Analyst firm Gartner has urged CIOs to ensure their teams have the right skills to exploit in-memory computing.
Gartner defines in-memory computing as an emerging paradigm enabling users to develop applications that run advanced queries or perform complex transactions, on very large datasets, at least one order of magnitude faster — and in a more scalable way — than when using conventional architectures.
In-memory computing achieves this by storing application data in a computer's main memory, rather than on discs, and without compromising the availability, consistency and integrity of that data.
Massimo Pezzini, a research director at Gartner said in-memory computing can boost a wide range of applications, both analytical and transactional, because it improves applications' performance and scalability and enables very deep data analysis.
Pezzini said areas include risk management, fraud detection, e-commerce, on-line entertainment (gaming and betting), internet banking, online trading, profitability analysis, manufacturing resource planning and corporate performance management. Software-as-a-service (SaaS) and social networking often use in-memory computing technologies.
However Pezzini warned: “Organisations that do not consider adopting in-memory application infrastructure technologies risk being out-innovated by competitors that are early mainstream users of these capabilities."
In a recent interview with Computer Weekly, in-memory database guru, Alexander Zeier, Accenture’s global lead for in-memory solutions, explained that one of the benefits of in-memory technology was that it would give users the perception of instant responses in an application. “The speed of thought is 700ms, so you can build really cool mobile applications with instant response times.”
Unilever is one major business starting to use the technology. It is assessing how to use SAP’s Hana in-memory database appliance to speed up order processing. Unilever believes in-memory computing enables it to reduce the number of instances it runs around the world. With fewer instances, business managers will be able to have a real-time view of the company’s operations, which could allow them to make better-informed decisions on product pricing.
Pezzini said CIOs should ensure their teams focus on three core skills. First, they must have the development skills for designing in-memory computing-based applications, which assume data is in the memory – compared with traditional applications, which assume the data is on disk.
The second core skill area lies in architectural skills. Pezzini said: “In-memory computing requires application and deployment architectures more complex than those needed for traditional applications."
Finally, Pezzini recommends CIOs develop skills to assess the impact of in-memory computing on IT operations. “In-memory technology is different from traditional technologies, therefore IT operation personnel must be retrained.”
While enterprise suppliers such as Oracle and SAP provide in-memory database appliances, such as SAP's Hana and Oracle's Exalogic, Pezzini warned that applications may require large amounts of re-engineering to make the most of the in-memory platform.
“In general, to achieve the full benefits of in-memory computing, there is some significant re-engineering to be made,” he said.
Existing applications may not be the best candidates for in-memory computing. “The in-memory computing design principles are rather different than those of traditional systems,” Pezzini explained.