In-memory computing powers digital transformation

Digital transformation will be accelerated via in-memory computing in 2019 as businesses look to harness the value of their enterprise wide data sets.
26 December 2018 | 832 Shares

Members of the BuzzFeed News team in New York City.(Source: Drew Angerer/Getty Images/AFP)

As more companies push towards becoming digital enterprises, the digital transformation agenda will continue to gather momentum, and in-memory computing (IMC) will play a key role in fueling this vision.

Companies will be able to reinvent their business models, redefine their relationships with customers, and be a disruptive force in their chosen marketplaces.

Speaking to CBR, GridGain Systems’ Founder and CTO, Nikita Ivanov believes that companies would do well to understand that if they don’t become a data-driven enterprise, they will likely lose out to a company that is one.

This digital transformation trend, with IMC as a ‘key enabler’, is expected to become more prevalent, as the technology is becoming increasingly mainstream with lower costs and vendor solutions maturing as well.

Moreover, IMC solutions are becoming more flexible as vendors provide native integrations with common solutions like Apache Kafka, Apache Spark, and protocols like SQL, Java and .NET

She also cites the dramatic increase of technical professionals with experience in implementing distributed computing solutions as a major advantage for companies looking towards the deployment of these systems.

Apparently, establishments like American Airlines are using IMC to accelerate response times, automate processes, and meet SLAs for business applications that process huge data sets from multiple sources.

IMC platforms are also being used in computational drug discovery projects, with organizations like eTherapeutics accelerating projects that used to take weeks, to just a few hours or minutes in some cases.

A company that supplies financial services software, Finastra, also uses IMC solutions for real-time processing of trades and transaction data, thereby reducing bottlenecks to facilitate next-generation types of services.

IMC solutions are also being utilized by institutions like ING, to support growth, satisfy customer expectations for mobility and provide an enhanced user experience as it continues to rely on its legacy mainframe computers.

In 2019, Ivanov believes that the addition of machine learning capabilities to IMC platforms will enable companies to expand and refine their omni-channel customer experience initiatives.

New computing approaches, such as hybrid transactional/analytical processing (HTAP), also known as hybrid operational/analytical processing (HOAP) will unleash data in real-time and lead to better business results.

Machine learning models can now be retrained in real-time based on new operational data, and it can support mission-critical applications such as fraud detection, credit approvals, price settings, and even vehicle or package routing.

TensorFlow , a deep learning system, is also integrated into IMC solutions, so companies can now reduce costs dramatically as the need to maintain a separate analytical architecture to their operational database is eliminated.

According to Ivanov, 2019 will also see the emergence of in-memory-computing-platform-as-a-service (imcPaaS), where databases are consumed as service from cloud providers.

451 Research believes that 60 percent of workloads will be deployed in cloud environments, including on-premises private clouds, hosted private clouds, IaaS, and SaaS in 2019.

Furthermore, by 2022, Gartner predicts that at least two-thirds of in-memory-computing data grid implementations (IMDG) will be driven by application platform as service (aPaas) deployments.

Ivanov believes that in the quest to be data-driven businesses, companies will have to accelerate their digital transformations or risk being left behind.

She is of the opinion that IMC will become an integral component of the ‘fast data’ discussion, as companies look to harness the sheer complexity of their enterprise-wide datasets towards driving their commercial aspirations.