riener. ‘The key with a managed services model is you effectively have one person who is the key point of contact. Behind the scenes there are a lot of very experienced individuals doing the work. ‘Scientists are paid to do science


and research – they are not paid to manage scientific computing resources. We focus on what we do really well and that allows them to focus on what they do really well,’ Riener added.


Changing requirements While there are similarities in the goals and challenges these organisations face, the technologies, software and implementation models they choose can be vastly different, which means RCH is constantly adapting to meet new challenges and driver RCEs that can support scientific innovation. ‘The interesting thing with this job


is, there really isn’t too much that is typical,’ noted Eschallier.


‘Everyday brings a new set of challenges and the type of people we hire really embrace that. They are not the kind of people that use standard operating procedures that like to rinse and repeat problems. ‘We will get involved with


anything based on: computing at scale, specalised applications, HPC, leveraging public cloud, curating data, analytics, mining and driving ML/DL,’ Eschallier continued. ‘Each customer has a similar problem: “I need to find a target, synthesise it and move it forward and look for things like toxicity and efficacy”. ‘But how they do it and the


technologies they want to use and the processes they want to follow are all different,’ he stressed. ‘Sometimes they want a holistic,


cradle-to-grave approach, and other times customers will say I just need you to do these three niche items. ‘It is about standing up software, making data available and performance tuning is a growing requirement,’ Eschallier concluded.


Adapting to change Riener also gave a practical example based on the company CelGene, which has now been acquired by Bristol Myers Squibb. The company had already developed its own computing environment for scientific research, but found it did not meet the requirements of its researchers upon completion. ‘CelGene was growing rather quickly and they were in a state of change. They


were acquiring a lot of companies and leadership decided it was going too fast and they needed to determine some parameters and take control,’ stated Riener. ‘There were many demands placed upon them from the business, but the one that stood out was the need for a scientific computing environment. ‘Corporate realised that, if they were


going to succeed collectively, they required a platform that would allow them to [do] scientific computing, share data and collaborate,’ he continued. ‘Effectively, they built this RCE, opened up shop and nobody came. When they investigated why, the feedback they got was not very positive. It was largely based on the fact this RCE was not built with researchers in mind. It was designed with good intentions, but it was designed around IT.’ At this point RCH was allowed to


come and investigate and recommend how this might be fixed. After working with CelGene and investigating the challenges and solutions, it helped the company redefine the RCE to better suit the researcher and scientists. ‘We happened to be very fortunate at the time we arrived,’ notes Riener. ‘We were asked by the chief architect of this platform to come in and evaluate it. We respectfully and constructively pointed out some things that needed to be changed and they embraced most of what we had to say and we redesigned this platform with them. We worked with them and brought the teams together to redesign this RCE and deployed it,’ Riener concluded. l


22


www.scientific-computing.com


metamorworks/shutterstock


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25  |  Page 26  |  Page 27  |  Page 28  |  Page 29  |  Page 30  |  Page 31  |  Page 32  |  Page 33  |  Page 34  |  Page 35  |  Page 36  |  Page 37  |  Page 38