In-Memory Is Faster, Better, And More Flexible

by CXOtoday News Desk    Feb 23, 2015


Huge data volumes and a focus on analytics in business, has prompted the growth of in-memory data grids. Oracle’s Database In-Memory is designed to speed up workloads driven by analytics with the capacity to cope with billions of data values per second. In an exclusive interaction with CXOtoday, Sanjay Gupta, Vice President, Fusion Middleware Business, Oracle India throws light on how in-memory data grids are meeting some of the critical storage challenges by CIO.

In light of growing user demand and shrinking technology budgets, what are the challenges faced by IT departments while making new systems or supporting existing systems especially in a context where Mobile, cloud and the growth in the number of connected devices — are driving demand for faster, more scalable computer systems.

Today, devices of all sizes, ranging from credit cards tohousehold appliances and even to our homes and cars,are being electronically integrated with each other andthe Internet.Existing technology that is gaining popularity, such asmobile devices, social media, cloud computing, and always-connected devices, requires the management oflarge volumes of data from an ever-expanding universeof highly varied data sources. Supporting these consumer and device-to-device, oftentermed the Internet of Things trends requires a robust IT architecture. However with shrinking IT budgets and growing user demands, IT departments are facing the following challenges:

·         Computer systems use shared infrastructure to ensure that stable and trusted components areused within the enterprise.

·         Application usage and workload are often difficult to forecast accurately at the system’s inception.

·         Unpredictable usage spikes and new or changing requirements are difficult to plan for in advance with enough lead time to take corrective action.

·         Performance problems within highly specialized components or databases or infrastructure can be expensive and time-consuming to correct.


What are In-memory data grids (IMDG) and how do they help IT departments address the above challenges?

IMDG is a low-cost, high-efficiency technology that is expected to gradually replace the conventional disk-based technology owing to its inherent capacity enabling efficient data storage, access and management. IMDGs are middleware software that manages data objects across multiple distributed servers in a horizontally scaled architecture. To put it simply,if a website is slow or a sales application is cumbersome, previously loyal customers won’t hesitate to move to a competitor in a matter of seconds. In this context, leveraging caching and memory-grid technologies has a direct impact on the customer experience. Technology like IMDG provides the speed, reliability, and scaling required by computer systems to support an environment that is responsive and reliable even under spikes in user demand. IMDGs provide faster and more stable service to customers while solving difficult scaling and reliability challenges thereby giving customers the user experience they demand.

Under what real-world circumstances would IMDG’s be a good choice?

The ideal use cases are situations where a larger number of distributed, concurrent users access an online system and repeatedly access similar data, inwhich users won’t tolerate slowness or failures. Take for example an online travel and hotel reservation application. Prospective travellers know where they want to vacation, but the details are often in flux while they evaluate their options for the best price before committing to a purchase.

Consumers, enabled by mobile devices and always on broadband connections, access the website from around the world 24/7. Since the travel service is free, customers use it when they have no intent on buying. This puts strain on the website infrastructure, which increases cost to run it. In addition consumers may seek to see 100 quotes at a time, so a single query can result in aggregating a large number of quotes from a variety of sources. They will also be searching for similar data on competing websites as they compare flights, hotels, dates, prices, and promotions to find the deal that fits their needs the best.

In this scenario performance and system stability is critical. Attempting to access data directly from each source without the benefit of caching is far too slow and prone to failures. If consumers become frustrated, they’ll go to a competitor. Besides, workload and usage spikes occur when online promotions are offered, special events are advertised, or seasonal events come up, even something as simple as summer vacation months.Cost-effective IMDG solutions are leveraged by such websites to ensure that customers have an optimal user experience, which directly translates to increased sales and repeat business.