In-memory computing can enable companies to deliver real-time, massively scalable applications for a range of use cases. Organisations in financial services, fintech, IoT, software, SaaS, ecommerce, healthcare and more already use in-memory computing to solve the challenges presented by complex, data-intensive applications. By attending the In-Memory Computing Summit Europe 2019, you can better understand the uses and benefits of today’s IMC solutions, see how they can be optimised for your specific needs, and learn about the newest solutions coming to market.
What Is In-Memory Computing?
In-memory computing is now a widely adopted technology that Gartner predicts will be incorporated into the majority of web-scale applications by next year. IMC platforms provide the application speed and scalability organisations need to meet the demands of data-intensive applications. These platforms typically include some or all of the following capabilities:
·IMDG and IMDB – Deployed on a cluster of servers on-premises, in a public or private cloud, or in a hybrid environment, IMDGs (in-memory data grids) and IMDBs (in-memory databases) pool available memory and compute to enable all data to be kept in memory, eliminating the need for applications to constantly read and write data to disks. IMDGs are deployed on top of an existing database and keep the database in synch. IMDBs hold data in memory with all data written to disk for backup and recovery and the ability to process against the disk-based dataset for fast restarts and to allow tradeoffs between application speed and infrastructure costs by maintaining only some of the data in memory.
·HTAP or HOAP – HTAP (hybrid transactional/analytical processing) or HOAP (hybrid operational/analytical processing) are defined by systems that access a single database for simultaneous transactional and analytical processing, eliminating the need for the time-consuming extract, transform, load (ETL) process of moving data to a separate analytical infrastructure which results in higher costs and delays in data availability for analytical processing.
·Streaming data processing – Streaming data processing solutions manage the complexity around dataflow and event processing, enabling users to rapidly ingest, transact, and analyse high volume data streams with real-time performance.
·ML and DL – By incorporating ML (machine learning) and (DL) deep learning capabilities, an IMC platform can provide what Gartner refers to as “in-process HTAP,” which is HTAP combined with real-time updates to machine learning models for optimal decision making at the point where transactions are occurring. For ML use cases, a continuous learning framework enables organisations to dramatically increase the frequency at which an ML model is updated and deployed to production. For DL use cases, a continuous learning framework can make it much easier and faster to move data into a separate deep learning infrastructure for model training.
How Companies Are Benefitting
Just a couple of examples can show the tremendous potential of in-memory computing.
·By providing real-time performance for data-intensive and compute-intensive applications, IMC is enabling a drug discovery and development company to perform hundreds of thousands of analyses across multiple parameter sets and assumptions—and complete analyses in just a few hours or even minutes that used to take weeks while also allowing them to launch new initiatives that were previously computationally infeasible. This is opening new approaches to tackling diseases and perhaps laying the groundwork for new treatments for cancer.
·An IMC platform is also enabling a sports betting platform to constantly update huge amounts of data from multiple events in multiple sports and make this data immediately available to a vast number of clients. With dynamically scalable IMC technology, the sports betting platform is able to track multiple live events, instantly sharing new betting opportunities and odds for each event. The platform currently supports over 500 casino and live dealer games and transacts more than £2 million in bets daily, while processing over 700 bets per second.
·Powering trading platforms for a variety of European banks, IMC platforms support high speed trading and power FRTB (fundamental review of the trading book) for real-time compliance reporting using an HTAP/HOAP architecture which delivers simultaneous transactional and analytical processing.
·By enabling credit card companies to update their ML models hourly instead of nightly, IMC is helping banks reduce their vulnerability to new fraud vectors.
The In-Memory Computing Summit Europe 2019
This year’s In-Memory Computing Summit Europe is the only in-memory computing conference in Europe focusing on the full range of in-memory computing-related technologies and solutions. It offers multiple tracks, including the latest advancements in IMC architectures, new hardware and software solutions for taking advantage of IMC platforms, the latest streaming data use cases, and lessons learned from actual IMC deployments. The In-Memory Computing Summit Europe agenda includes speakers from Barclays, CS Financial Group, Capital One, Cerner Corporation, GridGain Systems, GSI Technology, Hazelcast, Oracle, Redis Labs, Sberbank and ScaleOut Software. The event is attended by hundreds of industry experts and enthusiasts, creating a terrific setting for learning, networking and sharing experiences with in-memory computing.